The present invention relates to a battery equalizer. More particularly, the present invention is directed to the artificial intelligence based battery equalizer system.
PRIOR ART:
[002] Portable power equipment and electric vehicles require large-capacity energy storage batteries. At present, most of them use lead-acid battery packs. Although lead-acid batteries are relatively cheap, they have low energy efficiency to weight ratio, large volume, fewer cycles of charge and discharge, and self-discharge rate. High, the electrode plate is easy to vulcanize when it is not charged for a long time, the capacity decays greatly, and a small amount of hydrogen and sulfuric acid will overflow during the charging process, and the overflow volume is large during overcharge, and the safety performance is not high.
[003] Lithium iron phosphate battery has high single-cell voltage (the voltage in the concentrated area of stored power is 3.2V), high cycle charge and discharge times, low self-discharge rate, not afraid of over-discharge, light weight, small size, large energy storage capacity, and high energy efficiency to weight ratio. It has high safety performance and will not explode even if the two poles are short-circuited. The price is relatively moderate. It is an ideal large-capacity energy storage battery for portable power equipment and electric vehicles.
[004] However, lithium iron phosphate batteries have two major physical weaknesses:
1. Single-charge batteries are afraid of overcharging, and the rechargeable battery will cause permanent damage to the battery if it exceeds 3.7V;
2. The electrical parameters of the battery packs with the same capacity and the same production process are highly discrete and difficult to match. The overall external performance of the battery pack is not large, which limits its wide use.
[005] Reference may be made to the following prior arts:
[006] Publication No. CN207705834 relates to a lithium iron phosphate group battery heavy current intelligence equalizing charge management system, its characterized in that: including single chip logic controller, power adapter, constant current charging module and balanced current device, power adapter is connected with constant current charging module and single chip logic controller electricity, utilize it to charge to the lithium iron phosphate group battery, change heavy current equalizing charge promptly into as long as a lesson reaches to charge behind the 3.7V in the group battery, be full of back other how many at a lesson and economize on electricity the pond and can be full of the electric energy successively in the short time, solved the two major physics nature weakness problems of lithium iron phosphate group battery well for the popularized use is merged into to lithium iron phosphate group battery energy efficiency ratio maximize, full play battery advantage. Whole charging process need not artificial intervention, realizes intelligent charging.
[007] Publication No. WO2021086041 relates to a method of predicting a remaining battery life in a device. The method includes obtaining battery state information indicative of a current state of a charge of a battery, the battery being configured to supply power to a device; predicting, by using a machine learning algorithm, a remaining battery life of the device on which a specific application is to be executed, based on the obtained battery state information; and providing, to a user, an indication of the predicted remaining battery life.
[008] Publication No. US2021057920 relates to uses of artificial intelligence in battery technology including a method that includes receiving a trained model, receiving sensor data from at least one sensor associated with a battery, and executing the trained model by a processor. Executing the trained model includes providing the sensor data as input to the trained model to generate a model output. The method also includes sending, from the processor to a charge controller coupled to the battery, a control signal that is based on the model output and automatically, by the charge controller, initiating or terminating charging of the battery based on the control signal.
[009] Publication No. CN110346734 relates to a lithium ion power battery state-of-health estimation method based on machine learning. The method is used for estimating the state-of-charge (SOC) and the state-of-health of a power battery in real time. Parameter identification is performed on an equivalent circuit model by establishing the equivalent circuit model of the lithium ion battery, and then a Uoc-SOC model is established, and the SOC is estimated. Training is carried out by using a large amount of offline data to obtain a neural network model which takes Uoc-SOC model parameters as input and takes the maximum available capacity as output. The method further has the following steps: performing curve fitting on Uoc and the SOC at the same moment to obtain to-be-identified parameters in the model, inputting the to-be-identified parameters into the neural network model obtained by training to obtain the maximum available capacity, returning the obtained Uoc-SOC model parameters and the maximum available capacity to the SOC estimation step, and updating the parameters of a state equation and an observation equation. According to the lithium ion battery state-of-health estimation method provided by the invention, online estimation is carried out for the battery health state, parameter updating is carried out on SOC estimation, and the estimation precision is improved.
[010] Publication No. CN112379273 relates to a lithium ion battery charging curve reconstruction method based on artificial intelligence, so that estimation of multiple states of a battery can be achieved. According to the method, charging segment data is used as input, a complete charging curve is reconstructed by using a deep learning method, and then multiple states of the battery, including the maximum capacity, the maximum energy, the state of charge, the energy state, the power state, the capacity increment curve and the like of the battery, can be extracted from the complete charging curve. The battery state estimation method provided by the invention can be self-adaptively updated along with the change of the working state of the battery.
[011] Publication No. AU2021100545 relates to a method for estimating the state of health of lithium ion power batteries based on machine learning, which is used for estimating the state of charge and the state of health of the power battery in real time. The equivalent circuit model of lithium ion battery is established, and its parameters are identified. Establishing Uoc-SOC model and estimating SOC. The neural network model with Uoc-SOC model parameters as input and maximum available capacity as output is trained by using a large amount of offline data. The parameters to be identified in the model are obtained by curve fitting of Uoc and SOC at the same time, which are input into the trained neural network model to obtain the maximum available capacity. The Uoc-SOC model parameters and the maximum available capacity are returned to the SOC estimation step to update the parameters of its state equation and observation equation. The invention provides a method for estimating the state of health of lithium ion batteries, which estimates the the state of health of the battery on line, updates the parameters of SOC estimation, and improves the estimation accuracy. Establishing equivalent circuit model Parameter identification RD, Rps Cp Estimation of SOC based on RLS based on EKF Fitting Uoc-SOC curve ak bici a', hi' Training offline BP neural network data by BP neural model network Output C Fig. 2 A flowchart of the method for estimating the state of health of the lithium ion power battery according to the present invention.
[012] Publication No. US2021123975 relates to systems and methods for forecasting of State of Charge (SOC) of lithium ion batteries. A multi-step forecasting process with experimentally obtained decreasing C-Rate datasets together with machine learning can be used. The multi-step approach can combine a univariate technique with machine learning techniques. An Auto Regressive Integrated Moving Average (ARIMA) and/or Holt Winters Exponential Smoothing (HWES) can be combined with each other and/or with machine learning techniques such as Multilayer Perceptron (MLP) and Nonlinear autoregressive neural network with external input (NARX-net).
[013] Publication No. US2021005027 relates to a system and method for predicting a battery charge depletion of a vehicle. The system further configured to classify the vehicles according to the extracted data and generate a battery maintenance schedule for each vehicle. The system comprises an analytical data module, a remote computing system, and a weather forecast module. The analytical data module is configured to extract data, for example, battery status, from the vehicle. The weather forecast module is configured to detect weather forecasts for an area in which the vehicle is located. The remote computing system comprises a battery depletion prediction module, a machine learning module, and a database. The battery depletion prediction module is configured to predict the battery charge depletion based on the extracted data and weather forecasts using the machine learning algorithm. The remote computing system is connected to a user device to transfer the predicted battery status of the vehicle.
[014] Publication No. CN112467851 relates to a lithium iron phosphate battery pack equalization control method, aims to solve the problems of abnormal attenuation of battery pack capacity, limitation of battery pack discharge capacity and the like caused by aging and inconsistency of single batteries in the use process of a lithium iron phosphate battery pack, and belongs to a battery management system. The control method comprises an aging monomer identification module and an equalization control module; because the charging and discharging cut-off voltage is firstly reached in the charging and discharging process of the aging monomer to influence the charging and discharging capacity of the battery pack, the aging degree of the battery monomer is identified by using an extreme learning machine, and the inconsistency is judged by taking the state of charge (SOC) of the battery as an equalization variable; equalization control is carried out based on a fuzzy logic control algorithm, thereby improving capacity attenuation of the battery pack and optimizing use performance of the battery pack.
[015] Publication No. CN112083333 relates to lithium ion batteries, and provides a power battery pack state-of-charge estimation method based on a machine learning model. The method comprises the following steps: training a battery model through employing a long-term and short-term memory neural network algorithm according to the temperature, current and state-of-charge, obtained through testing, of a power single battery as input and terminal voltage as output; adopting a square root volume Kalman filtering algorithm to calculate the charge states of the single battery with the maximum voltage and the single battery with the minimum voltage in the power battery pack in real time; retraining the battery model by using a rolling learning method; and calculating the state of charge of the battery pack by using a weight method according to the obtained states of charge of the single battery with the maximum voltage and the single battery with the minimum voltage in the power battery pack. According to the invention, the state of charge of the power battery pack can be accurately estimated after the ambient temperature changes and the battery is aged, the efficiency and accuracy of estimating the state of charge of the power battery pack are improved, and the anti-interference capability is strong.
[016] Publication No. CN112083334 relates to a lithium ion battery state-of-charge estimation method based on data driving, and relates to the technical field of battery management systems. A state-of-charge estimation model based on an extreme learning machine is trained by utilizing the acquired battery voltage, current and temperature and the calculated state-of-health value and state-of-charge value. Then the state-of-charge value estimated based on the extreme learning machine is used as an observed value of Kalman filtering, Kalman filtering is used for filtering the observed value, and interference of measurement noise is filtered out, so that the estimation precision of the state of charge is improved. The method belongs to a data-driven method, integrates the advantages of the extreme learning machine and Kalman filtering, does not need to establish an accurate electrochemical model, can accurately estimate the state of charge only by learning historical data of a battery, and has very high generalization performance and applicability.
[017] Publication No. AU2020102231 relates to solar photovoltaic (PV) power is in demand, it is advisable to maximize and track the power point. As there is a deviation in receiving the irradiation due to the variation in the atmospheric condition, the four-sun technology is implemented to eradicate the same. Also, to maximize the power point, a (Maximum Power Point Tracker) MPPT system was introduced. This will accept the irradiation and temperature as the input and improve the voltage using the DC-DC (Direct Current) converter. Then it is subjected to the machine learning technique. In (Support Vector Machine) SVM the actual reference (Vref) and the predicted reference is compared, and it generates an output which is further fed into (Proportional Integral Derivative) PID for maintaining stability in the voltage. Later it is fed through the micro controller which will invert the DC to AC (Alternative Current) voltage and produces a text using the Arduino micro controller within a range of 0 to 1023. Battery is also used as an alternative measure for making the inversion. Thus, the maximum power is tracked and the PV charge is controlled using a machine learning technique yielding high efficiency of voltage.
[018] Publication No. CN111563576 relates to a lithium battery capacity estimation method based on a bat detection-extreme learning machine, and belongs to the technical field of batteries. The method comprises the following steps: S1, performing a charge-discharge cycle condition test on a lithium battery, recording test data, and determining input variable s output variables through sensitivity analysis; S2, forming a training set and a test set by the input variables and the output variables; designing a bat detection algorithm and importing the training set for iterative optimization to obtain an optimal output weight; S3, calculating an input connection weight and a hidden layer neuron threshold, and constructing a feed forward neural network structure extreme learning machine; and S4, importing the test set into the extreme learning machine constructed in the step S3 to perform lithium battery capacity estimation, and performing performance evaluation of lithium battery capacity estimation. The method is good in generalization capability, can effectively reduce the lithium battery capacity estimation error, and improves the lithium battery capacity estimation precision.
[019] Publication No. CN111693868 relates to a lithium battery state-of-charge estimation method based on density feature clustering integration. Aiming at the problems that the precision of a lithium ion battery state of charge (SOC) intelligent estimation model is difficult to improve and the stability of a traditional estimation method is poor, an integrated modeling method based on data feature clustering is designed to estimate the SOC by combining the characteristics of actual operation data of a battery. The method is characterized by comprising the following steps: firstly, designing a data selection strategy for clustering modeling data with a wide dispersion range according to characteristics to obtain a sub-learning machine training data set by combining the characteristics of large SOC data fluctuation range and frequent inter-state conversion, and improving the performance of a sub-learning machine by reducing the distribution range of the sub-learning machine training data; secondly, during ensemble learning, because the training data of each sub-learning machine has similar characteristics, updating the weight of the data by utilizing the correlation between the data and the cluster, so that the sub-learning machines have relatively strong pertinence during training; afterwards, integrating a plurality of sub-learning machines, and further improving the estimation precision of the model.
[020] Publication No. US2018166878 relates to a computer-implemented method, system, and computer program product for demand charge management. The method includes receiving an active power demand for a facility, a current load demand charge threshold (DCT) profile for the facility, and a plurality of previously observed load DCT profiles. The method also includes generating a forecast model from a data set of DCT values based on the current load DCT profile for the facility and the plurality of previously observed load DCT profiles. The method additionally includes forecasting a monthly DCT value for the facility using the forecast model. The method further includes preventing actual power used from a utility from exceeding the next month DCT value by discharging a battery storage system into a behind the meter power infrastructure for the facility.
[021] Publication No. CN207938239 relates to a real standard platform of BMS battery management system belongs to new energy automobile teaching field. Real standard platform of BMS battery management system is by the lithium iron phosphate battery, simulated battery, battery management system, the mechanical maintenance switch, the on -vehicle machine that charges, the relay charges, charging relay, the preliminary filling relay, preliminary filling resistance, load simulating motor, the intelligent trouble sets up the system, the alternating -current charging mouth, the signal detection station, other auxiliary assembly such as MINIPC machines constituted, simulated battery can change the voltage parameter of group battery under intelligent trouble setting device's operation, carry out voltage signal detection and change group battery output parameter through the change of parameter through battery management system, accessible signal detection station detects each battery signal of group battery simultaneously, thereby can solve true lithium cell cannot be in the short time and carries out the difficulty that the charge -discharge changes the parameter for a long time, make learning process more oversimplify.
[022] Publication No. CN107843843 relates to an online vehicle-mounted battery SOC (state of charge) prediction method based on big data and an extreme learning machine. The external characteristic parameters of the battery such as voltage, current, temperature and internal resistance are selected, the large amount of online acquired external characteristic parameters of the battery are integrated through a big data method, a big data system for SOC prediction is formed, the data thus can be effectively excavated later, and the prediction precision is ensured; and through the extreme learning machine method, effective data most closely related to SOC prediction are found out, and the SOC is further accurately predicted according to the excavated effective data. The method has the advantages of high prediction precision and strong practicability and the like.
[023] Publication No. JP2007240308 relates to a battery and to accurately control the battery based on this. This control device comprises a learning means for learning the state of the battery with an NN circuit, a charge/discharge amount determining means for receiving the input of a target value of the battery state amount at a predetermined time and determining the charge/discharge amount of the battery until reaching the predetermined time, and a controlling means for controlling the battery state amount to the target value based on the determined charge/discharge amount of the battery. The state of the battery of an actual machine can be learned by the NN circuit, so that the charge/discharge amount of the battery can be optimally controlled to the target value of the battery state amount.
[024] US patent no 5952815 relates to an apparatus and a method for regulating the charge voltage of a number of electrochemical cells connected in series. Equalization circuitry is provided to control the amount of charge current supplied to individual electrochemical cells included within the series string of electrochemical cells without interrupting the flow of charge current through the series string. This invention does not include the fuse protection and fuse blown indication and protection in case the batteries connected are bad.
[025] US patent no 6,031,354 discloses the on-line battery management and monitoring system and method for monitoring a plurality of battery cells identifies. The invention deals with online management, monitoring and controlling of plurality of battery cells. This invention does not include any cell or battery equalizer feature.
[026] US publication no 20020074985 pertains to a voltage equalizer circuit in which each of plural windings P1 to Pn are electromagnetically coupled to each other, each of plural storage elements E1 to En series are connected to each other, and each of plural first switching elements S1 to Sn are connected to each other in a series connecting manner so as to constitute a plurality of closed circuits. This invention does not provide any protection against any wrong wiring and also does not provide any indication of failure.
[027] US patent no 5,982,142 is directed to a three terminal battery equalizer which includes a DC--DC converter having a first filter inductor in the switched, current conducting path connected to the battery ground and a second filter inductor in the switched, current path connected to the non-grounded. This invention does not provide any protection against any wrong wiring and also does not provide any indication of failure.
[028] US patent no 5,594,320 relates to an equalizer for equalizing the charge on several series-connected cells includes a transformer having plural windings on a core corresponding to the number of cells. The windings are tightly coupled to one another. This invention does not include any protection against wrong connection or faulty battery it also does not provide any indication to show if the system is working properly or not.
[029] US patent no 6452363 is directed to the charge equalizer for a string of series-connected batteries. The equalizer includes a shunt path for each of the batteries. The equalizer measures the voltage of each of the batteries of the string and then closes a switch in the shunt path associated with the highest voltage battery for a predetermined time. Then the switches of all of the shunt paths are opened and the cycle is repeated.
[030] US patent no 6452363 and 6,150,795 discloses charge equalizer for a string of series connected batteries in automotive electrical system.
[031] US patent no 5,710,504 relates to a switched capacitor system for automatic battery equalization that can be used with series coupled batteries as well as primary and backup batteries which are alternately couplable to a load. This does not include any protection against a wrong connection or faulty battery. It also does not provide any indication if the system is working properly or not.
[032] US publication no 20080272739 relates to the battery monitoring device of a battery pack. The device, configured for powering a cordless power tool may include an integrated circuit connected to a microprocessor of the pack that is external to the integrated circuit and is connected to each of the N battery cells of the pack.
[033] US publication 20080272739 relates only to the battery monitoring device and does not contain battery equalization.
[034] US publication no 20070171965 relates to an adaptive equalizer which provides receiving symbols to generate an equalizer output. The adaptive equalizer comprises a plurality of tap cells, a coefficient updater, a plurality of multiplexers, a controller and an integrator. Each tap cell generates a filter value from a tap data value and a coefficient.
[035] Publication no WO9310589 relates to an apparatus for balancing the state of charge of a plurality of serially connected sub-units of a battery. The apparatus comprises state of charge monitoring means operative to monitor the state of charge reached by each sub-unit.
[036] The following patents talk about battery equalization, but do not provide protection in case of fault condition.
[037] US patent no 6,452,363 relates to a charge equalizer for a string of series-connected batteries. The charge equalizer includes a shunt path for each of the batteries, measures the voltage of each of the batteries of the string and then closes a switch in the shunt path associated with the highest voltage battery for a predetermined time. This patent does not provide protection in case of fault condition.

[038] US patent no 5,528,122 relates to the battery equalizer for equalizing the voltage on series connected batteries. It synchronously switches the opposite ends of a tapped autotransformer in alternately reversing connection to the distally opposite terminals of the battery by alternately turning transistor switches on and off in alternate pairs while maintaining a center tap of the autotransformer connected to an intermediate terminal of the series batteries.
[039] US publication no 20050140335 relates to the terminal voltage equalization circuit to equalize the terminal voltage of the series of connected battery strings so that each battery in the series of connected battery strings can be equally charged.
[040] US publication no 20070103121 relates to a method and system for battery protection. In some aspects, a battery pack configured to be interfaced with a power tool includes housing a cell, a controller and a circuit. The circuit is operable to enable the controller to operate when the voltage supplied by the cell to the controller is below an operating voltage threshold off the controller.
[041] Publication no EP0652620 relates to the method and device of equalizing the voltage across drive batteries for electric vehicles, connected in series during recharging.
[042] US patent no 4,331,911 relates to equalizer including DC to DC converter which increase the losses in the system and decreases the overall efficiency of the system. Also it does not provide protection in case of fault condition.
[043] US patent no 6,008,623 relates to a charge equalizer by using flyback converter. Thus it will have higher losses.
[044] US patent no 5,666,041 and 5,982,143 relate to the electronic battery equalization circuit that equalizes the voltages of a plurality of series connected batteries in a battery pack. The current waveform is in the shape of a ramp for providing zero current switching.
[045] US patent no 6,801,014 relates to the method and system for equalizing the voltage of batteries in a battery string to a desired voltage. An equal string current is drawn from the batteries of the battery string and redistributed as a plurality of secondary currents to each battery, depending upon the comparative voltage of the individual batteries.
[046] Publication no WO2007038898 provides an accumulator battery equalizer and an energy balance accumulator battery. The equalizer is equipped between the plates of individual battery cell, at least two of which are connected in series so as to compose an accumulator battery set.
[047] Application no 3389/DELNP/2007 relates to an implementation of a method for cell equalization. The present invention uses other information such as the individual cell state-of-charge (SOC) estimates and individual capacities and/or cell coulombic efficiencies, possibly available from a dual extended Kalman filter.
[048] Reference may be made to an article by “Vanner Incorporated, July 15, 2002”. The article explains an efficient and highly reliable method of obtaining a 12-volt DC power source from a 24-volt DC electrical system. The equalizer makes the batteries look like they are in series and parallel at the same time. In addition to providing regulated 12-volt power, the system ensures that battery voltages remain equal. This significantly extends battery life. This module monitors the battery system's voltage and balance and provides fault signals that can be wired to warning lights, buzzers or other control/warning devices.
[049] Reference may be made to an article by “GSL Electronics, 2009”. The article explains 12V/ 24V battery equalizer. The unique feature of the equalizer is that they are 12V TRUE battery equalizers designed for centre tapped 12V dual battery systems. This is due to the micro controller measuring both sides of the system and charging the 12V tapped battery so that it is never out of balance.
[050] Reference may be made to an article by “Ming Tang; Stuart, T., IEEE Transactions on Aerospace and Electronic Systems, Vol 36, Issue 1, P(s):201–211, Jan 2000”. The article explains a new selective equalizer developed from the earlier ramp equalizer. A set of bipolar junction transistors (BJTs) controlled by a microcontroller is used to route equalization current to the lowest voltage batteries. Since only the lowest voltage batteries are connected to the equalizer, the need for uniform transformer leakage inductance is avoided and a lower power level can be used since no excess current flows to the other batteries.
[051] Reference may be made to an article by “A. Handed, T.A. Stuart”. The article explains a special round robin (RR) algorithm has been developed to equalize nickel metal hydride (NiMH) battery packs using a new selective equalizer. This algorithm detects batteries either at a very low state of charge (SOC) or at an extremely high SOC. In this system, a set of electromechanical relays are connected in a matrix to route boost current to the weaker batteries. The relay switching is controlled by a 32-bit microcontroller, and the boost current is supplied by a boost charger. Once the RR algorithm detects a weak battery, it schedules the detected battery for a specific boost time.
[052] Reference may be made to an article by “Analytic Systems Ware Ltd, 2002”. The article explains a battery equalizer that safely permits 12 Volts to be drawn of a 24V battery bank. The unit is connected across the 24V battery bank, and its output is connected to the 12V midpoint of the battery bank.
[053] Reference may be made to an article by “Hot Juice Electric, LLC, 2007-08”. The article explains BEQ1, a rugged, simple to use, economical shunt regulator that is designed to keep a series string of AGM batteries equalized by enhancing the recombinant charging phase. The red over temp LED will illuminate if the BEQ1 temperature exceeds 100° C. Once tripped, the over temp circuit reduces the shunt current and is latched on until the voltage drops below the BEQ1’s set point. This means the charger should be turned down or turned off until the cause of over heating is determined.
[054] There are many mature lithium iron phosphate battery pack charging equipment on the market, and it is equipped with a balanced charging circuit. Its principle is: first charge the battery pack with a larger current, and charge it as long as one of the battery packs reaches 3.7V. Equalizing charging, the equalizing charging loop current is very small, generally only tens of milliamperes, which has no effect on the battery pack, especially the power battery pack. Only one part of the entire battery pack is fully charged, and the other batteries are (seriously) under charged for a long time, resulting in the battery The external performance of the group is that the storage capacity decays too fast, which leads to the misunderstanding that the lithium iron phosphate battery is less cost-effective.
[055] None of the above listed prior art describe an equalizer that intelligently equalizes the battery charge during the charging of the batteries and increases the life of the battery.
[056] With all the above discussed restrictions or limitations, it is essential to have an intelligent equalizer and its method of charge equalization which enhances the life of the batteries connected in series.
OBJECTS OF THE INVENTION
[057] The primary object of the present invention is to provide an artificial intelligence based battery equalizer and a method for equalizing charge with a means for continuously equalizing the voltages of the individual cells of storage batteries under conditions of service use.
[058] Another object of the present invention is to provide an artificial intelligence based battery equalizer and a method for equalizing charge which increases the life of the battery.
SUMMARY
[059] According to this invention, an intelligent battery equalizer is provided for equalizing charge on a string of batteries connected in series/ parallel as the batteries are being charged/discharged/remain idle. The system comprises an artificial intelligence based central server at remote location which controls and programs the charge controller 2 using PC or hand held device. The batteries are being charged through the equalizer. The control unit 2 is connected to the charging connect/ disconnect unit which is in connection with the charging/discharging device. The cutoff limits are being set in the equalizer through the central server which are being set as per the need through communication interface. If any of the battery from the string reaches to such as but not limited to the overcharge, deep discharge etc, it will cut off the power backup/charging system. It calculates the charging time of each cell and will monitor the variation in charging time. Based on the charging time enhancement the central server analyzed the condition of cell/ battery and accordingly decides the charging time of the battery/ cell.
[060] In another embodiment of the present invention, display device can be liquid crystal display, light emitting diodes or any other
BREIF DESCRIPTION OF THE ACCOMPANYING DRAWINGS:
[061] Further objects and advantages of this invention will be more apparent from the ensuing description when read in conjunction with the accompanying drawings and wherein:
[062] Fig 1 shows block diagram of improved battery charge equalizer according to the present invention.
DETAILED DESCRIPTION OF THE INVENTION WITH REFERENCE TO THE ACCOMPANYING DRAWINGS:
[063] Reference may be made to fig 1 which shows block diagram of an artificial intelligence based battery equalizer system. The system comprises an artificial intelligence based central server at remote location which controls and programs the charge controller 2 using PC or hand held device. The charge controller 2 signals a plurality of switches. The equalizer is positioned between power backup system and a plurality of batteries 1. The current protection unit 3 having in which a voltage/temperature sense unit collects data from batteries (and/or individual battery cells thereof) as they operate in various environments.; stores and compares the data at cloud based server. The PWM driver controlled by said charge controller 2 driving isolated multiple gate output, power transfer section 7 wherein charging and discharging current is settable and is changed according to the need characterized that pattern identification performed.
[064] A communication interface is providing to receives and sends data messages from one equalizer unit to another; a switch to connect or disconnect for the charging or discharging device (15).
[065] The battery may generally be any type of battery. Examples of battery types include, but are not limited to, rechargeable, non-rechargeable, lead acid, alkaline, nickel-metal hydride (NiMH), lithium ion, nickel cadmium (NiCad), etc.
[066] A paralleling interface section 10 having connector is provided to receive (Rx) and transmit (TX) the data messages from one equalizer unit to another. The display 11 unit shows the measured parameters and paralleling interface section 10 is provided to have parallel inter unit monitoring and control of a number of battery equalizer units connected to N number of batteries in series.
[067] The batteries are being charged through the equalizer. The control unit 2 is connected to the charging connect/ disconnect unit 14 which is in connection with the charging/discharging device 15. The cutoff limits are being set in the equalizer through the central server which are being set as per the need through communication interface. If any of the battery from the string reaches to such as but not limited to the overcharge, deep discharge etc, it will cut off the power backup/charging system. It calculates the charging time of each cell and will monitor the variation in charging time. Based on the charging time enhancement the central server analyzed the condition of cell/ battery and accordingly decides the charging time of the battery/ cell.
[068] Equalization of charge on multiple series/ parallel connected batteries is accomplished in accordance with the present invention rapidly and substantially without unnecessary dissipation of power. Equalization is accomplished automatically without requiring comparison of voltages across individual cells or batteries. Further, the present invention supplies current to a cell unit in proportion to the difference in the voltages between cell units so that the lowest charged cell unit receives the greatest charging current from the highest charged cell/ battery unit while cell/ battery units at voltages intermediate the highest and lowest (where more than two cell units are being charged) receive lesser charge currents. Energy is transferred in this manner from the most highly charged cell unit to the cell unit or units having lesser charge.
[069] In case of large imbalance in the series connected batteries, the main controller regulates the equalization process by controlling the active and inactive time of the equalization. With a large imbalance, the equalizer turns on and then remains off for a certain time period so that the switching devices can be protected from extremely high temperature which could lead to failure of switching devices. The present system and method are usable irrespective of the battery as well as power back up system technology. Voltage will be matched between adjacent batteries and power backup system regardless of chemistry, manufacturer or capacity. The system and method yield a low cost implementation. Precise equalization is achieved without any requirements for device matching or tight tolerances. This contrasts strongly with active methods that can equalize only if several different circuits match precisely. The concept is modular and extends to arbitrary numbers of batteries. Batteries can be added without any system redesign by providing each additional battery with a module. The process is self-limiting. The battery voltage equalizer is bidirectional. It transfers energy from any battery at a higher voltage to the other battery at a lower voltage.
[070] Numerous modifications and adaptations of the system of the present invention will be apparent to those skilled in the art, and thus it is intended by the appended claims to cover all such modifications and adaptations which fall within the true spirit and scope of this invention.

WE CLAIM

1. An artificial intelligence based battery equalizer system comprising of
a. an artificial intelligence based central server (1A) at remote location which controls and programs the charge controller using PC or hand held device;
b. charge controller (2) signaling a plurality of switches wherein said equalizer is positioned between power backup system and a plurality of batteries;
c. current protection unit (3) in which a voltage/temperature sense unit collects data from batteries (and/or individual battery cells thereof) as they operate in various environments; stores and compares the data at cloud based server;
d. PWM driver (2A) controlled by said charge controller driving isolated multiple gate output;
e. power transfer section (7) wherein charging and discharging current is settable and is changed according to the need characterized that pattern identification performed;
f. a communication interface (2C) section to receive and send data messages from one equalizer unit to another;
g. a switch to connect or disconnect for the charging or discharging device (15).
2. The artificial intelligence based battery equalizer system, as claimed in claim 1, wherein the cutoff limits are being set in the equalizer through the central server which are being set as per the need through communication interface.
3. The artificial intelligence based battery equalizer system, as claimed in claim 1, wherein the central system calculates the charging time of each cell and will monitor the variation in charging time and based on the charging time enhancement the central server analyzed the condition of cell/ battery and accordingly decides the charging time of the battery/ cell.