Method and system for building a state of health (SOH) prediction model for a battery

Abstract

A method and system for building a SOH prediction model for a battery, is disclosed. The method comprises receiving a dataset comprising gas emission data from at least one QD sensor and performance data of at least one battery cell of battery. Then, processing dataset to derive at least one feature indicative of performance of battery and to predict states of concentration of at least one type of gas and at least one parameter. Further building SOH prediction model to predict degradation of at least one component of at least one battery cell based on processed dataset and to compute a SOH of battery cell in relation to degradation of at least one component. The building includes training SOH prediction model to correlate concentration of at least one type of gas and at least one battery parameter based on derived at least one feature and predicted states.

Description

Internal Ref.: 2024P03445WOMETHOD AND SYSTEM FOR BUILDING A STATE OF HEALTH (SOH) PREDICTION MODEL FOR A BATTERYTECHNICAL FIELD

[0001] The present invention generally relates to the field of batteries, and more particularly relates to a method and system for building a state of health (soh) prediction model for a battery.BACKGROUND

[0002] The following description includes information that may be useful in understanding the present invention. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed invention, or that any publication specifically or implicitly referenced is prior art. Currently, vehicles are increasingly being equipped with powerful batteries for efficient functioning of the vehicles. To determine efficient functioning of a vehicle one of a key factor is to estimate a State of Health (SOH) of a battery associated with the vehicle.

[0003] The State of Health (SOH) of the battery in the vehicle is expressed as a ratio of possible charge capacity of the battery at a present time, as compared to an original or ideal charge capacity of the battery. The ratio is often expressed as a percentage, where 100% represents the ideal SOH, and 80% represents a battery SOH with a maximum presently possible charge capacity of 80% of the ideal charge capacity. The battery can be considered unusable when the battery SOH is less than 80%.

[0004] Conventionally, the SOH of the battery is determined by monitoring several battery parameters during a charging cycle and a discharging cycle of the battery. Experimental analysis method for battery SOH estimation is crucial during development phase. Advanced sensors are getting used under Indirect observational method of experimental analysis E.g., ultrasonic & temperature sensor. However, such methods are unable to provide accurate prediction and also have low accuracy of prediction during cell relaxation.

[0005] Conventional gas sensors exhibit non-discriminative response to a range of interfering gases due to cross sensitivity of the gas sensors that leads to difficulty in identifying desired gases. The conventional gas sensors may consume high power for measures the gases and the conventional gas sensors detects gases only during thermal runaway condition as high concentration of gases will be emitted. Gas Chromatography Mass Spectrometry (GC-MS) may not detect gases emitted at extremely low concentrations (sub-ppm or ppb levels), especially in normal battery operating. In some conditions such as complex mix of gasesInternal Ref.: 2024P03445WO(e.g., CO2, H2, CH4, volatile organic compounds) that can overlap in chromatographic peaks, complicating identification, quantification at low concentrations, “Particularly dynamic load test” conditions where gas emissions are minimal, conventional gas sensors does not detect the gases accurately.

[0006] Data processing and feature extraction is difficult with the conventional gas sensors due increase in a response time and limitation towards detection the gases. Thus, leading to inaccurate estimation of the SOH. Also, conventionally the gases are considered only during thermal runaway condition as high concentration gases are release from the battery. Further, in some cases a gas spectrometer is used for determining the gases, however the gas spectrometer is costly and heavy. Further, the gas spectrometer does not accurately detect the gas concentration when there is an emission of low-gas concentration. To overcome the above problems there is a need for accurately determine the SOH the battery.

[0007] Few of the existing technologies available in the art which deals with battery sensing system are US20230387483A1 that discloses a test platform for scanning the batteries using ultrasonic scans.SUMMARY

[0008] The present disclosure overcomes one or more shortcomings of the prior art and provides additional advantages. Embodiments and aspects of the disclosure described in detail herein are considered a part of the claimed disclosure.

[0009] In one non-limiting embodiment of the present disclosure, a method for building a State of Health (SOH) prediction model for a battery, is disclosed. The method comprises receiving a dataset comprising gas emission data from at least one Quantum Dot (QD) sensor and performance data of at least one battery cell of the battery. The gas emission data is indicative of concentration of at least one type of gas released from the at least one battery cell. The performance data comprises at least one battery parameter relating to performance of the at least one battery cell measured during the release of the at least one type of gas. The method further comprises processing the dataset to derive at least one feature indicative of performance of the battery and to predict states of the concentration of at least one type of gas and the at least one parameter. Moving ahead, the method further building the SOH prediction model to predict degradation of at least one component of the at least one battery cell based on the processed dataset and to compute a SOH of the battery cell in relation to the degradation of the at least one component. The building includes training the SOH prediction model to correlate the concentration of the at least one type of gas and the at least one battery parameter based on the derived at least one feature and the predicted states.Internal Ref.: 2024P03445WO

[0010] In another non-limiting embodiment of the present disclosure, wherein the at least one type of gas comprises at least of: hydrogen, carbon dioxide, volatile organic compounds, and carbon monoxide.

[0011] In another non-limiting embodiment of the present disclosure, wherein the at least one parameter comprises at least one of: a current, a voltage, a temperature, a resistance, type of the battery, battery information, driving profile of a vehicle, charging and discharging profile of the battery cell.

[0012] In another non-limiting embodiment of the present disclosure, wherein the gas emission data is measured using at least one of an electro chemical resistive technique and photo luminous based technique.

[0013] In another non-limiting embodiment of the present disclosure, wherein the at least one component comprises at least of: a Solid Electrolyte Interphase (SEI), an electrolyte, a dendrite, a separator, an anode and a cathode.

[0014] In yet another non-limiting embodiment of the present disclosure, estimating SOH of the at least one battery bell using the trained SOH prediction model. The method comprises receiving data relating to the performance of the at least one battery cell, wherein the data comprises at least one battery parameter measured during an operational cycle of the vehicle. The method further comprises predicting degradation of the at least one component of the at least one battery cell by processing the data using a SOH prediction model. Moving ahead, the method further comprises estimating the SOH of the battery based on the predicted degradation of the at least one component by using the SOH prediction model.

[0015] In another non-limiting embodiment of the present disclosure, wherein the trained SOH prediction model estimates SOH for the battery associated with a vehicle and the battery in storage condition.

[0016] In another non-limiting embodiment of the present disclosure, a system for estimating a State of Health (SOH) of a battery, is disclosed. The system comprises a memory and a processor which is electronically coupled to the memory. The processor is configured to receive a dataset comprising gas emission data from at least one Quantum Dot (QD) sensor and performance data of at least one battery cell of the battery. The gas emission data is indicative of concentration of at least one type of gas released from the at least one battery cell. The performance data comprises at least one battery parameter relating to performance of the at least one battery cell measured during the release of the at least one type of gas. The processor is configured predicts process the dataset to derive at least one feature indicative of performance of the battery and to predict states of the concentration of at leastInternal Ref.: 2024P03445WO one type of gas and the at least one parameter. Moving ahead the processor is configured to build the SOH prediction model to predict degradation of at least one component of the at least one battery cell based on the processed dataset and to compute a SOH of the battery cell in relation to the degradation of the at least one component. The building includes training the SOH prediction model to correlate the concentration of the at least one type of gas and the at least one battery parameter based on the derived at least one feature and the predicted states.

[0017] In yet another embodiment of the present disclosure, wherein the at least one QD sensor are arranged at different locations in the battery, wherein the different locations comprise at least one of: current collector of at least one battery cell of the battery, and a cover of the battery.

[0018] In another non-limiting embodiment of the present disclosure, a battery module, is disclosed. The battery module comprises at least one Quantum Dot (QD) sensor, a battery comprising at least one battery cells, and a battery management unit. Each of the at least one battery cell comprises at least one component comprising of: a Solid Electrolyte Interphase (SEI), an electrolyte, a dendrite, a separator, an anode and a cathode. The battery management unit comprises a SOH prediction model to estimate a SOH of the battery. The SOH prediction model is built using a dataset comprising gas emission data from at least one Quantum Dot (QD) sensor and performance data of at least one battery cell of the battery, wherein the gas emission data is indicative of concentration of at least one type of gas released from the at least one battery cell, and wherein the performance data comprises at least one battery parameter relating to performance of the at least one battery cell measured during the release of the at least one type of gas.

[0019] The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description.BRIEF DESCRIPTION OF DRAWINGS

[0020] The features, nature, and advantages of the present disclosure will become more apparent from the detailed description set forth below when taken in conjunction with the drawings in which like reference characters identify correspondingly throughout. Some embodiments of system and / or methods in accordance with embodiments of the present subject matter are now described, by way of example only, and with reference to the accompanying Figs., in which:Internal Ref.: 2024P03445WO

[0021] FIG. 1 depicts an exemplary environment for building a State of Health (SOH) prediction model for a battery, in accordance with embodiments of the present disclosure;

[0022] FIG. 2 depicts an exemplary block diagram illustrating a system for building a State of Health (SOH) prediction model for a battery, in accordance with embodiments of the present disclosure;

[0023] FIG. 3 depicts an exemplary block diagram illustrating a battery module for estimating a State of Health (SOH) of a battery, in accordance with embodiments of the present disclosure;

[0024] FIG. 4 represents an exemplary block diagram illustrating prediction of degradation of components of the battery, in accordance with embodiments of the present disclosure;

[0025] FIG. 5a-5b represents an exemplary circuit illustrating measurement of gases using QD sensor, in accordance with embodiments of the present disclosure;

[0026] FIG. 6 represents a graph of SOH of a battery, in accordance with embodiments of the present disclosure; and

[0027] FIG. 7 represents flowchart of an exemplary method for building a State of Health (SOH) prediction model for a battery, in accordance with embodiments of the present disclosure;

[0028] It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems embodying the principles of the present subject matter. Similarly, it will be appreciated that any flow charts, flow diagrams, state transition diagrams, pseudo code, and the like represent various processes which may be substantially represented in a computer readable medium and executed by a computer or processor, whether or not such computer or processor is explicitly shown.DETAILED DESCRIPTION

[0029] The foregoing has broadly outlined the features and technical advantages of the present disclosure in order that the detailed description of the disclosure that follows may be better understood. It should be appreciated by those skilled in the art that the conception and specific embodiment disclosed may be readily utilized as a basis for modifying or designing other structures for carrying out the same purposes of the present disclosure.

[0030] The novel features which are believed to be characteristic of the disclosure, both as to its organization and method of operation, together with further objects and advantages will be better understood from the following description when considered in connection with the accompanying figures. It is to be expressly understood, however, that each of the figures isInternal Ref.: 2024P03445WO provided for the purpose of illustration and description only and is not intended as a definition of the limits of the present disclosure.

[0031] Conventionally, gases released from a battery is not considered for estimating a State of Health (SOH) of a battery as conventionally known gas sensors are low in accuracy and performance. Further, the conventional gas sensors exhibit non-discriminative response to a range of interfering gases due to cross sensitivity of the gas sensors that leads to difficulty in identifying desired gases. Thus, leading to inaccurate estimation of the SOH.

[0032] To overcome the above-mentioned challenges, the present disclosure provides a method for building a State of Health (SOH) prediction model for a battery. In the present disclosure, the SOH prediction model may be trained using a dataset comprising concentration of gases released form the battery, that may be measured using Quantum Dot (QD) sensors and values associated with battery parameters may be received from sensors. Then, the SOH prediction model may process the dataset to derive features and predict states of the concentration of at least one type of gas and the at least one parameter. Further, based on the processing and predicting the SOH prediction model may be built to predict degradation of at least one component of the at least one battery cell. The SOH prediction model may be built by training the SOH prediction model to correlate the concentration of the at least one type of gas and the at least one battery parameter based on the derived at least one feature and the predicted states. Thus, by correlating the gas concentration with battery parameters, the present invention estimates the SOH of the battery using the SOH prediction model more accurately.

[0033] FIG. 1 depicts an exemplary environment 100 for building a State of Health (SOH) prediction model for a battery, in accordance with embodiments of the present disclosure. The exemplary environment 100 particularly depicts a test bench 108 that may incorporate the battery 101, system 102 and a sensor 106. The test bench 108 may be used to build the SOH prediction model. In an exemplary embodiment, the test bench may be any experimental setup for building the SOH prediction model. In an exemplary embodiment, the test bench 108 may include any know equipments / components. The battery may be a Lithium-ion battery, Nickel-Metal Hydride battery, Lead-Acid battery, Lithium- Sulphur battery or any other battery that may be used for performing operations of a vehicle.

[0034] In some implementations, the battery 101 may comprise at least one battery cell 107. In some implementations, the system 102 may comprise, a processor 103, a memory 104, andInternal Ref.: 2024P03445WO a SOH prediction model 105. The system 102 may include other components (not shown in this fig.) to implement desired functions of the system 102. The SOH prediction model 105 may be any suitable supervised machine learning model such as but not limited thereto a regression model that is trained to perform desired functions of the present invention.

[0035] In an exemplary implementation, the processor 103 may receive a dataset comprising gas emission data from at least one Quantum Dot (QD) sensor and performance data of at least one battery cell 107 of the battery 101. In an exemplary embodiment, the gas emission data may be indicative of concentration of at least one type of gas released from the at least one battery cell 107. In an exemplary embodiment, the performance data comprises at least one battery parameter relating to performance of the at least one battery cell 107 during the release of the at least one type of gas. In a non-limiting embodiment, the at least one battery parameter may include a current, a voltage, a temperature, a resistance, type of the battery, battery information, driving profde of a vehicle, charging and discharging profile of the battery cell. The vehicle may not be limited to an electric vehicle, hybrid vehicle. In a nonlimiting embodiment, the vehicle may be any vehicle that may use the battery for operating the vehicle. Then, the processor 103 may process the dataset to derive at least one feature indicative of performance of the battery and to predict states of the concentration of at least one type of gas and the at least one parameter. In a non-limiting embodiment, at least one component may be a Solid Electrolyte Interphase (SEI), an electrolyte, a dendrite, a separator, an anode and a cathode. Further, the processor 103 may building the SOH prediction model 105 to predict degradation of at least one component of the at least one battery cell 107 based on the processed dataset and to compute a SOH of the battery cell 107 in relation to the degradation of the at least one component. The building includes training the SOH prediction model 105 to correlate the concentration of the at least one type of gas and the at least one battery parameter based on the derived at least one feature and the predicted states. A detailed explanation of the system 102 is provided in the forthcoming paragraphs in conjunction with FIG.s 2-8.

[0036] FIG. 2 depicts an exemplary block diagram illustrating details of a system 200 (which is system 102 of FIG.1) for estimating SOH of a battery which is a part of the vehicle, in accordance with embodiments of the present disclosure. As illustrated in the Figure, the vehicle may also comprise one or more sensors 201 and at least one QD sensor 204. In some implementations, the system 200 may comprise, a processor 202, a memory 203, a SOH prediction model 205.Internal Ref.: 2024P03445WO

[0037] In one implementation, the processor 202 may 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 202 may be configured to fetch and execute computer-readable instructions and other information stored in the memory 203.

[0038] In some implementations, the memory 203 may include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, one or more memory devices, flash memory devices, etc., and combinations thereof. In an embodiment, data / information such as data relating to concentration of at least one gas and data relating to performance of the at least one battery cell 107 may be stored within the memory 203 in the form of various data structures. The memory 203 may also store other data 203b such as temporary data and temporary files, generated by the processor 202 or any other parts of the system 200 including a component degradation prediction unit 205a, a SOH prediction unit 205b, for performing the various functions of the present invention. The data relating to concentration of at least one gas may comprise Parts Per Million (PPM) values of at least gas relative to degradation of the at least one component. The data relating to performance of the at least one battery cell 107 may comprise at least one battery parameter measured during emission of the at least one gas from the at least one battery cell 107.

[0039] In some implementations, the SOH prediction model 205 may comprise the component degradation prediction unit 205a, the SOH prediction unit 205b. In some implementations, the processor 202 may be operatively coupled to the memory 203 and the SOH prediction model 205 that cooperate to estimate the SOH of the battery 101 of the vehicle.

[0040] In the illustrated figure 2, the SOH prediction model 205 are shown to reside outside the processor 202 and may 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. In said implementation, the component degradation prediction unit 205a, the SOH prediction unit 205b may perform prediction of the degradation of the at least one component and estimate the SOH of the battery in the present invention. However, one of ordinary skill will appreciate that in other implementations, the SOH prediction model 205 may also formInternal Ref.: 2024P03445WO a part of the processor 202 and may be implemented through software or hardware or a suitable combination of software and hardware as per the implementation requirements of the present disclosure. In said implementation, the processor 202 may perform all the functions carried out by the different units / components such as the component degradation prediction unit 205a and the SOH prediction unit 205b.

[0041] In one non-limiting example, the SOH prediction model 205 are trained Al models that use supervised learning algorithms for prediction / estimation of capacity of the battery. In one example, but not limited thereto, the SOH prediction model 205 is a linear regression model. In another example, but not limited thereto, the SOH prediction model 205 is a nonlinear regression model. In some other examples, but not limited thereto, the SOH prediction model 205 may be a Recurrent Neural Network (RNN) or a Computational Neural Network (CNN) or a combination of the RNN and the CNN. However, a person of ordinary skill will appreciate that any other suitable Al model that serves the purpose of the present disclosure may be utilized for estimating the SOH of the battery 101.

[0042] Though Fig. 2 is described below for building a State of Health (SOH) prediction model for estimating the SOH of the battery, in this description the battery in context of a scenario where the test bench is built for test a State of Health (SOH) of the battery not installed in vehicle (i.e. the storage battery). It may also be appreciated that the present disclosure is equally applicable to a context when the battery is installed in the vehicle and the vehicle is in operating condition.

[0043] Continuing to Fig. 2, the test bench 108 may be built for estimating the SOH of the battery. The present description may be in context to building the SOH prediction model 205 in the test bench 108 for estimating the SOH of the battery. Operation power in the battery 101 of the vehicle may be used for operating the vehicle. The SOH prediction model 205 may receive a dataset comprising the gas emission data and the performance data from the at least one QD sensor 204 and the one or more sensors 201, respectively.

[0044] Further, the at least one QD sensor 204 may sense the gas emission data that may be indicative of concentration of at least one type of gas released from the at least one battery cell 107. The at least one QD sensor 204 may comprise several quantum dots. The quantum dot is a nanometre-sized semiconductor particle traditionally with a core-shell structure. The Quantum dots emit light of specific wavelengths if energy is applied to them. These wavelengths of light can be accurately tuned by changing various properties of the particle,Internal Ref.: 2024P03445WO including shape, material composition, and size. The quantum dot sensor is a type of advanced optical sensor that utilizes semiconductor nanocrystals, known as quantum dots, to detect and measure various forms of light and electromagnetic radiation. The at least one QD sensor 204 may comprise different semiconductor particles for capturing different gases. The semiconductor particles may be tuned to capturing particular gases. Thus, helps in accurately measuring the concentration of different gases released by the battery 101. Further, the present disclosure may prevent cross-sensitivity by measuring different gas concentration simultaneously by tuning the quantum dots of the at least one QD sensor 204 with respect to each of the respective gases.

[0045] In an exemplary embodiment, the processor 202 may receive the concentration of at least one type of gas from the one or more sensors 201 that are coupled to or otherwise included within the test bench 108 and may store the concentration of at least one type of gas in the memory 203. In a non-limiting embodiment, the at least one type of gas may include but is not limited to, hydrogen, carbon dioxide, volatile organic compounds, and carbon monoxide.

[0046] In some implementations, the at least one QD sensor 204 may be arranged at different locations of the battery 101. In a non-limiting embodiment, the at least one QD sensor 204 may be arranged at a current collector (no shown in Figures) of at least one battery cell 107 of the battery, and a cover of the battery 101.

[0047] In an exemplary embodiment, the at least one QD sensor 204 may continuously monitor each of the at least one battery cell 107 of the battery 101 for measuring the at least one type of gas concentration of at least one type of gas. When there is a release of the at least one type of gas, the at least one QD sensor 204 may measure the concentration of at least one type of gas by determining change in energy level of the quantum dots of the at least one QD sensor 204, when the quantum dots are exposed to the at least one type of gas released from the at least one battery cell 107. In case, the at least one type of gas is not released from any of the at least one battery cell, the battery 101 may be determined to be in a heathy state.

[0048] In an exemplary embodiment, an electro chemical resistive technique may be used for measuring the concentration of at least one type of gas. In the electro chemical resistive technique, a voltage 502 may supplied to QD sensor 501, the change in energy level may be determined in terms of resistance value 503 of the QD sensor 501, when response toInternal Ref.: 2024P03445WO exposure of the at least one gas released from the at least one battery cell 107 of the battery 101 as shown in Fig. 5a. Further, the concentration of at least one type of gas may be determined based on the resistance value 503.

[0049] In other exemplary embodiment, a photo luminous based technique may be used for measuring the concentration of at least one type of gas. In the photo luminous based technique, an Ultra Voltage (UV) source 507 may be used to supply power to QD sensor 506, the change in energy level may be determined by a change in amplitude 508 of the electrical signal passing through the QD sensor in response to exposure of the at least one gas released from the at least one battery cell 107 of the battery 101 as shown in Fig. 5b. Further, the concentration of at least one type of gas may be determined based on the change in the amplitude 508. As the present disclosure considers gas concentration for predicting the degradation of the at least one component, the SOH of the battery may be estimated more accurately. Further, in the present disclosure the gas concentration is measured using at least one QD sensor 204. Thereby, the gas emission data may be accurate, that in turn helps in accurately determining the SOH of the battery. As the present invention uses at least one QD sensor 204 for measuring gas concentration, the present system is unique and cost effective.

[0050] In an exemplary scenario, the one or more sensors 201 installed on the vehicle may sense performance parameter of the at least one battery cell 107 during the release of the at least one type of gas. In an exemplary embodiment, the processor 202 may receive the at least one parameter from the one or more sensors 201 that are coupled to or otherwise included within the vehicle and may store the at least one parameter in the memory 203. The nonlimiting examples of the one or more sensors 201 may include temperature sensors, pressure sensors, motion sensors, voltage sensors, current sensors, internal battery sensors, external battery sensors, energy management sensors, temperature sensor and / or other sensors, etc. In a non-limiting embodiment, the at least one battery parameter may comprise a current, a voltage, a temperature, a resistance, type of the battery, battery information, driving profde of a vehicle, charging and discharging profile of the battery cell. For example, the battery information may comprise information related to type of the battery, capacity of the battery and the like. For example, the driving profile of the vehicle may comprise speed of the vehicle, distance covered by the vehicle and the like.

[0051] In an exemplary embodiment, the processor 202 may receive the dataset comprising the gas emission data and the performance data with respect to each duty cycle of the batteryInternal Ref.: 2024P03445WO101 from the at least one QD sensor 204 and the one or more sensors 201, respectively for estimating the SOH of the battery 101. In a non-limiting embodiment, the duty cycle may refer to a combination of charging cycle and a discharging cycle of the battery 101 with respect to one compete operational cycle of the battery 101. In an exemplary embodiment, the dataset may multivariate data that includes the concentration of the at least one gas type and the voltage, the temperature, the resistance, the current and the like. The multivariate dataset may be pre-processed. The data of the dataset may be indicative of the time series data. Then, the time series data with respect to each of the datasets may be combined to a single dataset. Further, the time series dataset may be extracted to process further.

[0052] During building the SOH prediction model 205 may collect a dataset. The dataset may comprise data relating to concentration of at least one gas and data relating to performance of the at least one battery cell 107. The data relating to the concentration of at least one gas comprises Parts Per Million (PPM) values of at least gas relative to degradation of the at least one component. The data relating to the performance of the at least one battery cell 107 comprises at least one battery parameter measured during emission of the at least one gas from the at least one battery cell. In an exemplary embodiment, the at least one battery parameter may include but not limited to, the current, the voltage, the temperature, the resistance, the type of the battery, the battery information, the driving profde of the vehicle, the charging and discharging profde of the battery cell 107. The at least one gas may include but not limited to, the hydrogen, the carbon dioxide, the volatile organic compounds, and the carbon monoxide.

[0053] In an exemplary embodiment, the data relating to the concentration of each of the at least one gas released by the at least one battery cell 107 may be in relative by the degradation of the respective at least one component. For example, the degradation of the SEI may release a combination of the hydrogen and the carbon dioxide. For example, the degradation of the electrolyte may release a combination of the volatile organic compounds and the carbon monoxide. For example, the degradation of the dendrite may release a combination of the hydrogen, the carbon dioxide and the carbon monoxide. For example, the degradation of the dendrite may release a combination the carbon dioxide, the volatile organic compounds, and the carbon monoxide.

[0054] In an exemplary embodiment, the release of the at least one gas due to the degradation of the respective at least one component may be determined using a chemical reaction that may occur inside each of the at least one battery cell 107, due to the operation of the at leastInternal Ref.: 2024P03445WO one battery cell 107. For instance, the below equation (1) may show the SEI degradation by releasing the combination of the hydrogen and the carbon dioxide.(CH2OCO2Li2)2 > Li2CO2 + C2H4 + CO2 + I2O2 . (1)

[0055] For instance, the below equation (2) may show the degradation anode with electrode by releasing the combination of the volatile organic compounds and the carbon monoxide. C3H6O3 (DMC) + e- + 2Li + H2> L12CO3 + 2CH4. (2)

[0056] For instance, the below equation (3) may show the degradation cathode by releasing the combination of the oxygen and other gases.2L1N1O2 (DMC) > L12O + 2N1O + 1 / 202 . (3)

[0057] Referring to Fig. 4, with respect to each of the duty cycles comprising a charging and a discharging cycle 409. For receiving the at least one battery parameters, a capacity test 402 may be implemented to obtain holding capacity 410 of the at one battery cell 107. Then, an impendence test 403 may be implemented to obtain resistance value 411 of the at one battery cell 107. An Independent component analysis (ICA) may be implemented to obtain voltage value and current value 412 of the at one battery cell 107. A temperature sensor 405 may provide the temperature in terms of degree Celsius 413. Then, the QD sensor 406 may provide the concentration of the at least one type of gas based on the change in the energy level of the quantum dots of the QD sensor 406.

[0058] Referring back to Fig. 2, upon receiving the dataset, the processor 202 may process the dataset to derive at least one feature indicative of performance of the battery 101. Then, the processor 202 may predict states of the concentration of the at least one gas and the at least one battery parameter. For instance, the at least one feature may include but not limited to ratios of the at least one gas, trends of the at least one component over time and a battery health feature. The at least one feature may be in relative to a particular duty cycle. For instance, the at least one features may be the ratios of the at least one gas determined for a first duty cycle, values associated with the at least one component such as the data relating to performance may be only the current, voltage, and temperature. At the same duty cycle the degradation of the at least one component may be the dendrite and the SEI. In a nonlimiting exemplary embodiment, the at least one feature may be the concentration of carbon dioxide (CO2) produced correlated with both voltage and temperature, methane (CH4) and ethene (C2H4) concentrations with temperature, carbon dioxide (CO2) concentration and evolution rate when the temperature above 40°C, and the upper cut-off voltage to 4.5 V. InInternal Ref.: 2024P03445WO another non-limiting exemplary embodiment, the predicted states for the above at least one feature may the state of current is 1c, the state of voltage is stable, and the state of temperature is 25-30 °C (as shown in exemplary Table 1). The predicted state with respect to degradation of the dendrite is <1 PPM and the SEI is <5 PPM (as shown in Table 2).

[0059] Further, the processor 202 may build the SOH prediction model 205 to predict the degradation of at least one component of the at least one battery cell 107 based on the processed dataset and to compute the SOH of the battery cell 107 in relation to the degradation of the at least one component. The building of the SOH prediction model 205 may include training the SOH prediction model 205 to correlate the concentration of the at least on gas with values associated with the at least one battery parameter based on the derived at least one feature and the predicted states. In a non-limiting embodiment, the SOH prediction model 205 may be trained using supervised machine learning (Regression model) for known relations and the SOH prediction model 205 may be trained using unsupervised machine learning model for unknown relation prediction between the PPM and other values associated with the at least one component. Combination of Recurrent Neural Network (RNN) and Computational Neural Network (CNN) may be used to train the SOH prediction model 205 to capture the relationships between the data relating to concentration of the at least one gas and the data relating to performance of the at least one battery cell 107. Finally, the degradation model 414 may receive all the above values with respect to each of the charging and discharging cycle 409 (as shown in Fig. 4).

[0060] For instance, the SOH prediction model 205 may be trained to correlate at a first duty cycle, if the current is 1c, voltage is stable, and temperature is 25-30 °C (as shown in exemplary Table 1), then the degradation of the at least component that is the dendrite and the SEI may be <1 PPM and <5 PPM respectively (as shown in exemplary Table 2). Upon training the SOH prediction model 205 for predicting the degradation of the at least one component, the SOH prediction model 205 may be trained to computing the SOH of the battery 101 in relation to the degradation of the at least one component. In a non-limiting example, the degradation percentage of all the components may be accumulated to form a total battery degradation value. The total battery degradation value may be subtracted with 100% to obtain the SOH of the battery as shown in equation (4) below.SOH = 100% - Total battery degradation percentage . (4)Total battery degradation percentage = degradation percentage of each of the component (5)Internal Ref.: 2024P03445WO

[0061] For instance, in case the degradation percentage of the SEI is 2%, the electrolyte is 1%, the anode is 1%, the cathode is 3% and the separator is 1%. Then, the total battery degradation percentage may be 10% obtained using equation (5). Further, the SOH of the battery 101 may be 100% - 10% = 90% using equation (4) as shown in exemplary Table 3.Table 1Table 2Internal Ref.: 2024P03445WOTable 3

[0062] Further, when the SOH prediction model 205 receives data relating to performance of the at least one battery cell 107, the SOH prediction model 205 may predict the degradation of the at least one component present in the battery of the at least one battery cell 107 by processing the data. In a non-limiting exemplary embodiment, the data comprises at least one battery parameter measured during an operational cycle of the vehicle. The operational cycle may refer to the duty cycle of the battery 101. The at least one battery parameter may include, but not limited to the current, the voltage, the temperature, the resistance, the type of the battery, the battery information, the driving profile of the vehicle and the charging and discharging profile of the battery cell 107. In a non-limiting embodiment, the SOH prediction model 205 may predict the degradation of the at one component by correlating the values associated with the at least one battery parameter, received in real time with predefined relationship built during the training stage between the gas emission data and the performance data, to predict the degradation percentage of each of the at least one component. Then, the SOH prediction model 205 may estimate the SOH of the battery based on the predicted degradation of the at least one component. Therefore, the present disclosure predicts the degradation of the at least one component of the battery more accurately as the prediction the degradation is based on the correlation between the gas emissions data and the performance data of the battery. For instance, in real time, if the SOH prediction model 205 receives the values of current as 1 c, the state of voltage as stable, and the state of temperature as 25-30 °C (as shown in exemplary Table 1), the SOH prediction model 205 may predict the degradation of the dendrite as <1 PPM and the SEI as <5 PPM. In an exemplary embodiment, in real time the SOH prediction model may not receive the gas emission data, however based on the correlation between the gas emission data and the performance data during the training, the SOH model predicts the degradation of the at least one component of the battery 101.

[0063] Further, the processor 202 may estimate the SOH of the battery based on the predicted degradation percentage of the at least one component as shown in FIG. 6. A y axis may represent a SOH of the battery, and the x axis may represent the duty cycle of the battery. The graph shows that a first cell 601 may represent degradation of the Solid Electrolyte Interphase (SEI)A, the electrolyte a and the separator 0. Similarly, the second cell 602 mayInternal Ref.: 2024P03445WO represent degradation of the SEI A, the electrolyte a and the separator 0. The present disclosure determines the SOH of the battery more accurately, as the present invention estimates the SOH based on the prediction of the degradation of the components by correlating the gas emission data and the performance data of the battery.

[0064] FIG. 3 depicts an exemplary block diagram illustrating a battery module for estimating a State of Health (SOH) of a battery, in accordance with embodiments of the present disclosure. The exemplary environment 300 depicts a storage battery scenario. In particularly to the storage battery scenario, the battery may be in a storage condition and not placed inside any vehicle. The battery module 301 may be incorporated with the battery 101, a QD sensor 302, a battery management unit 303. The battery 101 may comprise the at least one battery cells 107 and the at least one component 304. The at least one component may include, but not limited to, the Solid Electrolyte Interphase (SEI), the electrolyte, the dendrite, the separator, the anode and the cathode. In storage battery scenario, the battery management unit 303 may estimate the SOH of the battery 101. The battery management unit 303 may receive the gas emission data and the performance data. The gas emission data is indicative of the concentration of the at least one type of gas released from the at least one battery cell 107. The performance data comprises at least one parameter relating to performance of the at least one battery cell 107 during the release of the at least one type of gas. Further, the battery management unit 303 may predict the degradation of the at least one component of the at least one battery cell 107 by processing the gas emission data and the performance data by using the SOH prediction model. In the storage battery scenario, the SOH prediction model may be coupled with the battery management unit 303. In other exemplary embodiment, the SOH prediction model may reside out the battery management unit 303. Finally, the battery management unit 303 may estimate the SOH of the battery based on the predicted degradation of the at least one component by using the SOH prediction model. Therefore, the present disclosure provides the battery pack for estimating the SOH of the battery in the storing conditions.

[0065] FIG. 7 represents flowchart of an exemplary method for estimating resistance of a battery, in accordance with embodiments of the present disclosure. The order in which the method 700 is described is not intended to be construed as a limitation, and any number of the described method blocks may be combined in any order to implement the method. Additionally, individual blocks may be deleted from the methods without departing from the spirit and scope of the subject matter described. Furthermore, the method can be implemented in any suitable hardware, software, firmware, or combination thereof.Internal Ref.: 2024P03445WOHowever, for ease of explanation, in the embodiments described below, the method 700 may be considered to be implemented by the respective components and / or by the processor 202 and / or the SOH prediction model 205 of FIG. 2.

[0066] At step 701, the method may include receiving the gas emission data from the at least one Quantum Dot (QD) sensor and the performance data of the at least one battery cell of the battery. The gas emission data is indicative of concentration of the at least one type of gas released from the at least one battery cell. The performance data comprises at least one parameter relating to performance of the at least one battery cell 107 during the release of the at least one type of gas. In one implementation, the processor 202 may receive the gas emission data battery and the performance data. In another implementation, the SOH prediction model 205 may receive the gas emission data battery and the performance data.

[0067] At step 702, the method may include processing the dataset to derive at least one feature indicative of performance of the battery and to predict states of the concentration of at least one type of gas and the at least one parameter. In one implementation, the processor 202 may processing the dataset and predict the states. In another implementation, the SOH prediction model 205 may processing the dataset and predict the states.

[0068] At step 703, the method may include building the SOH prediction model to predict degradation of at least one component of the at least one battery cell based on the processed dataset and to compute the SOH of the battery cell in relation to the degradation of the at least one component. The building includes training the SOH prediction model to correlate the concentration of the at least one type of gas and the at least one battery parameter based on the derived at least one feature and the predicted states. In one implementation, the processor 202 may trained to estimate the SOH of the battery. In another implementation, the SOH prediction model 205 may be trained to estimate the SOH of the battery 101.

[0069] The order in which the method 700 is described is not intended to be construed as a limitation, and any number of the described method blocks may be combined in any order to implement the method. Additionally, individual blocks may be deleted from the methods without departing from the spirit and scope of the subject matter described.

[0070] 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 blocksInternal Ref.: 2024P03445WO 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.

[0071] 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 and spirit of the disclosed embodiments.

[0072] 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., are non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, non-volatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.

[0073] Suitable processors include, by way of example, a general-purpose processor, a special purpose processor, a conventional processor, a digital signal processor (DSP), a graphic processing unit (GPU), a plurality of microprocessors, one or more microprocessors in association with a DSP core, a controller, a microcontroller, Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) circuits, any other type of integrated circuit (IC), and / or a state machine.

[0074] Advantages of the embodiment of the present disclosure are illustrated herein-

[0075] As previously indicated, aspects of the present disclosure enable predicting degradation of the at least one component of the battery. Further, the present disclosure estimated the SOH of the battery based on the degradation of the at least one component of the battery. As the present disclosure considers gas concentration for predicting the degradation of the at least one component while training the SOH prediction model, the SOH of the battery may be estimated more accurately. Further, in the present disclosure the gas concentration is measured using QD sensor. Thereby, the gas emission data may be accurate, that in turnInternal Ref.: 2024P03445WO helps in accurately determining the SOH of the battery. As the present invention uses QD sensors for measuring gas concentration, the present system is unique and cost effective. In the present invention, as the degradation of the at least one component is determined based on the gas concentration and the performance data of the battery, the degradation of each of the at least one component is predicted more accurately. Then, the SOH of the battery is determined accurately as the SOH of battery is determined based on the degradation of the at least one component. Further, the present disclosure estimates the SOH of the battery during the storage condition.

Claims

Internal Ref.: 2024P03445WOCLAIMS:

1. A method of building a State of Health (SOH) prediction model for a battery, the method comprising: receiving a dataset comprising gas emission data from at least one Quantum Dot (QD) sensor and performance data of at least one battery cell of the battery, wherein the gas emission data is indicative of concentration of at least one type of gas released from the at least one battery cell, and wherein the performance data comprises at least one battery parameter relating to performance of the at least one battery cell measured during the release of the at least one type of gas; processing the dataset to derive at least one feature indicative of performance of the battery and to predict states of the concentration of at least one type of gas and the at least one parameter; and building the SOH prediction model to predict degradation of at least one component of the at least one battery cell based on the processed dataset and to compute a SOH of the battery cell in relation to the degradation of the at least one component, wherein the building includes training the SOH prediction model to correlate the concentration of the at least one type of gas and the at least one battery parameter based on the derived at least one feature and the predicted states.

2. The method as claimed in claim 1, wherein the at least one type of gas comprises at least of: hydrogen, carbon dioxide, volatile organic compounds, and carbon monoxide.

3. The method as claimed in claim 1, wherein the at least one battery parameter comprises at least one of: a current, a voltage, a temperature, a resistance, type of the battery, battery information, driving profile of a vehicle, charging and discharging profile of the battery cell.

4. The method as claimed in claim 1, wherein the gas emission data is measured using at least one of: an electro chemical resistive technique and photo luminous based technique.

5. The method as claimed in claim 1, wherein the at least one component comprises at least of: a Solid Electrolyte Interphase (SEI), an electrolyte, a dendrite, a separator, an anode and a cathode.

6. The method as claimed in claim 1, further comprising estimating SOH of the at least one battery bell using the trained SOH prediction model, wherein the estimating comprising: receiving data relating to performance of the at least one battery cell, wherein the data comprises at least one battery parameter measured during an operational cycle of the vehicle;Internal Ref.: 2024P03445WO predicting degradation of the at least one component of the at least one battery cell by processing the data using a SOH prediction model; and estimating the SOH of the battery based on the predicted degradation of the at least one component by using the SOH prediction model.

7. The method as claimed in claim 1, wherein the trained SOH prediction model estimates SOH for the battery associated with a vehicle and the battery in storage condition.

8. A system for building a State of Health (SOH) prediction model for a battery, the system comprises: a memory; at least one processor coupled with the memory, wherein the processor is configured to: receive a dataset comprising gas emission data from at least one Quantum Dot (QD) sensor and performance data of at least one battery cell of the battery, wherein the gas emission data is indicative of concentration of at least one type of gas released from the at least one battery cell, and wherein the performance data comprises at least one battery parameter relating to performance of the at least one battery cell measured during the release of the at least one type of gas; process the dataset to derive at least one feature indicative of performance of the battery and to predict states of the concentration of at least one type of gas and the at least one parameter; and build the SOH prediction model to predict degradation of at least one component of the at least one battery cell based on the processed dataset and to compute a SOH of the battery cell in relation to the degradation of the at least one component, wherein the building includes training the SOH prediction model to correlate the concentration of the at least one type of gas and the at least one battery parameter based on the derived at least one feature and the predicted states.

9. The system as claimed in claim 8, wherein the at least one QD sensor are arranged at different locations in the battery, wherein the different locations comprise at least one of: current collector of at least one battery cell of the battery, and a cover of the battery.

10. A battery module comprises: at least one Quantum Dot (QD) sensor; a battery comprising at least one battery cells, wherein each of the at least one battery cell comprises at least one component comprising of: a Solid Electrolyte Interphase (SEI), an electrolyte, a dendrite, a separator, an anode and a cathode; andInternal Ref.: 2024P03445WO a battery management unit comprises a SOH prediction model to estimate a SOH of the battery, wherein the SOH prediction model is built using a dataset comprising gas emission data from at least one Quantum Dot (QD) sensor and performance data of at least one battery cell of the battery, wherein the gas emission data is indicative of concentration of at least one type of gas released from the at least one battery cell, and wherein the performance data comprises at least one battery parameter relating to performance of the at least one battery cell measured during the release of the at least one type of gas.

Images