System and method for detecting electrolyte and coolant leaks of lithium ion battery system

Abstract

A computer-implemented method includes monitoring, using a first gas sensor, a gas analyte level associated with a battery system and monitoring at least one variable of the battery system. The method includes determining whether there is a correlation between the monitored gas analyte level and the monitored at least one variable of the battery system. The method includes determining whether there is an electrolyte leak in the battery system based on a result of the determination of the correlation.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 370,555, filed on August 5, 2022, entitled “SYSTEMS AND METHODS FOR DETECTING ELECTROLYTE AND COOLANT LEAKS IN LITHIUM-ION BATTERY SYSTEMS,” the contents of which are incorporated herein by reference. Technical Field

[0003] The technology includes systems and methods for detecting electrolyte and coolant leaks in a battery system. Background Art

[0004] A battery leak occurs when the hermetic seal of a battery within a battery system (e.g., a battery energy storage system (BESS) or an electric vehicle (EV) battery pack) is compromised, causing the internal contents of the battery to be open to the atmosphere. This is sometimes referred to as "cold outgassing." As a result of the battery being open, carbonate solvents at high vapor pressure in the battery electrolyte will slowly leak out over time, causing the battery to dry out, which can affect the performance of the battery. If the solvent leaking from the battery is contained in a confined space (such as in an electric vehicle battery pack), the leaking contents of the battery may also result in a flammable condition. The flash points of these solvents are very low (e.g., 18-25 degrees Celsius, °C), which may create a flammable environment. In addition, the presence of a leak in the battery may also be a place where oxygen and moisture enter the battery, which may lead to failures such as thermal runaway. Systems and methods for detecting electrolyte leaks are needed. Summary of the invention

[0005] In one embodiment, a computer-implemented method includes monitoring a gas analyte level associated with a battery system and monitoring at least one variable of the battery system using a first gas sensor. The method includes determining whether there is a correlation between the monitored gas analyte level and the monitored at least one variable of the battery system. The method includes determining whether there is an electrolyte leak in the battery system based on the determination of the correlation.

[0006] In one embodiment, the monitoring system includes at least one gas sensor configured to monitor a gas analyte associated with a battery system and at least one sensor configured to monitor one or more variables of the battery system. The monitoring system includes a controller, the controller including a memory for storing machine-readable instructions and a processor for accessing the memory and executing the machine-readable instructions. The machine-readable instructions cause the processor to perform the following operations: monitor the gas analyte using the at least one gas sensor, monitor one or more variables of the battery system using the at least one sensor; determine the correlation between the monitored gas analyte level and the one or more variables; and determine whether the battery system has an electrolyte leak based on the correlation.

[0007] In one embodiment, a computer-implemented method includes modulating a change in a value of a sensor operating variable of each gas sensor configured to monitor a gas analyte associated with a battery system. The method includes using the gas sensor to monitor the gas analyte throughout a range of values ​​of the modulated sensor operating variable. The method includes obtaining a data matrix including sensor signals generated by the gas sensor that vary as the values ​​of the modulated sensor operating variable vary. The method includes distinguishing gas species of the gas analyte based on a comparison of features in the data matrix. The method also includes determining a condition of the battery system based on the distinction of the gas species.

[0008] In another embodiment, a monitoring system includes at least one gas sensor and a controller, wherein the at least one gas sensor is configured to monitor a gas analyte associated with a battery system. The controller includes a memory for storing machine-readable instructions and a processor for accessing the memory and executing the machine-readable instructions. The machine-readable instructions cause the processor to perform the following operations: modulate the value variation of a sensor operating variable of each of the at least one gas sensor; use the at least one gas sensor to monitor the gas analyte level in the entire value variation range of the modulated sensor operating variable to obtain a data matrix, wherein the data matrix includes a sensor signal generated by the at least one gas sensor that varies with the value variation of the modulated sensor operating variable; distinguish the gas type of the gas analyte based on the comparison results of each feature in the data matrix; and determine the condition of the battery system based on the differentiation result of the gas type.

[0009] Brief description of the attached figure

[0010] In the accompanying drawings, chemical formulas, chemical structures, and experimental data are given, which together with the detailed description provided below describe exemplary embodiments.

[0011] Figure 1A is a block diagram of an exemplary monitoring system.

[0012] Figure 1B is a block diagram of another exemplary monitoring system.

[0013] Figure 2 An exemplary method for monitoring gas analytes of a battery system based on data correlation is provided.

[0014] Figure 3A Exemplary correlations between monitored gas analyte levels and monitored variables of the battery system are shown.

[0015] Figure 3B Another exemplary correlation between monitored gas analyte levels and monitored variables of a battery system is shown.

[0016] Figure 4 An exemplary method of monitoring gas analytes of a battery system based on modulated temperature variations and monitored data is shown.

[0017] Figures 5A-5D Shown based on Figure 4 An exemplary method processes an exemplary analysis of a data matrix to distinguish gas species.

[0018] Figure 6 An exemplary machine learning (ML) classification design process is shown.

[0019] Figure 7 is an exemplary flow chart illustrating a process of pre-training a gas sensor using an ML algorithm based on known gas analytes.

[0020] Figure 8 An exemplary output summary of gas analytes monitored using the systems and methods disclosed herein is shown.

[0021] Fig. 9 Another exemplary output summary of gas analytes monitored using the systems and methods disclosed herein is shown. DETAILED DESCRIPTION

[0022] The present disclosure generally relates to systems and methods for detecting electrolyte and / or coolant leaks in batteries. Over time, batteries may gradually degrade, which may lead to reduced capacity, cycle life, and safety issues. Degraded batteries may release gases. These gases may come from electrolyte leaks, coolant leaks, or battery venting / outgassing events. Electrolyte leaks can cause capacity loss issues, quality issues, potential compliance issues, and potential safety issues. Care should be taken to regularly inspect batteries for any signs of degradation, and repairs should be performed, especially in the case of electrolyte leaks and insulation failures (e.g., IEC 62485-5).

[0023] In the event of an electrolyte leak, the gases released are electrolyte solvent vapors, which are the primary gases in battery exhaust / battery outgassing events. In a battery exhaust / battery outgassing event, trace amounts of hydrogen (H2), carbon monoxide (CO), and carbon dioxide (CO2) may be released, allowing these gases to be detected by an H2 detector and / or CO detector in close proximity to the outgassing cell. However, in the event of an electrolyte leak, theoretically only solvent vapor is present, which cannot be detected using an H2 sensor and / or CO sensor. In particular, electrolyte leakage may occur very slowly, while exhaust / battery outgassing occurs quickly. Unlike exhaust / battery outgassing, electrolyte leakage detection is difficult because it may not occur at different time points and there may be no ability to establish a baseline or reference (for example, if the battery leaks prior to monitoring, it is likely that no changes will be detected).

[0024] The systems and methods described herein can detect leakage of electrolyte and / or coolant. In addition, the systems and methods described herein can be configured to monitor electrolyte leakage and / or coolant leakage in any type of battery (such as lithium-ion batteries and lead-acid batteries).

[0025] The term "gas analyte" is used herein to refer to the gas released by the battery. The gas analyte may include the emitted gas (i.e., "released gas" and "gas analyte"), which includes electrolyte gas (such as volatile electrolyte solvents, volatile components of the electrolyte mixture of the battery, etc.) and coolant (e.g., ethylene glycol / water mixture). The volatile electrolyte or emission gas analyte types may include one or more of the following flammable or toxic gases: lithium ion battery emission gas, dimethyl carbonate, diethyl carbonate, ethyl methyl carbonate, ethylene carbonate, propylene carbonate, vinylene carbonate, carbon dioxide, carbon monoxide, hydrocarbons, methane, ethane, ethylene, propylene, propane, benzene, toluene, hydrogen, oxygen, nitrogen oxides, volatile organic compounds, toxic gases, hydrogen chloride, hydrogen fluoride, hydrogen sulfide, sulfur oxides, ammonia and chlorine, etc. The gas sensor used herein is capable of detecting the gas analyte (if the gas analyte is present in the atmosphere).

[0026] In addition, the systems and methods described herein can be configured with multiple battery housings. Therefore, the systems and methods described herein can be used to monitor gaseous analytes released by one or more batteries located within a battery housing. The term "battery housing" as used herein refers to any housing that can at least partially encapsulate one or more batteries. In one example, the battery housing may include ventilated and non-ventilated battery housings. The ventilated battery housing may include a ventilation system that may include an air inlet and an exhaust port. In another example, the battery housing may include a battery cabinet. In another example, the battery housing may include a battery housing for a battery system for a vehicle. In another example, the battery housing may include a battery shipping container.

[0027] In addition, the term "processor" as used herein may refer to any device capable of executing machine-readable instructions, such as a computer, a controller, an integrated circuit (IC), a microchip, or any other device capable of implementing logic. The term "memory" as used herein may refer to non-transitory computer storage media, such as volatile memory (e.g., random access memory), non-volatile memory (e.g., hard drive, solid state drive, flash memory, etc.), or a combination thereof.

[0028] Figure 1A An example of a system 100 for monitoring electrolyte and / or coolant leakage is shown. The system 100 includes at least one gas sensor 102 configured to monitor a gas analyte 104 released from a battery 106. The at least one gas sensor 102 may include any type of gas sensor, such as a chemiresistor sensor, an electrochemical sensor, a semiconductor metal oxide sensor, a catalytic sensor, a thermal conductivity sensor, a metal oxide semiconductor, a potentiometric sensor, an optical sensor, an infrared (IR) sensor, an amperometric sensor, a micro hot plate sensor, etc. The at least one gas sensor 102 may be disposed near the battery 106. The at least one gas sensor 102 is configured to generate a (real-time) sensor signal 110 to convey information of the gas analyte 104 of the monitored battery 106.

[0029] The battery 106 may be any type of battery or battery system including an electrolyte. The battery 106 may also include a cooling system having a coolant for cooling the battery 106. The battery 106 may be a lithium-ion battery, a lead-acid battery, or any other type of rechargeable or non-rechargeable battery. The battery 106 is fully or partially surrounded by a housing 108.

[0030] Unlike most other systems, the deployed gas sensor 102 does not require the use of a separate reference sensor in the system 100 to calculate a moving average from the (real-time) sensor signal 110 to detect the gas analyte 104 released by the cell 106 .

[0031] The system 100 also includes at least one sensor 112 to monitor at least one variable of the battery 106. The at least one variable of the battery 106 includes one or more of a temperature of the battery 106 (e.g., battery temperature or overall temperature of the battery system), a current of the battery 106 (e.g., charge / discharge current), a relative humidity in an environment in which the battery 106 is located, an airflow (e.g., flow rate) of ambient air around the battery 106, or a combination thereof.

[0032] At least one sensor 112 includes a temperature sensor, a humidity sensor, an air flow sensor, and a current sensor or a resistivity sensor. At least one sensor 112 may be disposed on or near the battery 106, inside the housing 108, and / or outside the housing 108. In some embodiments, at least one gas sensor 102 and / or at least one sensor 112 may be a sensor of a battery management system (BMS) of the battery 106 or may include a sensor of a battery management system (BMS) of the battery 106. At least one sensor 112 is configured to generate a (real-time) sensor signal 114 to transmit at least one variable of the monitored battery 106.

[0033] The system 100 includes a controller 116 configured to operate and coordinate the operation of various components in the system 100. The controller 116 may include any suitable processor 118 (e.g., a microprocessor, a MOSFET, an IGBT, etc.) and a memory 120. The controller 116 may be programmed to execute certain programs or predetermined programs. In one example, an analysis program or algorithm 122 is stored in the memory 120 for processing / analyzing the sensor signals 110 and 114 to determine in real time any released gas analyte 104 as an event including one or more of the following events: an electrolyte leak event, a coolant leak event, a water ingress event, a contaminated metal oxide sensor event, an exhaust event (OGE), a thermal runaway event (TRE), and an interfering gas release event (i.e., non-OGE). The sensor contamination event discussed herein refers to the sensor being contaminated by chemicals that inhibit the sensor from properly performing its function, and the sensor contamination event is not limited to a contaminated metal oxide sensor.

[0034] In another example, the analysis program or algorithm 122 may include a machine learning (ML) or deep learning (DL) algorithm (program code) executed by the processor 118 to enable the at least one gas sensor 102 to detect any released gas analyte 104 in real time and classify it into events including one or more of the following: electrolyte leak, coolant leak, water ingress, contaminated metal oxide sensor, outgassing event (OGE), thermal runaway event (TRE), and interfering gas release event (i.e., non-OGE).

[0035] The controller 116 may include any suitable wired or wireless communication device / mechanism to output a signal or alarm 124 based on the analysis. The output signal or alarm 124 may be a warning alarm or may be a logic signal that is sent for warning or for display on a screen to take preventive measures, which output signal or alarm indicates the condition of the battery 106 (e.g., electrolyte leak, coolant leak, water ingress, contaminated metal oxide sensor, OGE, TRE, non-OGE, etc.).

[0036] The at least one gas sensor 102 may be pre-trained and the ML or DL ​​algorithm 122 may be stored in the memory 120 as a candidate model to distinguish the sensor signal 110 detected by the at least one gas sensor 102 without requiring any reference gas sensor and without requiring retraining of the at least one gas sensor 102 once the at least one gas sensor 102 is deployed in the field.

[0037] The machine learning and training of the algorithm 122 steps may be performed a priori in the factory during the manufacturing process, or offline at any time prior to physical commissioning or installation of the at least one gas sensor 102 in the system 100. Once the at least one gas sensor 102 is commissioned in the system 100, no real-time adaptation is required.

[0038] Alternatively, however, in another option, the ML or DL ​​algorithm 122 may be retrained or updated by the at least one gas sensor 102 to learn new exposure to other gas analytes that have not been pre-retrained or listed in the database. The goal of such pre-training using the ML or DL ​​algorithm is not only to detect OGE and coolant, but also to be able to identify other gas sources detected by the at least one gas sensor 102, thereby eliminating the need for a reference sensor.

[0039] exist Figure 1AIn the illustrated system 100, the algorithm 122 may be implemented in an embedded microcontroller of at least one gas sensor. The at least one gas sensor 102 and the controller 116 may be an integrated chip 126, such as an ASIC semiconductor chip. Alternatively, the at least one gas sensor 102 and the controller 116 may each be a discrete component electrically connected by a wiring harness or mounted on a printed circuit board (PCB).

[0040] In such Figure 1B In another embodiment shown, the controller 116 may be a separate computer (eg, a computer of a BMS of the battery 106 ), and the sensor signal 110 may be sent to the controller 116 via wired or wireless communication.

[0041] Figure 2 An exemplary computer-implemented method 200 performed by the system 100 is shown for determining whether an electrolyte leak exists in a battery system. The method 200 includes monitoring a gas analyte level using a first gas sensor (step 202). At least one gas sensor 102 (e.g., any suitable gas sensor) is configured to detect a gas analyte 104. The at least one gas sensor 102 is positioned on or in close proximity to a battery 106 so that the at least one gas sensor continuously monitors / measures the gas present and detects any gas analyte 104 released by the battery 106. The at least one gas sensor 102 generates a sensor signal 110 corresponding to the amount of gas detected.

[0042] Method 200 includes monitoring at least one variable of the battery system (step 204). At least one sensor 112 (eg, any suitable sensor) is configured to monitor at least one variable of battery 106, such as temperature, relative humidity, charge / discharge current, airflow around battery 106, etc.

[0043] The method 200 includes determining whether a correlation exists between the monitored gas analyte level and at least one monitored variable of the battery system 106 (step 206 ), and determining a condition of the battery system and / or a condition of the first gas sensor based on the correlation (step 208 ).

[0044] In step 208 , determining the condition of the battery system may include determining whether there is electrolyte leakage in the battery system based on the correlation determination result. The sensor signals 110 and 114 , data or information are processed or analyzed by the controller 116 to determine whether there is a correlation between the sensor signals 110 and 114 . Figure 3AAn example chart 300 is shown in which a reference data series trend 302 and a test data series trend 304 are plotted on an X-axis corresponding to at least one monitored variable 306 (e.g., a variable of the battery 106 or a variable corresponding to an environmental condition of the battery 106) and a Y-axis corresponding to a monitored gas analyte level 308.

[0045] Reference data series trend 302 shows the expected correlation. When at least one of the monitored variables changes (e.g., increases or decreases), the monitored gas analyte level remains relatively constant / unchanged. In this case, the monitored gas analyte level has no correlation with at least one variable of the battery 106 (e.g., temperature, relative humidity, charge / discharge current, airflow around the battery 106, etc.), indicating that there is no electrolyte leakage.

[0046] In the test data series trend 304, the monitored gas analyte level increases relatively linearly as the at least one monitored variable increases, indicating that a relatively linear correlation exists between the two variables. In this case, the correlation of the monitored gas analyte level with at least one variable of the battery 106 (e.g., temperature, relative humidity, charge / discharge current, airflow around the battery 106, etc.) deviates from the correlation of the reference data series trend 302. This deviation in correlation indicates the presence of an electrolyte leak.

[0047] The linear correlation in the graph 300 is shown only as a non-limiting example. A non-linear correlation may exist between the monitored gas analyte level and the at least one variable of the monitored battery 106. Any correlation (e.g., linear, non-linear, exponential, logarithmic, cubic, etc.) between the monitored gas analyte level and the at least one variable of the monitored battery 106 that deviates from an expected correlation (e.g., reference data series trend 302) indicates electrolyte leakage.

[0048] The controller 116 or the algorithm 122 may be configured to determine the presence of an electrolyte leak based on the correlation coefficient. For example, the controller 116 or the algorithm 122 may be configured to determine the presence of an electrolyte leak if the absolute value of the correlation coefficient of the test data sequence trend 304 is greater than a predetermined value. As another example, the controller 116 or the algorithm 122 may be configured to determine the presence of an electrolyte leak if the difference between the correlation coefficient of the reference data sequence trend 302 and the correlation coefficient of the test data sequence trend 304 is greater than a predetermined threshold.

[0049] In step 208, determining the condition of the first gas sensor may include determining whether the first gas sensor is contaminated based on the correlation determination result. The sensor signals 110 and 114, data, or information are processed or analyzed by the controller 116 to determine whether there is a correlation between the sensor signals 110 and 114 and to determine the degree of data distribution. Figure 3B An exemplary graph 310 is shown in which a reference data series trend 312 and a test data series trend 314 are plotted on an X-axis corresponding to at least one monitored variable 306 (e.g., a variable of the battery 106 or a variable corresponding to an environmental condition of the battery 106) and on a Y-axis corresponding to a monitored gas analyte level 308.

[0050] The reference data series trend 312 shows an expected correlation with an expected data distribution, where there is no correlation between the two variables. The data distribution shows expected natural fluctuations because the at least one gas sensor 102 is expected to respond to natural fluctuations in the gas in the background gas (e.g., ambient gas) in the battery housing 108. This results in a relatively large data distribution around the expected correlation 312 over time. The reference data series trend 312 indicates that the gas sensor (e.g., the at least one gas sensor 102) is not contaminated.

[0051] In the test data series trend 314, the monitored gas analyte level remains relatively constant / unchanged as the at least one monitored variable changes. The degree of data distribution / dispersion is lower than the degree of data distribution / dispersion of the reference data series trend 312, indicating that the gas sensor (e.g., at least one gas sensor 102) is contaminated. If the gas sensor is contaminated, the response to background gas fluctuations will be lower, resulting in less distribution of data from the gas sensor.

[0052] The controller 116 or the algorithm 122 may be configured to determine whether a gas sensor (e.g., at least one gas sensor 102) is contaminated based on the degree of data distribution or the fluctuation index. In one example, the controller 116 or the algorithm 122 may be configured to determine that the gas sensor is contaminated if the data distribution or fluctuation of the test data sequence trend 314 is below a predetermined value or threshold. As another example, the controller 116 or the algorithm 122 may be configured to determine that the gas sensor is contaminated if the difference between the data distribution or fluctuation of the test data sequence trend 314 and the data distribution or fluctuation of the reference data sequence trend 312 is greater than a predetermined value or threshold.

[0053] In one non-limiting example, Figure 3A and Figure 3B At least one variable monitored in is the temperature of the battery 106 .

[0054] In addition to detecting electrolyte leakage based on correlation between monitored gas analyte levels and monitored variables of the battery (e.g., correlation methods for determining electrolyte leakage), the system 100 is also configured to distinguish leaking electrolyte solvent vapor in the presence of other gases. At least one gas sensor 102 may be sensitive to other gases that may come from other volatile organic compounds (VOCs), such as adhesives or vent gasket materials in battery modules. The system 100 can reliably distinguish when the sensor response is caused by a small amount of electrolyte vapor (true positive) or caused by other VOCs (false positive).

[0055] In an embodiment where the system 100 is configured to distinguish between gas species, at least one gas sensor 102 may include a gas sensor with an induced or modulated gas sensor operating variable. For example, at least one gas sensor 102 may include multiple micro hot plate sensors (e.g., a second gas sensor), and in order to distinguish between true positive scenarios and false positive scenarios, the operating temperature of the second gas sensor is modulated. Changes in temperature cause different gas species to react differently with the sensor electrodes. This produces different features on the raw gas sensor signal, which can be distinguished to distinguish gases from positive sources (e.g., battery solvent vapor) or false positive sources (e.g., exhaust adhesive). In addition, based on this method, the system 100 can detect and classify battery coolant leaks. The coolant used in the liquid cooling module of the battery typically contains an ethylene glycol-water mixture (e.g., a 50:50 ethylene glycol / water mixture used for engine coolant in the radiator of an internal combustion engine vehicle). The system 100 is also configured to detect leaks of battery coolant, thereby enabling monitoring of unique failure modes in the battery.

[0056] As an example, the system 100 is configured to classify the response of the second gas sensor (e.g., a micro hot plate sensor) into carbonate and non-carbonate. The solvent in the lithium-ion battery is a carbonate-based solvent, so the presence of a carbonate-based solvent indicates a cell leak (e.g., electrolyte leak) of the battery, while the presence of hydrogen may indicate electrolysis inside the battery module due to a coolant leak or water ingress.

[0057] Figure 4An exemplary computer-implemented method 400 performed by the system 100 is shown for distinguishing gas species and determining a condition of the battery 106 based on the results of distinguishing the gas species. The method 400 includes modulating the value change of the gas sensor operating variable of each second gas sensor (step 402). The gas sensor operating variables discussed herein refer to any variable that controls / operates the gas sensor, including but not limited to temperature and biases applied to the sensor element, such as power bias, voltage bias, current bias, or polarity bias. For the purpose of discussion, temperature modulation is described as an example below; however, the method 400 can be performed based on the modulation of any one or more of the gas sensor operating variables.

[0058] Step 402 includes modulating the electrode temperature of each second gas sensor (e.g., a micro hot plate sensor) with any suitable waveform (e.g., any periodic temperature variation over time). The modulation of the value variation of the gas sensor temperature can be achieved via any suitable modulation of the gas sensor operating variable (e.g., hot plate temperature, bias applied to the sensor element (such as power, voltage, current, polarity, etc.)).

[0059] In one example, the electrode temperature of each second gas sensor is modulated between an initial temperature (e.g., a temperature at 0% capacity or a minimum temperature) and a final temperature (e.g., a temperature at 100% capacity or a maximum temperature) at a predetermined change (e.g., increase or decrease) level and a predetermined time interval. Each predetermined change can be a predetermined temperature change such as 5 degrees Celsius (°C), 10°C, 15°C, 20°C, etc. or a predetermined gas sensor operating variable change (e.g., power or voltage change) such as 5%, 10%, 15%, etc. In one example, step 402 includes modulating the plurality of electrodes of the second gas sensor 102 from a minimum temperature of 100°C to a maximum temperature of 400°C at a rate of 10°C per second. Step 402 includes maintaining the plurality of electrodes of the second gas sensor 102 at each temperature for an effective time period. In step 402, the electrodes of the second gas sensor 102 can have the same or different modulated temperature value changes.

[0060] The method 400 includes monitoring the gas analyte over the entire value range of the modulated gas sensor operating variable (e.g., the entire value range of the temperature) (step 404). The method 400 includes: obtaining a data matrix including the monitored sensor data, the monitored sensor data changes as the value of the modulated gas sensor operating variable changes (e.g., the value of the temperature); and distinguishing the gas type of the monitored gas analyte based on the comparison results of various features (step 406). The processor 118 receives the sensor signal 110 (e.g., impedance) and temperature data from the second gas sensor 102. The sensor signal 110 (e.g., impedance data) collected at each sensor temperature variable step (e.g., impedance on all electrodes at each temperature) is a "feature" that can be used for pattern recognition / classification. Any suitable algorithm or analysis method 122 stored in the memory 120 can be used to perform step 406.

[0061] In one example, a machine learning or deep learning method may be applied to perform step 406. Specifically, based on machine learning, step 406 includes obtaining features (step 408) based on the monitored gas analyte data and the value changes of the modulated gas sensor operating variable (e.g., the value changes of temperature). The number of features depends on the collected data set. In the case where the monitored gas analyte data is collected from three electrodes at 5°C intervals under the value changes of the temperature rising from 100°C to 400°C, the total number of features is 183, that is, the average impedance at each temperature during a one-minute cycle of the temperature ramp from 100°C to 400°C (61 temperature intervals). Therefore, method 400 may optionally include reducing the number of features (step 410). Principal component analysis (PCA) may be applied to reduce the number of features. It is not necessary to perform this step to fit and evaluate the model, but this step is an option to reduce the size. In addition, additional environmental variables of the battery 106, such as information measured by at least one sensor 112 (e.g., temperature, relative humidity, battery current, airflow, etc.). Alternatively, in step 406, unique signatures may be generated via any suitable modulation of gas sensor operating variables (e.g., hot plate temperature, bias applied to the sensor element, power, voltage, current, polarity, etc.), which may be obtained based on the monitored gas analyte data and the modulated bias power, bias voltage, bias current, and / or bias polarity.

[0062] The method 400 includes feeding the features into a model (e.g., a machine learning model or ML model) and outputting a classification to determine the gas species of the gas analyte (step 412). Candidate ML models are stored in the memory 120 (e.g., the algorithm 122) to analyze the features. The candidate ML model can be any supervised machine learning model, such as a k-nearest neighbor (kNN) technique.

[0063] The method 400 includes determining a condition of the battery based on the determined types of the gas analytes 104 (step 414). Specifically, the system 100 and method 400 can sort the gas analytes (e.g., into carbonates and non-carbonates) and / or distinguish or classify the gas analytes 104 into specific types to determine the condition of the battery 106. For example, detection of hydrogen gas indicates electrolysis occurring within the battery 106 due to a coolant leak or water ingress, and detection of one or more electrolyte vapors indicates an electrolyte leak.

[0064] At least one gas sensor 102 and candidate models are pre-trained prior to deployment. Steps 402 to 410 may be completed offline to tune / fit the model, and then steps 402 to 414 may be performed in real time to perform classification and determination.

[0065] Figure 5A An example of a graph of impedance versus temperature for a system 100 having three electrodes (eg, three second gas sensors) is shown when the temperature variation is modulated to rise from 100°C to 400°C. Figure 5B For each of the three electrodes Figure 5A Based on PCA, the number of features is reduced, such as Figure 5C When these features are fed into the ML model, after one full sensor modulation cycle, the monitored gas analytes are distinguishable between typical VOC interferents (false positive detections) and actual electrolyte leaks (true positive detections), as shown in Figure 2. Figure 5D As shown. Figure 5D The classification shown may be performed after each modulation period, after each sampling period, or after both each modulation period and each sampling period.

[0066] Figure 6 and Figure 7An example of an ML classification design process or how an ML algorithm can be developed is depicted. A pre-training (supervised learning) method can combine many signal features (from at least one gas sensor 102 and at least one sensor 112) using both heuristic-based impedance information and physics-based impedance information in a data pre-processing step in the algorithm development phase. The pre-training (supervised learning) method can also include environmental measurements included in the sensor group product, such as temperature, relative humidity, airflow, charge / discharge current, etc. For example, signal features can include moving averages, Bollinger bands, minimum electrode impedance, maximum impedance change rate, maximum impedance recovery rate for each electrode, principal component analysis (PCA), and linear discriminant analysis. In addition, pre-training (supervised learning) of gas sensors using other techniques to distinguish OGE or TRE from non-OGE can also include, but are not limited to, classification techniques, such as: support vector machines, discriminant analysis or nearest neighbor algorithms, basic statistics of time series value distributions (e.g., location, diffusion, Gaussianity, outlier properties), linear correlations (e.g., autocorrelation, features of power spectra), stationarity (e.g., sliding window measurements, prediction errors), information theory, and entropy / complexity, etc.

[0067] exist Figure 7 , an exemplary flow chart illustrates a process 700 of pre-training at least one gas sensor 102 based on a plurality of known gas analytes using an ML algorithm. For example, in step 702, for each known gas analyte, a raw sensor signal 110 such as resistance and capacitance may be generated from multiple electrodes of the at least one gas sensor 102, and in step 704, the raw sensor signal is sent to the processor 118 for feature extraction (e.g., impedance or transfer function changes over time). The feature extraction step 704 may include a time-frequency transform (such as a discrete cosine transform DCT or a discrete Fourier transform DFT) to transform the time domain analog signal into a frequency domain signal. In step 706, the extracted features may be organized accordingly. In step 708, the ML algorithm is applied to the organized data to establish a candidate model 710 (such as a multi-dimensional decision boundary construction). By repeating steps 702 to 710 for the remainder of the plurality of known gas analytes, the candidate model 710 (see step 712) may be updated to build a database or construct a composite decision boundary graph to complete the ML algorithm 122 training, which is stored in the memory 120 for execution by the processor 118. More specifically, step 708 may be implemented by repeating these training steps. Each training step may include sequentially performing the following operations: convolution, rectified linear unit (ReLU), and pooling operations. When the deployed model 714 is operated in conjunction with a multi-electrode gas sensor to perform gas analyte classification, the deployed model will be a field-ready ML algorithm.

[0068] Alternatively, these desired classifications can be achieved through deep learning algorithms utilizing pre-trained convolutional neural networks (e.g., convolutional neural networks CNN and long short-term memory (LSTM)) and automatic signal feature extraction.

[0069] The methods and analyses disclosed herein enable differentiation of gas analytes 104 based on ML techniques. Figure 8 and Fig. 9 Two examples of output summaries based on the systems and methods disclosed herein are shown. Figure 8 An example is shown in which the methods and systems disclosed herein are able to accurately differentiate the gas analytes 104 into carbonates and non-carbonates without false positives and false negatives. Fig. 9 is an exemplary positive prediction rate matrix for various gas species.

[0070] In addition, the method 400 can be modified to determine the condition of the gas sensor, such as a contaminated gas sensor. For example, the method may include modulating the value variation of the gas sensor temperature of each gas sensor, monitoring the gas analyte throughout the value variation range of the modulated gas sensor temperature, and obtaining a data matrix including the monitored gas sensor data that varies with the value variation of the modulated gas sensor temperature, and distinguishing the gas sensor response based on the comparison results of various features. These features are fed into the model and output classification to determine the unique gas sensor response characteristics. Then, the method determines whether the gas sensor is contaminated based on the determined gas sensor response. The method may include using the contaminated gas sensor to train the model using a data set collected from a wide range of gas sensor environments to obtain data of the contaminated gas sensor. When the gas sensor is contaminated, the gas sensor characteristic responds uniquely; therefore, the model can be trained to distinguish the contaminated gas sensor.

[0071] Methods 200 and 400 disclosed herein may be used alone or in combination, for example, method 400 may be used to confirm the results of method 200 and further confirm the type of gas detected. In some embodiments, method 200 and / or method 400 may include a step of reporting or warning the determined battery condition (e.g., electrolyte leakage, coolant leakage, etc.) and / or the type of gas detected. For example, after determination, method 200 or 400 may include a step of sending an early warning, which includes a logic signal output, an audible alarm, a visual alarm, a fire extinguishing, and / or a communication with other systems and users.

[0072] To the extent that the term "includes" or "including" is used in the specification or claims, the term "includes" or "including" is intended to be inclusive in a manner similar to the term "comprising," which is interpreted as "comprising" when used as a transitional word in the claims. As used in the specification and claims, the singular forms "a," "an," and "the" include the plural. Further, to the extent that the term "or" is used (e.g., A or B), the term is intended to mean "A or B or both." Finally, when the term "about" or "approximately" is used in conjunction with a number, the term is intended to include within ±5%, within ±4%, within ±3%, within ±2%, within ±1%, or within ±0.5% of that number.

[0073] As mentioned above, although the present application has been described by the description of the embodiments, and although the embodiments have been described in considerable detail, the present application is not intended to constrain the scope of the appended claims to such details or to limit such details in any way. Those skilled in the art who benefit from the present application will easily understand the additional advantages and modifications. Therefore, in the broader aspects of the present application, the present application is not limited to the specific details and illustrative examples shown. These details and examples may be deviated from without departing from the spirit or scope of the general inventive concept.

Claims

1. A computer-implemented method, the method comprising: monitoring a gas analyte level associated with the battery system using a first gas sensor; monitoring at least one variable of the battery system; determining whether a correlation exists between the monitored gas analyte level and at least one monitored variable of the battery system, and Whether there is electrolyte leakage in the battery system is determined based on the determination result of the correlation.

2. The computer-implemented method of claim 1 , wherein: The correlation includes an increasing trend in the monitored gas analyte level as the at least one monitored variable of the battery system increases.

3. The computer-implemented method of claim 1 , wherein: The battery system is a lithium-ion battery system.

4. The computer-implemented method of claim 1 , wherein: The monitored gas analytes include electrolyte gases and interfering gases that are not exhaust events (non-OGE).

5. The computer-implemented method of claim 4, wherein: The interfering gas of the non-exhaust event includes hydrogen and / or coolant.

6. The computer-implemented method of claim 1 , wherein: The at least one variable of the battery system monitored includes a temperature of the battery system.

7. The computer-implemented method of claim 1 , wherein: The at least one variable of the battery system monitored includes a current of the battery system.

8. The computer-implemented method of claim 1 , wherein: The at least one variable of the battery system monitored includes relative humidity surrounding the battery system.

9. The computer-implemented method of claim 1 , wherein: The at least one variable of the battery system monitored includes air flow around the battery system.

10. The computer-implemented method of claim 1 , comprising: modulating a change in a value of a sensor operating variable of each second gas sensor configured to monitor a gas analyte associated with the battery system; monitoring the gas analyte using the second gas sensor over a range of values ​​of the modulated sensor operating variable; obtaining a data matrix including sensor signals generated by the second gas sensor that vary with the value of the modulated sensor operating variable; distinguishing gas types of the gas analyte based on the comparison results of each feature in the data matrix; and The condition of the battery system is determined based on the result of distinguishing the types of gas.

11. The computer-implemented method of claim 10, comprising pre-training the second gas sensor based on a machine learning (ML) algorithm prior to initial field deployment of the second gas sensor.

12. The computer-implemented method of claim 10, wherein: The condition of the battery system includes one or more of electrolyte leakage, coolant leakage, battery gassing, thermal runaway, water ingress, and gas discharge.

13. The computer implemented method of claim 10, comprising identifying a contaminated one of the second gas sensors.

14. The computer implemented method of claim 10, the sensor operating variables comprising temperature, power, voltage, polarity and / or current.

15. A monitoring system, comprising: at least one gas sensor configured to monitor a gas analyte associated with the battery system; at least one sensor configured to monitor one or more variables of the battery system; and A controller, the controller comprising: a memory for storing machine-readable instructions; and A processor, the processor is configured to access the memory and execute the machine-readable instructions, the machine-readable instructions causing the processor to perform the following operations: monitoring the gas analyte using the at least one gas sensor; monitoring one or more variables of the battery system using the at least one sensor; determining a correlation between the monitored gas analyte level and the one or more variables; and It is determined whether there is an electrolyte leak in the battery system based on the correlation.

16. The monitoring system according to claim 15, wherein: The battery system is a lithium-ion battery system.

17. The monitoring system according to claim 15, wherein: The at least one sensor includes one or more of a temperature sensor, a relative humidity sensor, a current sensor, and an air flow sensor.

18. The monitoring system according to claim 15, wherein: The machine-readable instructions cause the processor to: modulating a change in a value of a sensor operating variable of the at least one gas sensor; monitoring the gas analyte level using the at least one gas sensor over a range of values ​​of the modulated sensor operating variable, obtaining a data matrix comprising sensor signals generated by the at least one gas sensor as a function of the value of the modulated sensor operating variable; distinguishing gas types of the gas analyte based on the comparison results of each feature in the data matrix; and The condition of the battery system is determined based on the result of distinguishing the types of gas.

19. The monitoring system according to claim 18, wherein: The machine-readable instructions cause the processor to determine whether there is a coolant leak in the battery system based on a result of distinguishing the types of gas.

20. The monitoring system of claim 18, wherein: The machine-readable instructions cause the processor to determine whether water has entered the battery system based on a result of distinguishing the types of gas.

21. The monitoring system of claim 18, wherein: The machine readable instructions cause the processor to determine whether the at least one gas sensor includes a contaminated sensor.

22. The monitoring system of claim 18, wherein: The sensor operating variables include temperature, power, voltage, polarity and / or current.

23. A computer-implemented method, the method comprising: modulating a change in a value of a sensor operating variable of each gas sensor configured to monitor a gas analyte associated with the battery system; monitoring the gas analyte using the gas sensor over a range of values ​​of the modulated sensor operating variable; obtaining a data matrix comprising sensor signals generated by the gas sensor that vary as a function of a value of the modulated sensor operating variable; distinguishing gas types of the gas analyte based on the comparison results of each feature in the data matrix; and The condition of the battery system is determined based on the result of distinguishing the types of gas.

24. The computer-implemented method of claim 23, comprising pre-training the gas sensor based on a machine learning (ML) algorithm prior to initial field deployment of the gas sensor.

25. The computer-implemented method of claim 23, wherein: The condition of the battery system includes one or more of electrolyte leakage, coolant leakage, battery gassing, thermal runaway, water ingress, and gas discharge.

26. The computer-implemented method of claim 23, the method comprising: distinguishing gas sensor responses based on comparison results of features in the data matrix; and A contaminated sensor among the gas sensors is identified based on the differentiation result of the gas sensor responses.

27. The computer-implemented method of claim 23, wherein the sensor operating variables include temperature, power, voltage, polarity, and / or current.

28. A monitoring system, comprising: at least one gas sensor configured to monitor a gas analyte associated with the battery system; and A controller, the controller comprising: a memory for storing machine-readable instructions; and A processor, the processor being configured to access the memory and execute the machine-readable instructions, the machine-readable instructions causing the processor to perform the following operations: modulating a change in a value of a sensor operating variable of each of the at least one gas sensor; monitoring a gas analyte level using the at least one gas sensor over a range of values ​​of the modulated sensor operating variable; obtaining a data matrix comprising sensor signals generated by the at least one gas sensor as a function of the value of the modulated sensor operating variable; Distinguishing the gaseous analyte into gaseous species based on the comparison results of the features in the data matrix; and The condition of the battery system is determined based on the result of distinguishing the types of gas.

29. The monitoring system of claim 28, wherein: The machine-readable instructions cause the processor to determine whether there is a coolant leak in the battery system based on a result of distinguishing the types of gas.

30. The monitoring system of claim 28, wherein: The machine-readable instructions cause the processor to determine whether water has entered the battery system based on a result of distinguishing the types of gas.

31. The monitoring system of claim 28, wherein: The machine-readable instructions cause the processor to differentiate gas sensor responses based on the comparison of features in the data matrix; and identify a contaminated sensor of the at least one gas sensor based on the differentiation of the gas sensor responses.

32. The monitoring system of claim 28, wherein: The sensor operating variables include temperature, power, voltage, polarity and / or current.

Images