DESC:SYSTEM AND METHOD FOR DETECTING ANOMALIES IN POWERPACK
CROSS REFERENCE TO RELATED APPLICATIONS
The present application claims priority from Indian Provisional Patent Application No. 202421068719 filed on 11/09/2024, the entirety of which is incorporated herein by a reference.
TECHNICAL FIELD
Generally, the present disclosure relates to a battery monitoring mechanism(s). Particularly, the present disclosure relates to a method and a system for detecting anomalies in powerpack(s).
BACKGROUND
Battery packs serve as the fundamental energy storage components in electric vehicles, energy storage systems, and portable electronic devices, where performance, safety, and traceability are critical operational requirements. As battery systems become increasingly modular and interchangeable, the need for robust identification and integrity verification mechanisms has grown substantially. Specifically, ensuring that a battery pack is genuine, untampered, and composed of its original constituent cells is essential for maintaining operational reliability, preventing unauthorized modifications, and enabling secure integration with host systems. Further, advancements in smart battery management and connected mobility have introduced new challenges related to counterfeit detection, cell replacement monitoring, and lifecycle authentication.
Conventional battery identification systems rely primarily on static metadata stored within the Battery Management System (BMS), such as serial numbers, production codes, or firmware-assigned unique identifiers. Specifically, such identifiers are electronically stored in non-volatile memory during manufacturing and retrieved for traceability, warranty enforcement, or operational logging. In advanced implementations, the identifier may be cryptographically signed or linked to a vehicle’s onboard system, providing a form of logical security. Further, certain systems incorporate passive hardware tags (EEPROMs or RFID chips) embedded within the battery pack enclosure to externally read identification data without powering the cells. Furthermore, some approaches monitor pack-level performance metrics such as voltage under load, state-of-charge profiles, or thermal gradients, which are used for indirect verification of pack integrity over time. More specifically, battery integrity verification includes periodic measurement of voltage, current, and temperature profiles during standard charge-discharge cycles, comparing the measured values against predefined operational thresholds. The data collected from the BMS during normal operation is analysed to detect deviations from expected behavior, often using threshold-based rule engines or simple statistical models. Further, techniques such as Coulomb counting and open-circuit voltage estimation are used to assess degradation or abnormal operation.
However, there are certain underlying problems associated with the above-mentioned existing mechanism for detecting at least one anomaly in a battery pack. For instance, conventional approaches lack intrinsic physical validation of the battery pack identity. Specifically, static identifiers stored in firmware are susceptible to tampering, duplication, or spoofing through software manipulation, firmware overwriting, or hardware cloning techniques. Further, there exists no binding between the stored identifier and the underlying electrochemical configuration of the battery cells, making conventional systems ineffective against sophisticated counterfeiting or cell substitution attacks. Furthermore, operational measurements based on bulk electrical parameters fail to capture cell-level micro-variations that emerge from manufacturing, aging, or structural differences. As a result, conventional systems cannot reliably distinguish between an authentic pack and a maliciously altered one once the static metadata is compromised or in case substituted cells mimic nominal electrical behavior.
Therefore, there exists a need for a mechanism for detecting at least one anomaly in a battery pack that is efficient and overcomes one or more problems as mentioned above.
SUMMARY
An object of the present disclosure is to provide a system for detecting anomalies in a powerpack.
Another object of the present disclosure is to provide a method for detecting anomalies in a powerpack.
Yet another object of the present disclosure is to provide a system and method for detecting anomalies in a powerpack via physically-derived identity for a battery pack based on transient electrical characteristics.
In accordance with a first aspect of the present disclosure, there is provided a system for detecting at least one anomaly in a battery pack, wherein the battery pack comprises a plurality of battery cells, the system comprising:
- an excitation module configured to sequentially inject a predefined current pulse into each battery cell of the battery pack;
- a processing unit coupled to the excitation module, the processing unit comprising:
- a measurement module communicably coupled to the excitation module and configured to measure, for each battery cell, a plurality of transient electrical characteristics; and
- an arbitration module communicably coupled to the measurement module and configured to receive the plurality of transient electrical characteristics;
- an identity storage module operatively coupled to the processing unit; and
- a Battery Management System (BMS) operatively coupled to the identity storage unit,
wherein the arbitration module is configured to generate a unique identity of the battery pack based on a deterministic function of the transient electrical characteristics, and the BMS is configured to regenerate the unique identity after a predefined period to detect at least one anomaly based on a mismatch with the generated unique identity.
The system and method for detecting at least one anomaly in a battery pack, as described in the present disclosure, are advantageous in terms of enabling a hardware-rooted, physically unclonable identity for each battery pack, derived from inherent cell-level electrochemical properties, with safety from spoofed, overwritten, or externally manipulated. Specifically, the use of transient electrical characteristics captured through controlled excitation ensures high-resolution differentiation between cells, providing robust identity generation. Further, the integration of identity regeneration within the Battery Management System (BMS) allows for periodic integrity checks throughout the operational lifecycle of the battery, enabling real-time detection of anomalies such as unauthorized cell replacement, tampering, or degradation beyond acceptable limits. Furthermore, the approach eliminates reliance on static digital identifiers, enhances traceability, and strengthens security in battery-swapping ecosystems, connected electric vehicles, and safety-critical energy systems.
In accordance with another aspect of the present disclosure, there is provided a method of detecting at least one anomaly in a battery pack, wherein the battery pack comprises a plurality of battery cells, the method comprises:
- injecting a predefined current pulse into each battery cell of the battery pack, via an excitation module;
- measuring, for each battery cell, a plurality of transient electrical characteristics, via a measurement module;
- generating a characteristic vector for each battery cell based on the plurality of transient electrical characteristics, via an arbitration module;
- applying a cryptographic hash function to a composite vector to generate the unique identity for the battery pack, via an arbitration module; and
- comparing a regenerated identity with the stored unique identity, via a battery management system.
Additional aspects, advantages, features, and objects of the present disclosure would be made apparent from the drawings and the detailed description of the illustrative embodiments constructed in conjunction with the appended claims that follow.
It will be appreciated that features of the present disclosure are susceptible to being combined in various combinations without departing from the scope of the present disclosure as defined by the appended claims.
BRIEF DESCRIPTION OF DRAWINGS
The summary above, as well as the following detailed description of illustrative embodiments, is better understood when read in conjunction with the appended drawings. For the purpose of illustrating the present disclosure, exemplary constructions of the disclosure are shown in the drawings. However, the present disclosure is not limited to specific methods and instrumentalities disclosed herein. Moreover, those in the art will understand that the drawings are not to scale. Wherever possible, like elements have been indicated by identical numbers.
Embodiments of the present disclosure will now be described, by way of example only, with reference to the following diagrams wherein:
Figure 1 illustrates a block diagram of a system for detecting at least one anomaly in a battery pack, in accordance with an embodiment of the present disclosure.
Figure 2 illustrates a flow chart of a method of detecting at least one anomaly in a battery pack, in accordance with another embodiment of the present disclosure.
In the accompanying drawings, an underlined number is employed to represent an item over which the underlined number is positioned or an item to which the underlined number is adjacent. A non-underlined number relates to an item identified by a line linking the non-underlined number to the item. When a number is non-underlined and accompanied by an associated arrow, the non-underlined number is used to identify a general item at which the arrow is pointing.
DETAILED DESCRIPTION
The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practicing the present disclosure are also possible.
As used herein, the terms “battery pack”, “battery”, and “power pack” are used interchangeably and refer to a structured assembly comprising a plurality of interconnected electrochemical cells, each housed within a unified mechanical and electrical enclosure configured to deliver collective power output. Specifically, each cell within the battery pack exhibits intrinsic physical and electrochemical properties arising from manufacturing variations, internal materials, and geometric configuration. Further, the battery pack includes necessary structural, thermal, and electrical interfaces to enable controlled excitation and measurement of individual cell behavior. Furthermore, the pack integrates auxiliary electronics such as sensing circuits, connectors, and control buses that permit communication with external modules. The battery pack serves as the physical substrate upon which a unique electrical fingerprint is extracted, with each cell contributing a distinct response signature during identity generation. The types of battery packs applicable to the invention include lithium-ion, lithium iron phosphate, nickel-metal hydride, and other rechargeable multi-cell chemistries arranged in series-parallel configurations. Specifically, variations in cell geometry, such as cylindrical, pouch, or prismatic formats, introduce measurable differences in transient response under controlled electrical excitation. Further, cell interconnection topologies influence pack-level impedance and dynamic performance, thereby modifying the overall identity vector. Furthermore, packs with passive balancing or active cell management architectures exhibit differing electrical profiles suitable for identity differentiation. The technique associated with the battery pack includes injecting a predefined current waveform into each individual cell using an excitation module, measuring transient characteristics via a measurement module, and processing the results in an arbitration module to generate a unique identity. Subsequently, the battery management system regenerates the identity after a defined usage period and compares the result with a previously stored identity to detect deviations indicative of physical or chemical anomalies within the pack.
As used herein, the terms “battery cells”, “cells”, and “power cells” are used interchangeably and refer to a single electrochemical unit comprising an anode, cathode, electrolyte, and separator, configured to convert chemical energy into electrical energy through controlled redox reactions. Specifically, each cell is characterized by inherent physical properties such as internal resistance, capacitance, impedance spectrum, and transient voltage behavior that originate from its material composition, geometric configuration, and manufacturing tolerances. Further, the electrochemical dynamics within a cell, including ion diffusion rates, electrode porosity, and SEI (solid electrolyte interphase) formation, introduce unique temporal response characteristics under electrical excitation. Furthermore, surface morphology, current collector interface quality, and electrolyte homogeneity influence the measured output when subjected to low-amplitude diagnostic pulses. Each battery cell functions as the fundamental unit for generating input data for identity computation, where variations in cell-level behavior are preserved through a structured measurement and vector formation process. The Types of battery cells relevant to the present invention include lithium-ion cells based on graphite-LFP, NMC, or LCO chemistries, as well as nickel-metal hydride and lithium polymer formats. Specifically, each chemistry yields distinct transient responses owing to differing ionic mobility, open-circuit voltage profiles, and charge-transfer kinetics. Further, cells may be classified by form factor, including cylindrical (such as 18650, 21700), prismatic, or pouch types, each offering different thermal and dynamic behavior under excitation. Furthermore, manufacturing processes such as winding, stacking, or lamination introduce subtle but measurable inconsistencies across cells, which serve as a basis for distinguishing them electrochemically. The functioning of each battery cell within the invention involves sequential excitation using a predefined current pulse, during which the cell’s response in terms of voltage rise time, current decay, and residual charge behavior is captured. Subsequently, the acquired response data is converted into a numerical vector that feeds into an arbitration function, enabling the generation of a unique identity for the entire pack based on the collective behavior of its constituent cells.
As used herein, the terms “excitation module” and “current injection unit” are used interchangeably and refer to a hardware and software component configured to receive user-defined trip parameters, including a start location and a destination to a dedicated electrical subsystem configured to apply a controlled input stimulus to each battery cell within a battery pack for the purpose of extracting characteristic response data. Specifically, the excitation module generates a predefined current pulse or voltage waveform with known amplitude, duration, and timing parameters, which is sequentially injected into individual cells under controlled conditions. Further, the excitation is performed during an electrically quiescent period to isolate the intrinsic response of each cell from system-level noise or load-induced disturbances. Furthermore, the excitation waveform is designed to be non-intrusive, preserving cell integrity while eliciting transient behaviors that are sensitive to internal cell structure, electrochemical composition, and manufacturing-induced variability. The excitation module includes switching logic, waveform generation circuits, and routing control mechanisms to direct the signal across selected cells in a time-controlled sequence. The Types of excitation waveforms used in the excitation module include step current pulses, sinusoidal bursts, pseudo-random binary sequences (PRBS), and rectangular voltage-controlled pulses. Specifically, rectangular waveforms are preferred for simplicity and high temporal resolution in capturing transient response characteristics such as rise time, impedance lag, or decay slopes. Further, programmable waveform generators embedded in the excitation module enable on-demand variation of amplitude and frequency to support diagnostic re-scans during in-field operation. Furthermore, the excitation module interfaces with the measurement module to synchronize waveform injection and data acquisition, ensuring phase-aligned response capture. The functioning of the excitation module within the system involves initiating a stimulus, activating corresponding switching elements to target a specific cell, maintaining waveform fidelity during injection, and subsequently triggering the measurement module to record the response. The coordinated sequence forms the foundation for the downstream computation of cell-specific vectors used in generating the unique identity of the battery pack.
As used herein, the terms “processing unit” and “processor” are used interchangeably and refer to an integrated hardware and logic subsystem configured to coordinate excitation, measurement acquisition, signal conditioning, and identity computation operations within a battery pack analysis system. Specifically, the processing unit manages the temporal sequence of current pulse injection into individual battery cells and orchestrates the corresponding data capture from each cell’s transient electrical response. Further, the processing unit includes embedded computational logic, memory registers, signal preprocessing circuits, and interface buses that enable real-time communication between the excitation module, measurement module, and arbitration module. Furthermore, the processing unit serves as the central controller responsible for executing predefined measurement protocols and applying deterministic algorithms for identity generation based on measured cell responses. The types of processing units applicable to the invention include microcontroller-based embedded systems, Field-Programmable Gate Arrays (FPGAs), or Digital Signal Processors (DSPs), depending on the complexity and latency requirements of the system. Specifically, the processing unit includes firmware routines that manage excitation sequencing, sampling synchronization, noise filtering, and digital transformation of analog response signals into numerical vectors. Further, high-resolution analog-to-digital conversion and time-synchronized data acquisition modules are integrated into the processing architecture to ensure accuracy and repeatability. Subsequently, the processing unit routes the extracted response vectors to the arbitration module for execution of the identity generation function and communicates the output to the identity storage module and battery management system. The processing unit thereby ensures deterministic and reproducible generation of the unique battery pack identity by maintaining strict control over the diagnostic measurement process.
As used herein, the term “predefined current pulse” refers to a precisely controlled electrical stimulus characterized by a fixed set of parameters including amplitude, duration, rise time, and fall time, configured to elicit a measurable transient response from an electrochemical system. Specifically, the current pulse is digitally or analogically synthesized by a waveform generator within the excitation module, with parameters calibrated to ensure consistency across multiple measurements and minimal interference with the electrochemical stability of the battery cells. Further, the current pulse is applied under quiescent or idle electrical conditions to isolate the intrinsic behavior of each cell, thereby enabling accurate detection of variations in cell-level dynamic response characteristics. Furthermore, the predefined nature of the pulse ensures reproducibility across manufacturing and operational environments, providing a consistent excitation profile for generating comparative electrical fingerprints. The types of predefined current pulses include rectangular pulses, step inputs, exponentially decaying pulses, and low-duty-cycle bursts, each selected based on the desired diagnostic resolution and cell chemistry under test. Specifically, rectangular current pulses are preferred for their abrupt onset and well-defined duration, which enable precise measurement of voltage rise time, impedance artifacts, and relaxation behavior. Further, pulse parameters are selected such that the injected energy remains below thresholds that would induce thermal or electrochemical stress, preserve cell integrity while ensuring response fidelity. Subsequently, the transient behavior elicited by the predefined current pulse is captured by the measurement module in real time and processed to extract electrical characteristics such as peak amplitude, decay slope, and residual voltage. The extracted characteristics are used to generate a feature vector that contributes to the computation of a unique identity for the battery pack via the arbitration module.
As used herein, the term “measurement module” refers to a dedicated subsystem configured to capture, digitize, and preprocess the electrical response of each battery cell following the injection of a predefined current pulse, with the objective of extracting transient electrical characteristics for identity generation. Specifically, the measurement module is coupled to the excitation module to operate in a time-synchronized manner, enabling acquisition of voltage, current, or impedance waveforms immediately following stimulus application. Further, the measurement module comprises high-resolution Analog-to-Digital Converters (ADCs), signal conditioning circuitry, and sampling controllers to preserve fidelity of transient features such as, but not limited to, rise time, decay rate, overshoot, and steady-state response. Furthermore, the measurement module performs initial filtering and normalization to compensate for measurement noise, temperature drift, and environmental variations, thereby enhancing consistency across measurement cycles. The types of measurable parameters within the scope of the measurement module include peak current amplitude, transient voltage response time, residual voltage decay, and impedance characteristics, each of which reflects the physical and electrochemical state of the battery cell. Specifically, the module supports multi-channel acquisition to monitor multiple cells either sequentially or in parallel, depending on the system architecture. Further, configurable sampling rates and resolution settings enable adaptation of the measurement process to cell chemistries and waveform characteristics. Subsequently, the measurement module outputs a vector of characteristic values for each cell, which is forwarded to the arbitration module for identity computation. The precise and repeatable functioning of the measurement module directly influences the uniqueness and reliability of the generated battery pack identity.
As used herein, the terms “transient electrical characteristics” and “electrical characteristics” are used interchangeably and refer to time-dependent electrical response parameters exhibited by an electrochemical cell immediately following the application of a predefined current pulse, reflecting the intrinsic physical and chemical properties of the cell structure. Specifically, the characteristics include dynamic changes in voltage, current, and impedance that occur over short durations, typically in the millisecond to second range, and are influenced by factors such as electrode composition, electrolyte behavior, internal resistance, and interfacial charge transfer dynamics. Further, transient electrical characteristics are highly sensitive to internal construction tolerances, cell degradation states, and manufacturing-induced microvariations, making them suitable for identity extraction and anomaly detection in multi-cell battery systems. Furthermore, the reproducibility of the characteristics under standardized excitation conditions ensures consistent fingerprinting across production and in-field environments. The types of transient electrical characteristics include, but are not limited to, transient voltage rise time, which denotes the time taken for the voltage to reach a predefined threshold after pulse initiation; peak current amplitude, which reflects the instantaneous current delivered and internal resistance of the cell; residual charge decay rate, which indicates the cell relaxation quickness to a quiescent state; and impedance response, which characterizes frequency-dependent resistance to charge movement. Specifically, the parameters form a multidimensional feature space that captures the unique electrochemical signature of each cell. Further, derived metrics such as phase shift, response overshoot, or relaxation curvature may be used to enhance fingerprint granularity. Subsequently, the extracted transient electrical characteristics are compiled into characteristic vectors and processed by the arbitration module to compute a unique and reproducible identity value for the battery pack.
As used herein, the term “arbitration module” refers to a dedicated processing unit configured to generate a unique identity of the battery pack based on a deterministic function of transient electrical characteristics measured from each battery cell. Specifically, the module receives a plurality of transient responses comprising electrical, electrochemical, or dynamic parameters sequentially collected from the injection of a predefined current pulse into each battery cell. The arbitration module further constructs a characteristic vector for each cell by processing the received parameters and subsequently applies a weighting function to the vectors to compute a composite vector. Consequently, the arbitration module applies a cryptographic hash function to the resulting composite vector to generate an identity that is unique to the electrical and electrochemical signature of the battery pack under normal operating conditions. The types of operations executed by the arbitration module include deterministic transformation, vector computation, linear combination with applied weighting, and cryptographic hashing. Each operation is sequentially integrated within the processing workflow to ensure consistent and repeatable identity generation. Furthermore, the arbitration module facilitates secure fingerprinting of the battery pack by ensuring that even minor deviations in transient response, resulting from cell degradation or unauthorized replacement, lead to a change in the generated identity. Subsequently, the identity storage module retains the computed unique identity, and the BMS periodically instructs regeneration of the same through the arbitration module. A mismatch detected in this regenerated identity, when compared to the stored reference, signifies an anomaly within the pack structure, integrity, or operational behavior.
As used herein, the terms “identity storage module” and “storage module” are used interchangeably and refer to a dedicated hardware memory unit operatively coupled to the processing unit and configured to store a unique identity of the battery pack generated by the arbitration module. Specifically, the module receives the output of a cryptographic hash function applied to a composite vector derived from a weighted combination of transient electrical characteristics of the battery cells. The identity storage module further transmits a copy of the stored unique identity to the Battery Management System (BMS), enabling distributed and synchronized anomaly detection operations. Furthermore, the storage architecture ensures persistent retention of the identity across power cycles, preserving the integrity of the stored fingerprint for long-term validation. The types of functions executed by the identity storage module include secure write operations, indexed retrieval, and time-correlated identity access. Each function is sequentially structured to support deterministic anomaly detection based on identity mismatch. Subsequently, after a predefined number of charge-discharge cycles, the BMS initiates a request to regenerate the identity through the arbitration module, which is then compared with the stored reference. Consequently, the identity storage module enables a closed-loop authentication mechanism for the battery pack by ensuring the availability and reliability of the original fingerprint. This configuration ensures that any structural degradation, unauthorized cell replacement, or tampering within the battery pack is traceable through deviations in the regenerated identity.
As used herein, the terms “Battery Management System” and “BMS” are used interchangeably and refer to a control and monitoring unit operatively coupled to the identity storage module and configured to manage the anomaly detection process through identity verification. Specifically, the BMS receives a copy of the unique identity generated by the arbitration module and retained by the identity storage module. The BMS further stores this identity and initiates regeneration of the same after a predefined number of charge-discharge cycles. Furthermore, the BMS performs a comparison between the regenerated identity and the stored reference to determine the presence of any anomaly in the battery pack structure or behavior. The types of operations performed by the BMS include identity acquisition, schedule-based identity regeneration, mismatch detection, event logging, and enforcement of control protocols. Each operation is sequentially orchestrated to maintain the integrity and operational safety of the battery pack. Subsequently, upon detecting a mismatch between the regenerated and stored identities, the BMS logs the anomaly detection event in system memory and issues control commands to restrict further charging or discharging operations. Consequently, the BMS ensures secure and autonomous health monitoring of the battery pack by leveraging identity-based fingerprinting and periodic authentication cycles, thereby preventing unauthorized use or operation under unsafe conditions.
As used herein, the terms “voltage-controlled waveform” and “waveform” are used interchangeably and refer to a predefined electrical excitation profile characterized by a controlled amplitude and duration, used by the excitation module to inject current pulses into individual battery cells. Specifically, the waveform is defined by a voltage-driven shape that governs the current injection profile, enabling consistent and repeatable stimulation of each battery cell in the battery pack. Further, the voltage-controlled waveform ensures that the transient electrical characteristics measured in response remain deterministic and sensitive to variations in internal cell properties, facilitating accurate identity generation by the arbitration module. The types of voltage-controlled waveforms include step, ramp, exponential, or custom pulse shapes, each selected based on the required sensitivity to particular electrical or electrochemical cell parameters. The excitation module generates the waveforms under precise timing control to maintain uniformity across cell stimulations. Subsequently, the controlled current pulse resulting from the waveform interacts with the internal impedance and dynamic properties of each cell, producing a measurable transient response. Consequently, the use of voltage-controlled waveforms standardizes the excitation process across the battery pack, ensuring that deviations in response directly reflect underlying anomalies or changes in cell behavior, which are later captured by the measurement module and utilized in the identity generation process.
As used herein, the term “electrical parameter” refers to a quantifiable property of each battery cell that characterizes its electrical behavior in response to an injected voltage-controlled current pulse. Specifically, the electrical parameter includes measurable quantities such as, but not limited to, voltage, current, resistance, capacitance, and impedance, which are captured by the measurement module immediately after excitation. Further, the parameters serve as foundational inputs for constructing characteristic vectors that represent the transient electrical identity of each cell. Furthermore, variations in electrical parameters under controlled excitation reflect internal cell conditions, thereby enabling the arbitration module to distinguish normal behavior from anomalies. The Types of electrical parameters include, but not limited to, steady-state voltage, instantaneous current, internal resistance, differential capacitance, and complex impedance across multiple frequencies. The measurement module records the parameters over a predefined observation window following each excitation event. Subsequently, the collected electrical parameters are processed to form feature vectors for each cell, which are weighted and combined into a composite vector. Consequently, accurate extraction of electrical parameters under controlled stimulation conditions provides a deterministic basis for identity generation and anomaly detection, supporting the system's ability to monitor cell health and detect deviations arising from aging, imbalance, or structural degradation.
As used herein, the term “electrochemical parameter” refers to an intrinsic property of each battery cell that reflects the state of internal electrochemical processes during and after the injection of a voltage-controlled current pulse. Specifically, the electrochemical parameter encompasses values such as, but not limited to, state of charge, diffusion coefficient, charge transfer resistance, double-layer capacitance, and open-circuit voltage behavior. Further, the parameters are indirectly inferred from the time-dependent response of the cell to the excitation pulse and are processed by the measurement module to extract transient features. Furthermore, the electrochemical behavior provides a unique signature linked to the cell’s chemistry, material condition, and degradation profile, which enhances the distinctiveness of the generated identity. The types of electrochemical parameters include kinetic and thermodynamic characteristics associated with ion transport, electrode reaction rates, and interfacial dynamics. The measurement module evaluates the parameters through analysis of response curves, including voltage relaxation, impedance profiles, and derivative metrics following excitation. Subsequently, the extracted electrochemical parameters are embedded within the characteristic vector for each cell and contribute to the final composite vector used by the arbitration module. Consequently, integration of electrochemical parameters into the identity generation process allows detection of subtle anomalies arising from changes in electrode composition, electrolyte degradation, or lithium plating effects, thereby enabling comprehensive health monitoring of the battery pack.
As used herein, the term “dynamic response parameter” refers to a time-dependent behavioral characteristic of a battery cell captured during its response to a voltage-controlled excitation pulse. Specifically, the dynamic response parameter includes temporal features such as, but not limited to, rise time, settling time, time constant, overshoot, and rate of voltage or current change, which are derived from the transient profile recorded by the measurement module. Further, the parameters reflect the dynamic stability and reaction kinetics of each cell under controlled stimulation, offering high sensitivity to internal inconsistencies or aging effects. Furthermore, dynamic response behavior provides a non-linear signature that complements electrical and electrochemical features for accurate fingerprinting of each cell. The types of dynamic response parameters include first-order and higher-order time constants, transient recovery profiles, and derivative-based response slopes calculated over specific time intervals. The measurement module continuously records the evolving output during and after pulse injection and extracts the parameters through signal processing techniques. Subsequently, the derived dynamic response parameters are integrated into the characteristic vector representing each cell and contribute to the composite vector used by the arbitration module for identity generation. Consequently, any shift in dynamic response due to cell degradation, internal short formation, or impedance buildup becomes detectable through identity mismatch, enabling early anomaly detection and predictive health management of the battery pack.
As used herein, the term “characteristic vector” refers to a structured numerical representation generated for each battery cell based on the extracted transient electrical characteristics resulting from a controlled excitation pulse. Specifically, the characteristic vector comprises a set of quantified values corresponding to electrical parameters, electrochemical parameters, and dynamic response parameters measured within a defined observation window. Further, the arbitration module constructs the vector by aggregating the measured features into a deterministic format that preserves the unique behavioral signature of each cell. Furthermore, the characteristic vector enables consistent comparison, mathematical manipulation, and transformation required for identity generation. The types of components embedded within the characteristic vector include normalized values of voltage response, internal resistance, charge transfer kinetics, time constants, and differential response rates. Each vector element reflects a distinct aspect of the cell’s transient behavior under excitation. Subsequently, the arbitration module applies a weighting function to the set of characteristic vectors corresponding to all cells and computes a composite vector through a linear combination. Consequently, the characteristic vector serves as a foundational unit for the identity generation process, ensuring that subtle variations across individual cells contribute to the final fingerprint used for anomaly detection within the battery management framework.
As used herein, the term “weighting function” refers to a mathematical operation applied by the arbitration module to assign relative importance to each element of the characteristic vectors corresponding to individual battery cells. Specifically, the weighting function modifies the influence of each transient electrical characteristic comprising electrical, electrochemical, and dynamic response parameters based on predefined criteria such as, but not limited to, sensitivity, diagnostic relevance, or variability. Further, the weighting function transforms the set of characteristic vectors into a format that emphasizes features with higher anomaly detection potential while suppressing noise or less significant attributes. Furthermore, the application of the weighting function enables the arbitration module to generate a composite vector that accurately reflects the global behavior of the battery pack. The types of weighting functions include scalar multipliers, normalized weighting coefficients, statistical weights derived from variance analysis, and adaptive weights based on the historical behavior of the battery pack. Each weighting operation adjusts the amplitude of the corresponding vector elements before aggregation. Subsequently, the arbitration module computes a linear combination of the weighted characteristic vectors to form a unified composite vector representative of the overall transient response of the battery pack. Consequently, the use of a weighting function ensures that critical signal components dominate the identity generation process, enhancing the resolution and reliability of anomaly detection through the subsequent application of the cryptographic hash function.
As used herein, the term “composite vector” refers to an aggregated numerical representation generated by the arbitration module through a linear combination of weighted characteristic vectors obtained from individual battery cells. Specifically, the composite vector encapsulates the collective transient electrical behavior of the entire battery pack by integrating electrical, electrochemical, and dynamic response parameters across all cells. Further, each characteristic vector is first subjected to a weighting function to emphasize diagnostically significant features before inclusion in the aggregation process. Furthermore, the composite vector forms a deterministic and high-resolution fingerprint that captures the pack-level identity with sensitivity to cell-level variations. The types of composite vectors vary based on the dimensionality and structure of the underlying characteristic vectors and the nature of the applied weighting scheme. The arbitration module executes vector summation or matrix-based combination to generate a singular vector representing the battery pack’s operational signature. Subsequently, the composite vector serves as the input to a cryptographic hash function, resulting in the generation of a unique identity for the battery pack. Consequently, the composite vector plays a central role in identity formation, ensuring that any deviation in the measured characteristics of one or more cells propagates through the vector structure, enabling reliable detection of anomalies during subsequent identity regeneration cycles initiated by the Battery Management System.
As used herein, the term “cryptographic hash function” refers to a deterministic algorithm applied by the arbitration module to transform the composite vector into a fixed-length unique identity representing the battery pack. Specifically, the cryptographic hash function processes the numerical contents of the composite vector and generates a non-reversible output that uniquely corresponds to the internal state of the battery pack at the time of measurement. Further, the hash function ensures collision resistance, meaning distinct composite vectors originating from different transient electrical behaviors produce distinct identity outputs. Furthermore, the cryptographic nature of the function provides security and integrity, preventing reconstruction or manipulation of the original measurement data from the generated identity. The types of cryptographic hash functions suitable for implementation include SHA-256, SHA-3, BLAKE2, or equivalent algorithms conforming to industry standards for data integrity and authentication. The arbitration module executes the selected hash function on the composite vector after the weighting and aggregation stages have been completed. Subsequently, the output of the hash function is stored as the unique identity in the identity storage module and transmitted to the Battery Management System for future reference. Consequently, the cryptographic hash function ensures that any deviation in the transient behavior of the battery cells due to degradation, tampering, or replacement results in a mismatch during identity regeneration, enabling reliable and secure anomaly detection within the defined system architecture.
In accordance with a first aspect of the present disclosure, there is provided a system for detecting at least one anomaly in a battery pack, wherein the battery pack comprises a plurality of battery cells, the system comprising:
- an excitation module configured to sequentially inject a predefined current pulse into each battery cell of the battery pack;
- a processing unit coupled to the excitation module, the processing unit comprising:
- a measurement module communicably coupled to the excitation module and configured to measure, for each battery cell, a plurality of transient electrical characteristics; and
- an arbitration module communicably coupled to the measurement module and configured to receive the plurality of transient electrical characteristics;
- an identity storage module operatively coupled to the processing unit; and
- a Battery Management System (BMS) operatively coupled to the identity storage unit,
wherein the arbitration module is configured to generate a unique identity of the battery pack based on a deterministic function of the transient electrical characteristics, and the BMS is configured to regenerate the unique identity after a predefined period to detect at least one anomaly based on a mismatch with the generated unique identity.
Referring to figure 1, in accordance with an embodiment, there is described a system 100 for detecting at least one anomaly in a battery pack 102, wherein the battery pack 102 comprises a plurality of battery cells 104 (104A…104N). The system 100 comprises an excitation module 106 configured to sequentially inject a predefined current pulse into each battery cell 104 of the battery pack 102; a processing unit 108 coupled to the excitation module 106. The processing unit 108 comprises a measurement module 110 communicably coupled to the excitation module 106 and configured to measure, for each battery cell 104, a plurality of transient electrical characteristics, and an arbitration module 112 communicably coupled to the measurement module 110 and configured to receive the plurality of transient electrical characteristics. The system 100 further comprises an identity storage module 114 operatively coupled to the processing unit and a Battery Management System (BMS) 116 operatively coupled to the identity storage module 114. Further, the arbitration module 112 is configured to generate a unique identity of the battery pack 102 based on a deterministic function of the transient electrical characteristics, and the BMS 116 is configured to regenerate the unique identity after a predefined period to detect at least one anomaly based on a mismatch with the generated unique identity.
The system 100 for detecting at least one anomaly in a battery pack 102 operates by generating a secure, deterministic identity of the pack based on transient electrical characteristics of individual battery cells 104. An excitation module 106 sequentially injects a predefined current pulse into each cell 104, using a voltage-controlled waveform with fixed amplitude and duration to ensure uniform excitation across the battery pack 102. A processing unit 108, comprising a measurement module 110 and an arbitration module 112, captures the resulting transient electrical characteristics, such as, but not limited to, voltage response, impedance, and dynamic behavior of each cell. The measurement module 110 records the characteristics with high temporal resolution, enabling detailed profiling of each cell’s electrical, electrochemical, and dynamic properties. The arbitration module 112 further constructs a characteristic vector for each cell using the measured parameters and applies a weighting function to emphasize features with greater diagnostic significance. A composite vector is subsequently generated through a linear combination of the weighted characteristic vectors, which serves as an intermediate representation of the overall battery pack behavior. Furthermore, applying a cryptographic hash function, such as SHA-256, to the composite vector produced by the arbitration module 112. The hash function transforms the composite vector into a fixed-length, non-reversible digital signature defined as the unique identity of the battery pack 102. The identity is stored in an identity storage module 114, which maintains the integrity and persistence of the data across multiple charge-discharge cycles. Further, a copy of the identity is also transmitted to a Battery Management System (BMS) 116, which manages identity regeneration. After a predefined number of charge cycles, the BMS 116 initiates a new excitation and measurement sequence, repeating the identity generation process. The regenerated identity is compared to the previously stored reference. A mismatch between the two identities indicates the presence of at least one anomaly in the battery pack, such as internal degradation, unbalanced aging, or unauthorized cell replacement. Specifically, the system 100 leverages deterministic identity generation mechanisms to provide robust tamper detection and operational integrity verification. Further, by converting raw transient electrical data into a cryptographically secure identity, the system 100 eliminates reliance on traditional threshold-based monitoring. Furthermore, the identity-based method supports scalable implementation across varied battery configurations without requiring real-time human calibration. Subsequently, the approach enables predictive maintenance, enhances battery safety, and reduces operational risks. Consequently, the invention delivers improved lifecycle management, early failure detection, and increased trust in battery pack authenticity and performance.
In an embodiment, the excitation module 106 is configured to inject the predefined current pulse with a voltage-controlled waveform, wherein the voltage-controlled waveform comprises a predefined amplitude and duration. The system 100 utilizes an excitation module 106 configured to inject a predefined current pulse into each battery cell 104 using a voltage-controlled waveform with fixed amplitude and duration. Specifically, the waveform is generated based on a reference voltage profile that dictates the shape, intensity, and timing of the current pulse delivered to each cell 104 in the battery pack 102. The excitation module 106 applies the waveform sequentially across all cells, ensuring that each cell experiences identical excitation conditions for accurate and comparative measurement. Further, the control parameters of the waveform, but not limited to, such as peak voltage, rise time, and pulse width, are tuned to stimulate a measurable transient response without inducing degradation or triggering safety mechanisms. Furthermore, the controlled nature of the waveform allows consistent input stimuli for each measurement cycle, forming the basis for deterministic identity generation. For instance, generating a trapezoidal voltage-controlled waveform where the excitation module 106 ramps up the voltage to a predefined peak, holds the peak steady for a specific duration, and then ramps down to zero. The resulting current response from each cell 104 is recorded by the measurement module 110, capturing voltage drop, impedance variation, and dynamic reaction parameters in real time. The shape and duration of the waveform are calibrated to trigger electrochemical activity within safe operational limits, allowing accurate extraction of parameters such as internal resistance, diffusion coefficients, and charge transfer dynamics. Subsequently, the measurement data from each waveform application contributes to a unique characteristic vector that reflects the individual cell’s behavior. Consequently, the responses form the input to the arbitration module 112, where the process of composite vector formation and identity generation is initiated. The technical effect of using a voltage-controlled waveform with predefined amplitude and duration includes improved repeatability and measurement precision during transient characterization of the battery cells 104. Specifically, uniform stimulation ensures that variations in the transient response are attributable solely to internal cell properties, not inconsistencies in excitation. Further, the voltage-controlled approach enables fine-tuned control of current injection, avoiding abrupt electrical stress that may otherwise distort the response. Furthermore, the waveform profile is selected to reveal both rapid dynamic effects and slower electrochemical phenomena, providing a comprehensive basis for anomaly detection. Subsequently, the injection leads to robust system behavior across temperature, aging, and operational variances. Consequently, the system 100 achieves higher diagnostic sensitivity, extended pack safety, and consistent performance in identity-based monitoring and anomaly detection applications.
In an embodiment, the plurality of transient electrical characteristics measured by the measurement module 110 comprises at least one of an electrical parameter, an electrochemical parameter, or a dynamic response parameter measured for each battery cell 104. The system 100 incorporates the measurement module 110 configured to capture a plurality of transient electrical characteristics for each battery cell 104 following excitation by the excitation module 106. Specifically, the measured characteristics include electrical parameters such as terminal voltage, instantaneous current, and internal resistance; electrochemical parameters such as charge transfer resistance, diffusion coefficient, and open-circuit potential behavior; and dynamic response parameters such as rise time, settling time, and transient response slope. Further, each measurement is initiated in response to a predefined current pulse delivered through a voltage-controlled waveform, allowing the measurement module 110 to operate under synchronized and deterministic conditions. Furthermore, the collected data reflects the internal state of each cell 104, providing a comprehensive basis for constructing a characteristic vector used in subsequent identity generation. The voltage-time response of a cell 104 is captured immediately after the application of a current pulse with known amplitude and duration. The measurement module 110 records the initial voltage drop, the slope of the recovery curve, and the time taken to reach a steady state. From the response, electrical parameters such as but not limited to, internal resistance and equivalent series resistance are extracted, electrochemical parameters such as but not limited to, double-layer capacitance and diffusion resistance are inferred, and dynamic response parameters such as, but not limited to, the first-order time constant and overshoot are computed. Subsequently, each parameter is assigned a numerical value and organized into a structured characteristic vector that preserves the cell-specific behavioral signature. The vector is transmitted to the arbitration module 112, contributing to the generation of the composite vector and ultimately the cryptographically secure identity of the battery pack 102. The measuring of the electrical, electrochemical, and dynamic response parameters in a transient regime provides high-resolution insight into the internal health, aging, and performance state of individual cells 104. Specifically, the transient nature of the measurement captures fast and slow processes that are not observable under steady-state conditions. Further, combining multiple parameter categories enables detection of a broad range of failure modes, including electrode degradation, electrolyte imbalance, and mechanical deformation. Furthermore, the deterministic extraction and structuring of the parameters enhance the uniqueness and reproducibility of the generated identity. Subsequently, deviations in any parameter during identity regeneration signal an anomaly in the early stages of degradation. Consequently, the system 100 achieves precise, non-invasive anomaly detection with minimal computational overhead, supporting predictive diagnostics and operational safety in real-world battery applications.
In an embodiment, the arbitration module 112 is configured to generate a characteristic vector for each battery cell 104 based on the plurality of transient electrical characteristics. The system 100 comprises an arbitration module 112 configured to generate a characteristic vector for each battery cell 104 using the plurality of transient electrical characteristics captured by the measurement module 110. Specifically, each characteristic vector includes numerically represented features extracted from electrical, electrochemical, and dynamic response parameters measured under a predefined excitation condition. Further, the arbitration module 112 organizes the parameters into a multi-dimensional vector structure where each element corresponds to a specific property, such as, but not limited to, internal resistance, charge transfer resistance, diffusion time constant, or voltage recovery slope. Furthermore, the vectorization process preserves the temporal and diagnostic fidelity of the measured response, ensuring accurate representation of each cell's internal state. The above-mentioned generation involves processing the voltage-time response of a cell 104 following a current pulse to compute features such as the initial voltage drop (electrical), relaxation curve curvature (electrochemical), and time-to-steady-state (dynamic). The arbitration module 112 assigns each value to a specific index within a fixed-length vector, forming the characteristic vector for that cell. Each vector is normalized to mitigate the effect of amplitude differences while preserving feature distribution. Subsequently, a complete set of characteristic vectors is generated across all battery cells 104 in the pack 102, forming a structured representation of the overall pack behavior at the individual cell level. Consequently, the characteristic vectors serve as the foundation for weighted aggregation and identity generation, enabling accurate anomaly detection. The generation of the characteristic vectors for each cell 104 provides localized defect detection, scalable pack-level identity formation, and enhanced signal resolution. Specifically, vector-based representation allows consistent comparison between cells across cycles, capturing progressive degradation or abnormal deviation. Further, the characteristic vector encapsulates multiple parameters, enhancing robustness against noise and single-feature failure. Furthermore, vectorization enables high-dimensional analysis and subsequent transformation through mathematical functions for identity computation. Subsequently, the use of characteristic vectors ensures modular, interpretable, and secure anomaly detection. Consequently, the system 100 achieves detailed fingerprinting of each cell, supporting cell-specific diagnostics and comprehensive battery pack validation.
In an embodiment, the arbitration module 112 is configured to apply a weighting function to each characteristic vector and compute a composite vector as a linear combination of the weighted characteristic vectors. The arbitration module 112 in the system 100 is configured to apply a weighting function to each characteristic vector and compute a composite vector as a linear combination of the weighted characteristic vectors. Specifically, the weighting function assigns predefined significance values to each element or entire vector based on diagnostic sensitivity, prior fault history, or statistical variability. Further, the process ensures that parameters contributing more effectively to anomaly detection receive higher influence during vector aggregation. Furthermore, after applying the weighting function, the arbitration module 112 performs a summation or matrix-based operation to compute the composite vector, which represents a global signature of the battery pack 102. The above-mentioned application of the weighting function involves assigning weights to each parameter within the characteristic vectors based on their standard deviation across cells or cycles. The arbitration module 112 multiplies each vector element by its respective weight, amplifying signal components that exhibit higher diagnostic correlation. For instance, charge transfer resistance may receive a higher weight in applications involving frequent thermal cycling. Subsequently, the module computes a linear combination across all weighted vectors, forming a single composite vector that captures both cell-level information and pack-level behavior in a compact structure. Consequently, the composite vector maintains diagnostic sensitivity while reducing dimensional complexity, preparing the data for identity generation via hashing. The application of the weighting function and computing a composite vector provides improved accuracy, enhanced anomaly localization, and optimized representation of battery pack condition. Specifically, the weighted aggregation accounts for parameter relevance and minimizes the influence of low-sensitivity or noisy features. Further, the composite vector acts as a compressive signature, retaining essential behavioral traits of the pack while reducing data volume. Furthermore, the weighting mechanism allows configuration-specific customization, enabling adaptation to different chemistries or operational profiles. Subsequently, the vector serves as a consistent and deterministic input for cryptographic hashing. Consequently, the system 100 achieves efficient identity formation with heightened robustness to environmental and structural variability.
In an embodiment, the arbitration module 112 is configured to apply a cryptographic hash function to the composite vector to generate the unique identity for the battery pack 102. The arbitration module 112 is configured to apply a cryptographic hash function to the composite vector to generate the unique identity for the battery pack 102. Specifically, the composite vector serves as the final deterministic representation of the battery pack’s electrical behavior, and the cryptographic hash function transforms this representation into a fixed-length identity string. Further, the hash function ensures one-way transformation, collision resistance, and integrity protection, making the generated identity secure and non-reversible. Furthermore, the output identity serves as a digital fingerprint, capturing the collective health and configuration state of the battery pack 102 at the time of measurement. For instance, the cryptographic hash function application involves applying a SHA-256 hash algorithm to the composite vector after the computation. The arbitration module 112 converts the numerical composite vector into a binary or hexadecimal string and passes the string through the SHA-256 hashing engine. The output is a 256-bit identity value that uniquely corresponds to the current configuration and condition of the battery pack 102. The identity is stored in the identity storage module 114 and transmitted to the Battery Management System 116 for future validation. Subsequently, during identity regeneration cycles, a new identity is computed using the same hashing method and compared with the stored reference to identify anomalies. Consequently, any change in internal parameters that affects the composite vector will result in a different identity, triggering anomaly detection. The use of the cryptographic hash function provides a tamper-proof identity formation, secure authentication of battery state, and immunity to data reconstruction. Specifically, the hash function guarantees that even minor changes in cell behavior lead to a completely different identity value. Further, the fixed-length output allows efficient storage and rapid comparison during regeneration cycles. Furthermore, the use of standard cryptographic algorithms ensures compatibility with secure firmware and cloud-based validation systems. Subsequently, the above-mentioned mechanism enables distributed and verifiable battery authentication across different platforms. Consequently, the system 100 ensures trusted anomaly detection, operational transparency, and long-term traceability of battery health and integrity.
In an embodiment, the identity storage module 114 is configured to store the generated unique identity of the battery pack 102 and transmit a copy of the generated unique identity to the battery management system 116. The identity storage module 114 in the system 100 is configured to store the generated unique identity of the battery pack 102 and transmit a copy of the same to the Battery Management System 116. Specifically, the identity storage module 114 receives the unique identity output from the arbitration module 112, which is derived by applying a cryptographic hash function to the composite vector representing the collective behavior of all battery cells 104. Further, the module performs secure, non-volatile storage of the identity to preserve integrity across charge-discharge cycles and power transitions. Furthermore, a communication protocol is initiated to transmit a verified copy of the identity to the BMS 116, where the verified copy becomes the reference for future comparison during anomaly detection. The mechanism involves the identity storage module 114 initiating a secure write sequence upon receiving the hashed identity from the arbitration module 112. The identity is encoded with a timestamp and stored using ECC (Error Correction Code) protected memory blocks to prevent corruption. A parallel communication interface then transmits a duplicate of the identity to the BMS 116, where the identity is logged into a secure memory segment indexed by cycle count or system uptime. Subsequently, both modules retain synchronized copies of the identity, enabling distributed and redundant validation. Consequently, the stored reference becomes the baseline for determining whether any structural or behavioral changes have occurred in the battery pack 102 over time. The above-mentioned mechanism provides secure identity preservation, fault-tolerant data replication, and preparation for integrity verification cycles. Specifically, storing the identity in both modules ensures redundancy and robustness against data loss or corruption. Further, the identity serves as a static fingerprint that encapsulates the complete transient behavior of the battery pack at the time of measurement. Furthermore, early transmission and synchronization between storage and BMS components facilitate timely anomaly detection in subsequent operational phases. Subsequently, the identity serves as a deterministic benchmark for evaluating future changes. Consequently, the system 100 achieves long-term traceability, enhanced data security, and consistent diagnostic readiness.
In an embodiment, the battery management system 116 is configured to store the copy of the unique identity of the battery pack 102 and initiate regeneration of the unique identity after a predefined number of charge cycles of the battery pack 102. The Battery Management System 116 is configured to store the copy of the unique identity of the battery pack 102 and initiate regeneration of the unique identity after a predefined number of charge cycles. Specifically, the BMS 116 maintains a counter to track the number of full or partial charge-discharge cycles completed by the battery pack 102. Further, upon reaching a predefined threshold, the BMS 116 triggers a new round of excitation, measurement, vector construction, and identity generation using the same deterministic pipeline. Furthermore, the regenerated identity is used for comparison with the previously stored reference to identify any internal or external anomalies. The mechanism involves the BMS 116 incrementing a persistent cycle count register every time a complete charge-discharge cycle is detected. When the cycle count reaches a predefined value, the BMS 116 sends an initiation signal to the excitation module 106, which begins a new excitation sequence across all battery cells 104. The transient responses are measured by the measurement module 110, processed by the arbitration module 112, and hashed into a new identity. Subsequently, the regenerated identity is temporarily stored within the BMS 116 for validation against the original reference identity. Consequently, any significant change in internal cell characteristics will reflect as a change in the regenerated identity, enabling anomaly detection through mismatch analysis. The regeneration of the identity provides scheduled identity verification, periodic condition monitoring, and proactive anomaly detection. Specifically, regeneration after a predefined number of cycles aligns with natural cell aging, allowing degradation tracking without additional hardware. Further, deterministic regeneration ensures reproducibility and minimizes false positives. Furthermore, embedded cycle-based scheduling enables autonomous operation without manual intervention or external triggers. Subsequently, changes due to aging, misuse, or tampering are identified systematically. Consequently, the system 100 supports lifecycle management, enhances reliability, and ensures continued safety compliance throughout battery usage.
In an exemplary embodiment, a system 100 for detecting anomalies in a battery pack 102 is implemented during the manufacturing and operational lifecycle of a high-capacity lithium-ion battery module intended for use in electric vehicles. A 6-cell lithium-ion battery pack installed in a shared electric scooter, where each cell 104 is sequentially excited during factory calibration using a 1A rectangular current pulse of 100 ms duration. The measurement module 110 records, for each cell 104, a peak voltage response (Cell 1: 3.71V, Cell 2: 3.74V, ..., Cell 6: 3.69V), a decay rate (Cell 1: -0.045 V/s), and a time constant associated with relaxation behavior. Further, the above-mentioned parameters form vectors V1 to V6, which are weighted and combined by the arbitration module 112 to form a composite vector that is hashed using SHA-256, yielding a unique 256-bit identity such as “8F4D...7A3B.” The arbitration module 112 applies a set of predetermined weights (W1 through W6) to compute a weighted composite vector: S(W? × V?). The identity is stored both in the battery pack’s 102 secure EEPROM and in the vehicle’s BMS. Subsequently, after 300 charge cycles, the BMS 116 re-executes the same excitation and measurement process. The regenerated identity derived from the current transient values is compared against the stored identity, and a deviation beyond a 5% similarity threshold triggers an anomaly alert, suggesting that Cell 3 may have been replaced or degraded abnormally. As a result, the scooter's control unit restricts fast charging and alerts the fleet operator for inspection, thereby ensuring safety and authenticity of the battery pack.
In an embodiment, the battery management system 116 is configured to compare the regenerated identity with the stored unique identity and detect an anomaly based on a mismatch between the regenerated identity and the stored unique identity. The Battery Management System 116 is configured to compare the regenerated identity with the stored unique identity and detect an anomaly based on a mismatch between the two. Specifically, after completion of a regeneration cycle, the BMS 116 performs a byte-level or hash-level comparison between the newly generated identity and the previously stored reference. Further, a strict comparison protocol ensures that even a single-bit difference is recognized as a potential indicator of abnormal behavior. Furthermore, the detection mechanism is embedded within the identity verification engine of the BMS 116, operating in real-time with minimal processing overhead. The mechanism involves, for instance, retrieving both the regenerated identity and the stored reference into volatile memory blocks. A bitwise XOR operation is executed to detect any mismatch, with the result monitored by a control logic that flags deviations beyond a defined Hamming distance. In case the difference exceeds the acceptable threshold, the BMS 116 classifies the event as an anomaly and logs the corresponding data into memory. Subsequently, an internal flag is raised to indicate failure or degradation, which is used to trigger additional safety or maintenance protocols. Consequently, the system transitions into a diagnostic or restricted operational state depending on severity. The identity mismatch comparison provides high-fidelity anomaly detection, tamper resistance, and early-stage fault identification. Specifically, identity-based comparison eliminates the need for multiple threshold-based metrics and condenses complex behavior into a single validation step. Further, the cryptographic nature of the identity ensures that any substantial deviation in cell condition results in a changed output. Furthermore, the mechanism provides a secure and non-intrusive means to detect internal or external anomalies. Subsequently, the integrity of the battery pack 102 is continuously validated over its operational life. Consequently, the system 100 achieves precise condition monitoring and strengthens its functional safety profile.
In an embodiment, the battery management system 116 is configured to log the anomaly detection event in a memory and restrict charging or discharging operations of the battery pack 102 based on the detected anomaly. The Battery Management System 116 is configured to log the anomaly detection event in memory and restrict charging or discharging operations of the battery pack 102 based on the detected anomaly. Specifically, upon confirming an identity mismatch, the BMS 116 updates the event log with a timestamp, severity level, and diagnostic metadata derived from the identity deviation. Further, a protection protocol is activated to limit energy flow into or out of the battery pack 102 to prevent further degradation or unsafe operation. Furthermore, system-wide control signals are issued to associated power controllers, chargers, or inverters to enforce the operational restrictions. The mechanism involves the BMS 116 writing an event record to a secure non-volatile log that includes the regenerated identity, stored identity, time of mismatch detection, and the operational context. Simultaneously, the BMS 116 evaluates the severity of the anomaly using a predefined classification scheme. In case the anomaly exceeds the criticality threshold, the BMS 116 transmits control commands to disable charging or discharging subsystems and isolate the battery pack 102 from the load or power source. Subsequently, a fault indicator is triggered to alert the user or connected control unit, while maintaining the pack in a safe, monitored state. Consequently, the system enters a fail-safe mode until service or revalidation occurs. The anomaly detection provides active fault containment, prevention of unsafe operating conditions, and enhancement of battery pack protection. Specifically, real-time logging ensures complete traceability and enables forensic diagnostics after fault events. Further, operational restrictions limit the potential for cascading failures due to hidden or emerging faults. Furthermore, integrating the identity-based anomaly signal with control logic ensures rapid and deterministic fault response. Subsequently, the system transitions from passive monitoring to active protection upon detection of unsafe conditions. Consequently, the system 100 delivers increased safety, reliability, and compliance with functional safety standards in high-performance energy storage applications.
In accordance with a second aspect, there is described a method of detecting at least one anomaly in a battery pack, wherein the battery pack comprises a plurality of battery cells:
- injecting a predefined current pulse into each battery cell of the battery pack, via an excitation module;
- measuring, for each battery cell, a plurality of transient electrical characteristics, via a measurement module;
- generating a characteristic vector for each battery cell based on the plurality of transient electrical characteristics, via an arbitration module;
- applying a cryptographic hash function to a composite vector to generate the unique identity for the battery pack, via an arbitration module; and
- comparing a regenerated identity with the stored unique identity, via a battery management system.
Figure 2 describes a method of detecting at least one anomaly in a battery pack, wherein the battery pack comprises a plurality of battery cells. The method 200 starts at a step 202. At the step 202, the method comprises injecting a predefined current pulse into each battery cell 104 of the battery pack 102, via an excitation module 106. At a step 204, the method comprises measuring, for each battery cell 104, a plurality of transient electrical characteristics, via a measurement module 110. At a step 206, the method comprises generating a characteristic vector for each battery cell 104 based on the plurality of transient electrical characteristics, via an arbitration module 112. At a step 208, the method comprises applying a cryptographic hash function to a composite vector to generate the unique identity for the battery pack 102, via an arbitration module 112. At a step 210, the method comprises comparing a regenerated identity with the stored unique identity, via a battery management system 116.
In an embodiment, the method 200 comprises receiving the plurality of transient electrical characteristics via the arbitration module 112.
In an embodiment, the method 200 comprises applying a weighting function to each characteristic vector and computing a composite vector as a linear combination of the weighted characteristic vectors, via the arbitration module 112.
In an embodiment, the method 200 comprises storing the generated unique identity of the battery pack 102 and transmitting a copy of the generated unique identity to the battery management system 116, via an identity storage module 114.
In an embodiment, the method 200 comprises storing the copy of the unique identity of the battery pack 102 and initiating regeneration of the unique identity after a predefined number of charge cycles of the battery pack 102, via the battery management system 116.
In an embodiment, the method 200 comprises detecting an anomaly based on a mismatch between the regenerated identity and the stored unique identity, via the battery management system 116.
In an embodiment, the method 200 comprises logging the anomaly detection event in a memory and restricting charging or discharging operations of the battery pack 102 based on the detected anomaly, via the battery management system 116.
In an embodiment, the method 200 comprises injecting a predefined current pulse into each battery cell 104 of the battery pack 102, via an excitation module 106. Furthermore, the method 200 comprises measuring, for each battery cell 104, a plurality of transient electrical characteristics, via a measurement module 110. In an embodiment, the method 200 comprises generating a characteristic vector for each battery cell 104 based on the plurality of transient electrical characteristics, via an arbitration module 112. Furthermore, the method 200 comprises receiving the plurality of transient electrical characteristics via the arbitration module 112. Furthermore, the method 200 comprises applying a weighting function to each characteristic vector and computing a composite vector as a linear combination of the weighted characteristic vectors, via the arbitration module 112. Furthermore, the method 200 comprises applying a cryptographic hash function to a composite vector to generate the unique identity for the battery pack 102, via an arbitration module 112. Furthermore, the method 200 comprises storing the generated unique identity of the battery pack 102 and transmitting a copy of the generated unique identity to the battery management system 116, via an identity storage module 114. Furthermore, the method 200 comprises storing the copy of the unique identity of the battery pack 102 and initiating regeneration of the unique identity after a predefined number of charge cycles of the battery pack 102, via the battery management system 116. Furthermore, the method 200 comprises detecting an anomaly based on a mismatch between the regenerated identity and the stored unique identity, via the battery management system 116. Furthermore, the method 200 comprises logging the anomaly detection event in a memory and restricting charging or discharging operations of the battery pack 102 based on the detected anomaly, via the battery management system 116.
It would be appreciated that all the explanations and embodiments of the system 100 also apply mutatis-mutandis to the method 200.
In the description of the present invention, it is also to be noted that, unless otherwise explicitly specified or limited, the terms “disposed,” “mounted,” and “connected” are to be construed broadly, and may for example be fixedly connected, detachably connected, or integrally connected, either mechanically or electrically. They may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Modifications to embodiments and combinations of different embodiments of the present disclosure described in the foregoing are possible without departing from the scope of the present disclosure as defined by the accompanying claims. Expressions such as “including”, “comprising”, “incorporating”, “have”, and “is” used to describe and claim the present disclosure are intended to be construed in a non-exclusive manner, namely allowing for items, components or elements not explicitly described also to be present. Reference to the singular is also to be construed to relate to the plural where appropriate.
Although embodiments have been described with reference to a number of illustrative embodiments thereof, it should be understood that numerous other modifications and embodiments can be devised by those skilled in the art that will fall within the spirit and scope of the principles of this disclosure. More particularly, various variations and modifications are possible in the component parts and/or arrangements of the subject combination arrangement within the scope of the present disclosure, the drawings, and the appended claims. In addition to variations and modifications in the component parts and/or arrangements, alternative uses will also be apparent to those skilled in the art.
,CLAIMS:WE CLAIM:
1. A system (100) for detecting at least one anomaly in a battery pack (102), wherein the battery pack (102) comprises a plurality of battery cells (104), the system (100) comprising:
- an excitation module (106) configured to sequentially inject a predefined current pulse into each battery cell (104) of the battery pack (102);
- a processing unit (108) coupled to the excitation module (106), the processing unit (108) comprising:
- a measurement module (110) communicably coupled to the excitation module (106) and configured to measure, for each battery cell (104), a plurality of transient electrical characteristics; and
- an arbitration module (112) communicably coupled to the measurement module (110) and configured to receive the plurality of transient electrical characteristics;
- an identity storage module (114) operatively coupled to the processing unit (108); and
- a Battery Management System (BMS) (116) operatively coupled to the identity storage unit (114),
wherein the arbitration module (112) is configured to generate a unique identity of the battery pack (102) based on a deterministic function of the transient electrical characteristics, and the BMS (116) is configured to regenerate the unique identity after a predefined period to detect at least one anomaly based on a mismatch with the generated unique identity.

2. The system (100) as claimed in claim 1, the excitation module (106) is configured to inject the predefined current pulse with a voltage-controlled waveform, wherein the voltage-controlled waveform comprises a predefined amplitude and duration.

3. The system (100) as claimed in claim 1, wherein the plurality of transient electrical characteristics measured by the measurement module (110) comprises at least one of an electrical parameter, an electrochemical parameter, or a dynamic response parameter measured for each battery cell (10).

4. The system (100) as claimed in claim 2, wherein the arbitration module (112) is configured to generate a characteristic vector for each battery cell (104) based on the plurality of transient electrical characteristics.

5. The system (100) as claimed in claim 2, wherein the arbitration module (112) is configured to apply a weighting function to each characteristic vector and compute a composite vector as a linear combination of the weighted characteristic vectors.

6. The system (100) as claimed in claim 2, wherein the arbitration module (112) is configured to apply a cryptographic hash function to the composite vector to generate the unique identity for the battery pack (102).

7. The system (100) as claimed in claim 2, wherein the identity storage module (114) is configured to store the generated unique identity of the battery pack (102) and transmit a copy of the generated unique identity to the battery management system (116).

8. The system (100) as claimed in claim 2, wherein the battery management system (116) is configured to store the copy of the unique identity of the battery pack (102) and initiate regeneration of the unique identity after a predefined number of charge cycles of the battery pack (102).

9. The system (100) as claimed in claim 2, wherein the battery management system (116) is configured to compare the regenerated identity with the stored unique identity and detect an anomaly based on a mismatch between the regenerated identity and the stored unique identity.

10. The system (100) as claimed in claim 1, wherein the battery management system (116) is configured to log the anomaly detection event in a memory and restrict charging or discharging operations of the battery pack (102) based on the detected anomaly.

11. A method (200) of detecting at least one anomaly in a battery pack (102), wherein the battery pack (102) comprises a plurality of battery cells (104), the method (200) comprises:
- injecting a predefined current pulse into each battery cell (104) of the battery pack (102), via an excitation module (106);
- measuring, for each battery cell (104), a plurality of transient electrical characteristics, via a measurement module (110);
- generating a characteristic vector for each battery cell (104) based on the plurality of transient electrical characteristics, via an arbitration module (112);
- applying a cryptographic hash function to a composite vector to generate the unique identity for the battery pack (102), via an arbitration module (112); and
- comparing a regenerated identity with the stored unique identity, via a battery management system (116).

12. The method (200) as claimed in claim 10, wherein the method (200) comprises detecting an anomaly based on a mismatch between the regenerated identity and the stored unique identity, via the battery management system (116).

13. The method (200) as claimed in claim 10, wherein the method (200) comprises logging the anomaly detection event in a memory and restricting charging or discharging operations of the battery pack (102) based on the detected anomaly, via the battery management system (116).