Claims:1. A processor implemented method to detect change in vibrational data for failure forecasting, the method comprising:
receiving, via one or more hardware processor, a vibrational data comprising a plurality of vibrational signals as an operation response obtained from one or more assets being monitored by one or more vibration sensors;
extracting, a plurality of statistical features from the plurality of vibrational signals for the one or more assets being monitored via the one or more hardware processor;
detecting by edge, via the one or more hardware processor, one or more abnormal changes from the one or more assets observed in the plurality of statistical features extracted from the plurality of vibrational signals, wherein the edge comprises one or more computing devices; and
forecasting failure from the one or more abnormal changes detected for the one or more assets being monitored via the one or more hardware processors, based on (i) a rank of each statistical feature from the plurality of statistical features, and (ii) a magnitude of respective defect frequency peak, wherein the abnormal changes are windows based on a fast signal segmentation.
2. The processor implemented method as claimed in claim 1, wherein detecting the one or more abnormal changes from the plurality of statistical features by,
obtaining, the plurality of statistical features from the plurality of vibrational signals;
determining, the rank of each statistical feature from the plurality of statistical features by applying monotonicity to identify trend of each statistical feature based on (i) a difference between the number of a positive trend and a negative trend, and (ii) a number of measurement time points within a predefined value; and

detecting, the one or more abnormal changes from each statistical feature is based on the top ranked statistical features.
3. The processor implemented method as claimed in claim 1, wherein the step of forecasting failure for the one or more abnormal changes is based on a predetermined threshold.
4. The processor implemented method as claimed in claim 2, wherein the plurality of statistical features includes a mean, a variance, a kurtosis, a crest factor, an impulse factor, an energy, a standard deviation, a skewness, a root mean square of acceleration, a shape factor, a margin factor, and a peak.
5. A system (100), comprising:
a memory (102) storing instructions;
one or more communication interfaces (106); and
one or more hardware processors (104) coupled to the memory (102) via the one or more communication interfaces (106), wherein the one or more hardware processors (104) are configured by the instructions to:
receive, a vibrational data comprising a plurality of vibrational signals as an operation response obtained from one or more assets being monitored by one or more vibration sensors;
extract, a plurality of statistical features from the plurality of vibrational signals for the one or more assets being monitored via the one or more hardware processor;
detect by edge, via the one or more hardware processor, one or more abnormal changes from the one or more assets observed in the plurality of statistical features extracted from the plurality of vibrational signals , wherein the edge comprises one or more computing devices; and
forecast failure from the one or more abnormal changes detected for the one or more assets being monitored, based on (i) a rank of each statistical feature from the plurality of statistical features, and (ii) a magnitude of

respective defect frequency peak, wherein the abnormal changes are windows based on a fast signal segmentation.
6. The system as claimed in claim 5, wherein detecting the one or more abnormal changes from the plurality of statistical features by,
obtaining, the plurality of statistical features from the plurality of vibrational signals;
determining, the rank of each statistical feature from the plurality of statistical features by applying monotonicity to identify trend of each statistical feature based on (i) a difference between the number of a positive trend and a negative trend, and (ii) a number of measurement time points within a predefined value; and
detecting, the one or more abnormal changes from each statistical feature is based on the top ranked statistical features.
7. The system as claimed in claim 5, wherein the step of forecasting failure for the one or more abnormal changes is based on a predetermined threshold.
8. The system as claimed in claim 6, wherein the plurality of statistical features includes a mean, a variance, a kurtosis, a crest factor, an impulse factor, an energy, a standard deviation, a skewness, a root mean square of acceleration, a shape factor, a margin factor, and a peak , Description:FORM 2

THE PATENTS ACT, 1970 (39 of 1970)
&
THE PATENT RULES, 2003

COMPLETE SPECIFICATION
(See Section 10 and Rule 13)

Title of invention:
METHOD AND SYSTEM TO DETECT CHANGE IN VIBRATIONAL DATA AT EDGE FOR FAILURE FORECASTING

Applicant:
Tata Consultancy Services Limited
A company Incorporated in India under the Companies Act, 1956 Having address:
Nirmal Building, 9th Floor, Nariman Point, Mumbai 400021, Maharashtra, India

The following specification particularly describes the invention and the manner in which it is to be performed.

TECHNICAL FIELD
[01] The disclosure herein generally relates to change detection, and, more particularly, to method and system to detect change in vibrational data at edge for failure forecasting.
BACKGROUND
[02] Recent trends in manufacturing realm such as industry 4.0, smart factories, and industrial systems, vibrational signals have shown to be the most important measurements to indicate asset health. Based on these measurements, a user with expert knowledge on the assets, industrial process and vibration monitoring can perform spectral analysis to identify failure modes. However, this is still a manual process that heavily depends on the experience and knowledge of the user analyzing the vibration signals. Condition monitoring of such industrial assets ensures that problems can be early detected to avoid unplanned downtimes as well as extensive damages. Moreover, when measurements are performed continuously, it becomes impossible to act in real time on these vibration signals.
[03] In one of the conventional industrial systems, analysis of vibration signals helped in determining the critical failure of the asset which can be replaced on time with focused maintenance. Also, the cost of replacing entire assets is decreased to repair work and overall, the analysis brings increased safety, revenue and employee efficiency. In another existing system, machine learning based condition monitoring is used for various purposes for failure diagnostics where occurrence of such failure can be detected (possibly even identifying its specific type) without stopping and disassembling the asset being monitored. However, such system lacks prediction of failures before they occur based on subtle changes in the asset without estimating the Remaining Useful Life (RUL).
[04] In the realm of condition monitoring of industrial and manufacturing assets vibration monitoring is done in more than one way either continuously or discretely at predefined intervals of time. In continuous measurements, the system

predominantly stores the summary of the measurements as compared to the complete data as the overall volume of data gets out of control. Each time the summary is compared with a standard or a threshold and decisions are taken whether to continue or maintain the asset. Such system lacks in determining when the defect has initiated. This system can provide a broad observation after the defect has grown by some significant magnitude. The frequency of measurement is relatively low. In discrete measurements, such system predominantly makes a time schedule of measurements and accordingly make measurements manually with hand-held instruments most of the times. In such system skill of the person making the measurements enters the measurement process as a variance confounded with the variance of the instruments and it is not easy to separate and observe. These observations are then compared with standards and based on the experience of the certified personnel the decisions are made. Due to the inherent manual nature of the process it is bound to introduce conservatism into the decision making. In the above described continuous measurements and discrete measurements, the features calculated are similar, by taking a holistic view of the measurement processes, their frequency, the data and the decision is instantaneous, and their validity is retained till the next measurement is made, as the history of the measurement is not being used in the decision making processes. These systems lack the capability to provide a warning at the onset of defect, follow-up with a forecast of the time to failure and repeats the process for predefined interval of time.
SUMMARY
[05] Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one embodiment, a system to detect change in vibrational data at edge for failure forecasting is provided. The system includes receiving vibrational data comprising a plurality of vibrational signals as an operation response obtained from one or more assets being monitored by one or more vibration sensors. Further, a plurality of statistical features from the

plurality of vibrational signals are extracted for the one or more assets being monitored. The edge detects, one or more abnormal changes from the one or more assets observed in the plurality of statistical features extracted from the plurality of vibrational signals, wherein the edge comprises one or more computing devices. The failure is forecasted for the one or more abnormal changes detected for the one or more assets being monitored based on (i) a rank of each statistical feature from the plurality of statistical features, and (ii) a magnitude of respective defect frequency peak, wherein the abnormal changes are windows based on a fast signal segmentation. In one embodiment, the one or more abnormal changes are detected from the plurality of statistical features by obtaining the plurality of statistical features from the plurality of vibrational signals. Then, a rank of each statistical feature are determined from the plurality of statistical features by applying monotonicity to identify trend of each statistical feature based on (i) a difference between the number of a positive trend and a negative trend, and (ii) a number of measurement time points within a predefined value.
[06] In another aspect, a method to detect change in vibrational data at edge for failure forecasting is provided. The method includes receiving a vibrational data comprising a plurality of vibrational signals as an operation response obtained from one or more assets being monitored by one or more vibration sensors. Further, a plurality of statistical features from the plurality of vibrational signals are extracted for the one or more assets being monitored. The edge detects, a one or more abnormal changes from the one or more assets observed in the plurality of statistical features extracted from the plurality of vibrational signals, wherein the edge comprises one or more computing devices. The failure is forecasted for the one or more abnormal changes detected for the one or more assets based on (i) a rank of each statistical feature from the plurality of statistical features, and (ii) a magnitude of respective defect frequency peak, wherein the abnormal changes are windows based on a fast signal segmentation. Further, the one or more abnormal changes are detected from the

plurality of statistical features by obtaining the plurality of statistical features from the plurality of vibrational signals. Then, a rank of each statistical feature are determined from the plurality of statistical features by applying monotonicity to identify trend of each statistical feature based on (i) a difference between the number of a positive trend and a negative trend, and (ii) a number of measurement time points within a predefined value.
[07] In yet another aspect, a non-transitory computer readable medium provides one or more non-transitory machine-readable information storage mediums comprising one or more instructions, which when executed by one or more hardware processors perform actions includes an I/O interface and a memory coupled to the processor is capable of executing programmed instructions stored in the processor in the memory to receive a vibrational data comprising a plurality of vibrational signals as an operation response obtained from one or more assets being monitored by one or more vibration sensors. Further, a plurality of statistical features from the plurality of vibrational signals are extracted for the one or more assets being monitored. The edge detects, a one or more abnormal changes from the one or more assets observed in the plurality of statistical features extracted from the plurality of vibrational signals, wherein the edge comprises one or more computing devices. The failure is forecasted for the one or more abnormal changes detected for the one or more assets based on (i) a rank of each statistical feature from the plurality of statistical features, and (ii) a magnitude of respective defect frequency peak, wherein the abnormal changes are windows based on a fast signal segmentation. Further, the one or more abnormal changes are detected from the plurality of statistical features by obtaining the plurality of statistical features from the plurality of vibrational signals. Then, a rank of each statistical feature are determined from the plurality of statistical features by applying monotonicity to identify trend of each statistical feature based on (i) a difference between the number of a positive trend and a negative trend, and (ii) a number of measurement time points within a predefined value.

[08] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
BRIEF DESCRIPTION OF THE DRAWINGS
[09] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles:
[10] FIG. 1 illustrates an exemplary block diagram of a system (alternatively referred as an asset monitoring system), in accordance with some embodiments of the present disclosure.
[11] FIG. 2 illustrates a functional block diagram of the asset monitoring using the system of FIG.1, in accordance with some embodiments of the present disclosure.
[12] FIG. 3 is a flow diagram illustrating the method for analyzing vibration signals of the asset being monitored using the system of FIG.1, in accordance with some embodiments of the present disclosure.
[13] FIG.4 illustrates graphical representation of samples with monotonicity for analyzing the vibration signals of the asset being monitored using the system of FIG.1, in accordance with some embodiments of the present disclosure.
[14] FIG.5 illustrates graphical representation of various frequencies versus with change points observed from the vibration signals for the asset being monitored using the system of FIG.1, in accordance with some embodiments of the present disclosure.
DETAILED DESCRIPTION OF EMBODIMENTS
[15] Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or

like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the scope of the disclosed embodiments.
[16] Embodiments herein provide a method and system to detect change in vibrational data at edge for failure forecasting. The method disclosed, enables to forecast a time to failure or RUL accurately from abnormal changes observed from a plurality of vibrational signals obtained from one or more assets being monitored. The method of the present disclosure identifies defects of each asset from the plurality of vibrational signals to forecast failure time. Further, change in failure times are stored along with raw data that includes the plurality of vibrational signals forecasting the failure time or alternatively referred as a warning time, wherein this step is iteratively performed for a predefined interval of time. Further, for every iteration repeated wherein a one or more abnormal changes are identified from the plurality of vibration signals and thus provides fidelity for the forecasted failure time based on historical data. Also, the present disclosure provides decision making process based on the historical data which are prestored in the system to forecast real time failure to improve the performance of the asset. Also, the system and method of the present disclosure are time efficient, and scalable for failure forecast of the asset being monitored. The disclosed system is further explained with the method as described in conjunction with FIG.1 to FIG.5 below.
[17] Referring now to the drawings, and more particularly to FIG. 1 through FIG. 5, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method.
[18] FIG. 1 illustrates an exemplary block diagram of a system (alternatively referred as an asset monitoring system), in accordance with some embodiments of the present disclosure. In an embodiment, the asset monitoring system 100 includes processor (s) 104, communication interface (s), alternatively referred as or input/output

(I/O) interface(s) 106, and one or more data storage devices or memory 102 operatively coupled to the processor (s) 104. The system 100, with the processor(s) is configured to execute functions of one or more functional blocks of the system 100. Referring to the components of the system 100, in an embodiment, the processor (s) 104 can be one or more hardware processors 104. In an embodiment, the one or more hardware processors 104 can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor(s) 104 is configured to fetch and execute computer-readable instructions stored in the memory. In an embodiment, the system 100 can be implemented in a variety of computing systems, such as laptop computers, notebooks, 10 hand-held devices, workstations, mainframe computers, servers, a network cloud, and the like.
[19] The I/O interface(s) 106 can include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface, and the like and can facilitate multiple communications within a wide variety of networks N/W and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular, or satellite. In an embodiment, the I/O interface (s) 106 can include one or more ports for connecting a number of devices (nodes) of the system 100 to one another or to another server. The memory 102 may include any computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes. The modules 108 can be an Integrated Circuit (IC) (not shown), external to the memory 102, implemented using a Field-Programmable Gate Array (FPGA) or an Application- Specific Integrated Circuit (ASIC). The names (or expressions or terms) of the modules

of functional block within the modules 108 referred herein, are used for explanation and are not construed to be limitation(s).
[20] FIG. 2 illustrates a functional block diagram of the asset monitoring using the system of FIG.1, in accordance with some embodiments of the present disclosure. The FIG.2 includes an edge, the one or more assets to be monitored, and a plurality of vibration sensors. The system receives the plurality of vibrational signals from the one or more assets for every predefined interval of time, wherein each vibration sensor is configured to each asset to capture the plurality vibrational signals for analysis. The edge includes the one or more cloud computing devices to identify failures based on the one or more abnormal changes observed from the plurality of vibrational signals.
[21] FIG. 3 is a flow diagram illustrating a method for analyzing vibration signals of the asset being monitored using the system of FIG.1, in accordance with some embodiments of the present disclosure. In an embodiment, the system 100 comprises one or more data storage devices or the memory 102 operatively coupled to the processor(s) 104 and is configured to store instructions for execution of steps of the method 300 by the processor(s) or one or more hardware processors 104. The steps of the method 300 of the present disclosure will now be explained with reference to the components or blocks of the system 100 as depicted in FIG.1 and the steps of flow diagram as depicted in FIG.3. Although process steps, method steps, techniques or the like may be described in a sequential order, such processes, methods and techniques may be configured to work in alternate orders. In other words, any sequence or order of steps that may be described does not necessarily indicate a requirement that the steps to be performed in that order. The steps of processes described herein may be performed in any order practical. Further, some steps may be performed simultaneously.
[22] Referring now to the steps of the method 300, at step 302, the one or more hardware processors 104 receive vibrational data comprising a plurality of

vibrational signals as an operation response obtained from one or more assets being monitored by one or more vibration sensors. Here, each asset is configured with the vibration sensor to capture the plurality of vibrational signals for further analysis which are obtained at predefined intervals by the system 100. The edge provides the one or more abnormal changes to remaining useful life or time to failure by one or more cloud computing devices connected to any network. The system 100 can be utilized in any IoT manufacturing for industrial health monitoring for vibrational analytics of each asset being monitored. Considering an example, where the one or more vibration sensors captures the plurality of vibration signals from the one or more assets. Further, each vibrational signal is analyzed to detect the one or more abnormal changes from the one or more assets.
[23] Referring now to the steps of the method 300, at step 304, the one or more hardware processors 104 extract a plurality of statistical features from the plurality of vibrational signals for the one or more assets being monitored. From the above example, the plurality of statistical features are extracted from the plurality of vibrational signals. The plurality of statistical features includes, but are not limited, a mean, a variance, a kurtosis, a crest factor, an impulse factor, an energy, a standard deviation, a skewness, a root mean square of acceleration, a shape factor, a margin factor, and a peak. The sample of each vibration signal is measured at high frequency for a short duration for example at 20.48 kHz frequency for a duration of 1 second which leads to 20480 rows of data. These 20480 rows of data are summarized to one row with a time stamp. The measured vibrational parameter is acceleration to identify a critical indicator trend of degradation that includes a positive trend and a negative trend.
[24] Referring now to the steps of the method 300, at step 306, the one or more hardware processors 104 detect via edge, a one or more abnormal changes from the one or more assets observed in the plurality of statistical features extracted from the plurality of vibrational signals, wherein the edge comprises one or more cloud

computing devices. Further, the plurality of statistical features are obtained from the plurality of vibrational signals to determine rank of each statistical feature from the plurality of statistical features by applying monotonicity to identify trend based on (i) a difference between the number of a positive trend and a negative trend, and (ii) a number of measurement time points within a predefined value. This trend triggers to detect the one or more abnormal changes from the top ranked statistical features. The top ranked statistical features are identified for further processing, and this selection can be surpassed. Upon detection, the respective raw data and the time index will be moved to separate storing and archival system for further processing. The method of identifying the one or more abnormal changes artefacts and stored for processing which reduces the data volume to <15% of the volume of the data if such a system was non- existent. The system is implemented in python with the ruptures library. The remaining data is purged, and all the artefacts are retained through the detection of one or more abnormal changes and are then stored. Here, degradation process is a continuous irreversible process which is always an increasing function. It becomes imperative to rank and identify each statistical feature that best emulates the degradation process for the one or more assets being monitored. This ranking is performed using monotonicity which quantifies the positive trend of each statistical feature. Monotonicity is the difference between the number of positive trend and the number of negative trend in the path as described below in equation 1,

???????????????????????? = ???????? (|

????????????????(????????(????))-(????????????????(????????(????)))
??-1
equation 1,

|) ------

where, i = 1,2,3,4……n, and n is the number of measurement time point

[25] Referring now to the steps of the method 300, at step 308, the one or more hardware processors 104 forecast failure with the one or more abnormal changes from the one or more assets being monitored, based on (i) a rank of each statistical feature from the plurality of statistical features, and (ii) a magnitude of respective

defect frequency peak, wherein the one or more abnormal changes are windows based on a fast signal segmentation. Now referring to the above example for identification of the time index of the one or more abnormal changes, failure identification is performed by frequency domain analysis and forecasting of the time to failure. Alternatively the warning time is estimated using the timeseries forecasting. Here, the failure identification involves performing a Fast Fourier transform (FFT) of the raw timeseries data and identifying the characteristic frequencies and the respective amplitude peaks. The values are arrived through engineering calculations and physics. In addition to Fast Fourier transform (FFT) several systems of frequency domain analyses are done to isolate and to identify the characteristic peaks including signal filtering. Further, the onset of failure and the abnormal change detection, the failure forecasting estimates the time to failure taking the change points and the summarized data feature coming from the application of monotonicity.
[26] Further, the method detects the one or more abnormal changes for the window based fast signal segmentation. The steps performed in the system uses two adjacent windows that slide along the data, and at each slide the plurality of statistical features properties in each window are compared and the one or more abnormal change are identified based on a minimum predetermined threshold. The predetermined threshold is set as 30. The one or more abnormal changes detected will start after accumulation of the predetermined threshold and within predetermined threshold. If the statistical properties are similar/same no change identified, while the abnormal change is identified when there is observation of change. The method further provides time instant of the change. Following this identification, the respective raw data is moved into a cloud or any computing device for further processing involving failure detection and forecasting the time to failure. The edge performs the volume of the plurality of vibrational signals which are significantly reduced at the same time and all the one or more abnormal changes or deviation of normality information is preserved.

[27] In an embodiment, for the experimental analysis the raw data has reduced to ~15% of the total volume of raw data when the referred system was non- existent. These change points along with the summary that are being calculated for further processing. In another offline analyses, the one or more abnormal changes are used to identify raw data which is subjected to frequency domain analyses and is used to estimate the frequencies that include bearing defect frequencies due to an inner race, an outer race, a rolling element and a cage. Similarly, for gear boxes it is gear meshing frequencies which are characteristic in their appearance and can be easily identified when the timeseries data is subjected to frequency domain analyses. The plurality of statistical features are identified for the one or more abnormal change and the timeseries summary of this feature is used for forecasting the time to failure – which is mostly the predetermined threshold and if reached it is deemed to have failed. Additionally, the present disclosure can be extended through a series of mathematical transformations of the data and forecasting. The mathematical transformations include transformation of the acceleration to velocity and computation of the root mean square (RMS) of the velocity which is a common feature in the referred standard [ISO 10816
– 3]. The forecasting is done on the root mean square (RMS) of the velocity feature which makes the comparison of the state of the system relatively simpler from forecasting the transformation of the state. The system is implemented through a Generalized Additive Model (GAM) which has trend, seasonality and an error as its terms. The seasonality is modelled through Fourier terms. The system has been implemented using FBProphet [Taylor et. al.,] library in python. Though this library as indicated by the developers is best for marketing and product launches it has been implemented for an engineering problem for the first time.
[28] FIG.4 illustrates graphical representation of samples with monotonicity for analyzing the vibration signals of the asset being monitored using the system of FIG.1, in accordance with some embodiments of the present disclosure. FIG.4 depicts the equation 1, and for each sample of each vibrational signal kurtosis is considered as

the top ranked feature to change detection. On the selected feature change detection is performed. The above figure depicts the kurtosis and the root mean square (RMS) which are the most sought features at times due to data monotonicity indicates some other feature, at these situations a vibration analysis expert can intervene and make the choice of the feature.
[29] FIG.5 illustrates graphical representation of various frequencies versus with change points observed from the vibration signals for the asset being monitored using the system of FIG.1, in accordance with some embodiments of the present disclosure. FIG.5 indicates defect frequencies obtained following a frequency domain analysis of the raw data coming from the change detection system. The frequency domain analyses include the Fast Fourier Transform (FFT), the power spectral density (PSD), Short Time Fourier Transform (STFT), physics-based calculations and sometimes signal filtering algorithms. The system is implemented using python with the scipy.signal library. Python is the programming tool which has been used for the development of this change detection system, in which the statistical features and monotonicity has been implemented.
[30] The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope of the subject matter embodiments is defined by the claims and may include other modifications that occur to those skilled in the art. Such other modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.
[31] The embodiments of present disclosure herein address unresolved problem of change detection. The embodiment thus provides method and system to detect change in vibrational data at edge for failure forecasting. Moreover, the embodiments herein further provide a system wherein change in failure time are stored along with raw data that includes the plurality of vibrational signals forecasting the

failure time or alternatively referred as a warning time. In the realm of vibration analytics owing to data storage challenges only summary is stored historically (or in the conventional approaches/techniques). There existed no technological intelligence to decide what data needs to be stored and what needs to be discarded. The present disclosure addresses this adequately and it brings in the IOT system with EDGE/CLOUD configuration, and the components residing on the EDGE enables optimization of the data storage wherein the data stored is relevant to the phenomena. This process provides an opportunity to perform a root cause analysis in-retrospect. Additionally, the system also performs failure identification and forecasting which is not the routine output of specialists in vibration analytics. Complimenting the change detection, as industrially once a defect is identified the primary interest is to estimate the warning time.
[32] It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device which can be programmed including e.g., any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g., hardware means like e.g., an application- specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g., an ASIC and an FPGA, or at least one microprocessor and at least one memory with software processing components located therein. Thus, the means can include both hardware means, and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g., using a plurality of CPUs.

[33] The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various components described herein may be implemented in other components or combinations of other components. For the purposes of this description, a computer- usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
[34] The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
[35] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable

storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
[36] It is intended that the disclosure and examples be considered as exemplary only, with a true scope of disclosed embodiments being indicated by the following claims.