Health state assessment method, apparatus and device, and storage medium

Abstract

The invention discloses a state of health assessment method, device and equipment and a storage medium. The method comprises the following steps: acquiring battery data measured by a target sensor within a preset time period; performing derivation processing on the battery data to obtain a target derivative group; the health state of the battery is evaluated based on the target derivative group, a target evaluation result is obtained, and the target evaluation result comprises early warning of the health state of the battery or no early warning of the health state of the battery. According to the scheme, the accuracy of the health state evaluation of the battery can be improved.

Description

Technical Field

[0001] The present application relates to the field of batteries, and in particular to a health status assessment method, device, equipment and storage medium. Background Art

[0002] The internal information of lithium batteries is of great value in thermal runaway warning and life assessment, especially the internal gas of lithium batteries (CO2, CO, H2, CH4, etc.) is the product of trace side reactions inside the battery. By detecting the concentration of gas in the battery, it is possible to determine whether the lithium battery has thermal runaway. However, the gas concentration measured by the gas sensor is inevitably biased, resulting in inaccurate judgment results. Summary of the invention

[0003] The present application at least provides a health status assessment method, apparatus, device and storage medium.

[0004] The present application provides a health status assessment method, including: obtaining battery data measured by a target sensor within a preset time period; performing derivative processing on the battery data to obtain a target derivative array; evaluating the health status of the battery based on the target derivative array to obtain a target assessment result, the target assessment result including issuing a warning for the health status of the battery or not issuing a warning for the health status of the battery.

[0005] In the above scheme, the battery data measured by the target sensor within a preset time period is obtained, and then the target derivative array is obtained by deriving the data of each battery. The data after derivation can reduce the error caused by sensor zero drift and other reasons compared to the data before derivation. Then, the health status of the battery is evaluated based on the target derivative array. Compared with directly using the battery data to evaluate the health status of the battery, the evaluation result obtained by this scheme is more accurate.

[0006] In some embodiments, the battery data is derivatized to obtain a target derivative array, including: performing a first derivation on the battery data to obtain an initial derivative array; and performing a second derivation on the data in the initial derivative array to obtain a target derivative array.

[0007] In the above scheme, the problem of sensor zero drift can be solved by taking a derivative of the sensor data once, and the data error caused by sensor aging and the like can be further reduced by taking a second derivative of the data in the initial derivative array.

[0008] In some embodiments, the health status of the battery is evaluated based on the target derivative array to obtain a target evaluation result, including: extracting features of the target derivative array, the features including one or more of the maximum value, minimum value, average value, standard deviation, period, and difference between the maximum value and the minimum value; sending the features of the target derivative array to a health status assessment model; and receiving the target assessment result obtained by the health status assessment model by evaluating the battery based on the features of the target derivative array.

[0009] In the above scheme, by extracting the features from the target to the array and then using the health status assessment model to perform assessment based on the extracted features, the assessment error rate caused by subjective factors of human assessment can be reduced, the objectivity of the assessment can be improved, and the assessment efficiency can also be improved.

[0010] In some embodiments, sending the characteristics of the target derivative array to a health status assessment model includes: determining a target operating state of the battery; determining a target health status assessment model corresponding to the target operating state; sending the characteristics of the target derivative array to the target health status assessment model; receiving a target assessment result obtained by the health status assessment model based on the characteristics of the target derivative array for evaluating the battery, including: obtaining a target assessment result of the health status of the battery based on the characteristics of the target derivative array by the target health status assessment model.

[0011] In the above scheme, the working state of the battery is divided, and the extracted features are evaluated using the health state assessment model corresponding to the working state. Compared with using a single health state assessment model, this scheme can refer to the differences between the extracted features caused by the working state, thereby improving the accuracy of the evaluation results.

[0012] In some embodiments, the target health status assessment model is trained based on battery data collected from target operating states of the battery in different health states, wherein the health state includes a normal state, an abnormal state, and an intermediate state between the normal state and the abnormal state.

[0013] In the above scheme, the target health assessment model is trained by acquiring training battery data of the battery in the target working state under different health states, so that the assessment results subsequently output by the trained health state assessment model are more accurate.

[0014] In some embodiments, the battery data includes gas concentration data and air pressure data, and the battery data is differentiated to obtain a target derivative array, including: respectively differentiating multiple gas concentration data and multiple air pressure data measured within a preset time period to obtain a concentration derivative array and a pressure derivative array; based on the target derivative array, the health state of the battery is evaluated to obtain a target evaluation result, including: based on the concentration derivative array, the health state of the battery is evaluated to obtain a first evaluation result; and, based on the pressure derivative array, the health state of the battery is evaluated to obtain a second evaluation result; based on the first evaluation result and the second evaluation result, a target evaluation result is obtained.

[0015] In the above scheme, by acquiring the gas concentration data and the air pressure data within a preset time period, the corresponding first evaluation result and the second evaluation result are obtained, and then the final target evaluation result is obtained based on the first evaluation result and the second evaluation result. Compared with using a single type of sensor, the target evaluation result obtained by this scheme is more accurate.

[0016] In some embodiments, based on the first evaluation result and the second evaluation result, a target evaluation result is obtained, including: in response to the first evaluation result and the second evaluation result being the same, determining that the target evaluation result is the same as the first evaluation result and the second evaluation result; or, in response to the first evaluation result and the second evaluation result being different and the target number of times being greater than or equal to a preset number of times, determining that the target evaluation result includes a warning of the battery health status.

[0017] In the above scheme, if the two evaluation results are the same, then it can be determined that the two evaluation results are relatively accurate, and the first evaluation result and the second evaluation result can be directly used as the target evaluation result. If the evaluation results are different, but the number of evaluation results that require warnings is large, then in order to reduce the situation where battery problems are not discovered in time, this scheme will determine the target evaluation result as requiring a warning of battery health.

[0018] The present application provides a health status assessment device, including: a data acquisition module, a data processing module and an assessment module. The data acquisition module is used to acquire battery data measured by a target sensor within a preset time period; the data processing module is used to perform derivative processing on the battery data to obtain a target derivative array; the assessment module is used to assess the health status of the battery based on the target derivative array to obtain a target assessment result, and the target assessment result includes giving an early warning of the health status of the battery or not giving an early warning of the health status of the battery.

[0019] The present application provides a device, including a memory and a processor, wherein the processor is used to execute program instructions stored in the memory to implement the above-mentioned health status assessment method.

[0020] The present application provides a computer-readable storage medium on which program instructions are stored. When the program instructions are executed by a processor, the above method is implemented.

[0021] In the above scheme, the battery data measured by the target sensor within a preset time period is obtained, and then the target derivative array is obtained by deriving the data of each battery. The data after derivation can reduce the error caused by sensor zero drift and other reasons compared to the data before derivation. Then, the health status of the battery is evaluated based on the target derivative array. Compared with directly using the battery data to evaluate the health status of the battery, the evaluation result obtained by this scheme is more accurate.

[0022] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present application and are used together with the specification to illustrate the technical solution of the present application.

[0024] Figure 1 is a partial cross-sectional schematic diagram of a battery cell provided by one or more embodiments of the present application;

[0025] Figure 2 is a flowchart of a health status assessment method provided by one or more embodiments of the present application;

[0026] Figure 3 It is a partial flow chart of the health status assessment method provided by one or more embodiments of the present application;

[0027] Figure 4 is a schematic diagram of the structure of a health status assessment device provided by one or more embodiments of the present application;

[0028] Figure 5 is a schematic diagram of the structure of an electronic device provided by one or more embodiments of the present application;

[0029] Figure 6 It is a schematic diagram of the structure of a computer-readable storage medium in one or more embodiments of the present application.

[0030] Figure numbers: battery cell 10; battery housing 100; target sensor 200; light emitting group 210; light emitter 211; first protective cover 212; light receiving group 220; light receiver 221; second protective cover 222; gas detection light 230; health status assessment device 40; data acquisition module 41; data processing module 42; assessment module 43; equipment 50; memory 51; processor 52; computer-readable storage medium 60; program instructions 61. DETAILED DESCRIPTION

[0031] The scheme of the embodiment of the present application is described in detail below in conjunction with the drawings of the specification.

[0032] In the following description, for the purpose of explanation rather than limitation, specific details such as specific subsystem structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0033] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the objects associated before and after are in an "or" relationship. In addition, "many" in this article means two or more than two. In addition, the term "at least one" in this article means any combination of at least two of any one or more of a plurality of, for example, including at least one of A, B, and C, can mean including any one or more elements selected from the set consisting of A, B, and C.

[0034] The internal information of lithium batteries is of great value in thermal runaway warning and life assessment, especially the internal gases of lithium batteries (CO2, CO, H2, CH4, etc.) are the products of trace side reactions inside the battery. By detecting the concentration of gas in the battery, it is judged whether the lithium battery has thermal runaway. However, the gas concentration measured by the gas sensor is inevitably biased, resulting in inaccurate judgment results. Considering that directly using the data measured by the sensor for health status assessment may result in inaccurate assessment results, this solution first processes the measured data. The error of the processed data is lower, and the results obtained by evaluating the processed data are more accurate.

[0035] The battery disclosed in the present application may be a battery cell, or a battery module or battery module formed by connecting multiple battery cells. The battery can be used in an electrical device that uses the battery as a power source or in various energy storage systems that use the battery as an energy storage element. The electrical device may be, but is not limited to, a mobile phone, a tablet, a laptop computer, an electric toy, an electric tool, a battery car, an electric car, a ship, a spacecraft, and the like. Among them, the electric toy may include a fixed or mobile electric toy, for example, a game console, an electric car toy, an electric ship toy, an electric airplane toy, and the like, and the spacecraft may include an airplane, a rocket, a space shuttle, a spacecraft, and the like.

[0036] This embodiment takes the battery as a battery cell as an example. Each battery cell can be a secondary battery or a primary battery, and can also be a lithium battery, a sodium ion battery or a magnesium ion battery, but is not limited thereto. The battery cell can be cylindrical, flat, rectangular or other shapes. A battery cell refers to the smallest unit that makes up a battery. Figure 1 As shown, the battery cell 10 includes a battery housing 100 and a target sensor 200, and the target sensor 200 is disposed in the battery housing 100. In other embodiments, if the battery is a battery module or a battery module, the target sensor 200 can be disposed outside each battery cell, for example, in a receiving cavity of the battery for accommodating each battery cell, and the location of the target sensor 200 is not specifically limited here. The battery housing 100 can isolate the internal environment of the battery cell 10 from the external environment, and the battery housing 100 can have a certain hardness and strength, so that the battery housing 100 is not easily deformed when squeezed and collided, thereby improving the safety performance of the battery cell 10. The battery housing 100 can be in any shape, for example, the shape of the battery housing 100 includes but is not limited to square, cylindrical, prismatic, etc. The battery housing 100 can be an internal hollow structure, and the interior of the battery housing 100 can be used to accommodate objects such as electrodes, electrolytes and target sensors 200. The target sensor 200 is disposed in the battery housing 100, so that the target sensor 200 can accurately detect the internal information of the battery cell 10. For example, when the target sensor 200 includes a gas sensor, the target sensor 200 can detect the content of a specific gas inside the battery cell 10, wherein the specific gas may include but is not limited to carbon dioxide (CO 2 ), carbon monoxide (CO), methane (CH 4 ) and ethane (C 2 H 6). The target sensor 200 includes a light emitting group 210 and a light receiving group 220. The light emitting group 210 includes a light emitter 211 and a first protective cover 212 covering the light emitter 211. The light receiving group 220 includes a light receiver 221 and a second protective cover 222 covering the light receiver 221. The first protective cover 212 and the second protective cover 222 can be made of a material with good corrosion resistance, which can remain in its original state for a long time in a corrosive environment. For example, the first protective cover 212 and the second protective cover 222 can be corrosion-resistant protective covers relative to the electrolyte and / or corrosive gas. Therefore, the first protective cover 212 covers the light emitter 211, and the second protective cover 222 covers the light receiver 221, which can effectively alleviate the problem of the light emitter 211 and the light receiver 221 being corroded by contact with the electrolyte. The first protective cover 212 and the second protective cover 222 can cooperate with the battery casing 100 to respectively cover the light emitter 211 and the light receiver 221. By way of example, taking the first protective cover 212 and the light emitter 211 in the light emission group 210 as an example, the first protective cover 212 can be a hollow sealed shell, and the light emitter 211 is built into the first protective cover 212, so that the light emitter 211 is completely sealed by the first protective cover 212, and then the first protective cover 212 is fixed to the inner wall of the battery casing 100, so as to achieve the fixing effect of the light emission group 210 and the battery casing 100 and the anti-corrosion effect of the light emitter 211 at the same time. Alternatively, the first protective cover 212 may be a hollow concave shell, the light emitter 211 is built into the groove of the first protective cover 212, and the open end of the first protective cover 212 is blocked by the inner wall of the battery housing 100, so that the first protective cover 212 cooperates with the battery housing 100 to completely seal the light emitter 211, and while the first protective cover 212 is fixed to the inner wall of the battery housing 100, the light emitting group 210 and the battery housing 100 are fixed and the anti-corrosion effect of the light emitter 211 is achieved. Similarly, the second protective cover 222 and the light receiver 221 in the light receiving group 220 can be arranged in the same manner as the first protective cover 212 and the light emitter 211 in the light emitting group 210, and will not be repeated here. The light emitter 211 is used to emit the gas detection light 230, the first protective cover 212 and the second protective cover 222 are spaced apart in the optical path direction x of the gas detection light 230, and the gas detection light 230 is received by the light receiver 221 via the first protective cover 212, the medium between the first protective cover 212 and the second protective cover 222, and the second protective cover 222. The light emitter 211 may include but is not limited to a small light bulb, a light-emitting diode (LED), an infrared chip of a micro-electro-mechanical system (MEMS), etc. The light emitter 211 can emit the gas detection light 230, and the gas detection light 230 may include but is not limited to visible light, infrared light, etc.The optical path direction x of the gas detection light 230 can be understood as the direction in which the gas detection light 230 is emitted. Furthermore, the first protective cover 212 and the second protective cover 222 can be spaced apart in the optical axis direction of the light emitter 211. The optical axis of the light emitter 211 can be understood as the center line of the gas detection light 230, that is, when the gas detection light 230 rotates around the optical axis of the light emitter 211, there will be no obvious change in optical properties, and the extension direction of the optical axis can be identified as the optical axis direction of the light emitter 211. The optical receiver 221 may include but is not limited to a photodetector made of a thermopile chip, etc. The optical receiver 221 can receive the gas detection light 230 and generate an electrical signal according to the light intensity of the received gas detection light 230 to detect the concentration of the gas. The first protective cover 212 and the second protective cover 222 are spaced apart in the optical path direction x of the light emitter 211, so that a space allowing gas to pass is formed between the first protective cover 212 and the second protective cover 222. The medium between the first protective cover 212 and the second protective cover 222 can be an air medium, etc. When gas is generated inside the battery, at least part of the gas will enter the space between the first protective cover 212 and the second protective cover 222. The light transmitter 211 emits a gas detection light 230. When the gas detection light 230 passes through the medium between the first protective cover 212 and the second protective cover 222, at least part of the gas detection light 230 of a specific wavelength will be absorbed by the specific gas. After the light receiver 221 receives the gas detection light 230, the light intensity of the specific wavelength in the gas detection light 230 can be detected, thereby detecting the concentration of the specific gas. For example, when the gas to be detected is carbon dioxide, since the specific wavelength of carbon dioxide is 4.26μm, the light receiver 221 can generate an electrical signal by receiving the gas detection light 230 and analyzing the light intensity of the gas detection light 230 with a wavelength of 4.26μm to detect the concentration of carbon dioxide. Through the above implementation, the light emitter 211 is protected by the first protective cover 212, and the light receiver 221 is protected by the second protective cover 222, which can alleviate the problem of corrosion of the light emitter 211 and the light receiver 221 by the electrolyte and / or corrosive gas inside the battery. At the same time, the light emitter 211 can emit gas detection light 230 to be received by the light receiver 221, which is convenient for collecting internal information of the battery, analyzing the working status of the battery, and reducing the occurrence of problems such as thermal runaway of the battery. The above-mentioned setting method of the target sensor 200 in the battery cell 10 is only an example, and the target sensor 200 can also be set in the battery cell 10 in other ways.

[0037] See also Figure 2The health status assessment method provided in the present application may include the following steps S11 to S13. Step S11: Obtain battery data measured by a target sensor within a preset time period. Step S12: Perform derivative processing on the battery data to obtain a target derivative array. Step S13: Evaluate the health status of the battery based on the target derivative array to obtain a target evaluation result.

[0038] The battery data measured within the preset time period may be multiple, and the multiple may be two or more. As mentioned above, the battery may be a battery cell, and the target sensor may be set in the battery cell. The number of target sensors may be at least one. Among them, the at least one may be one or more than two. The target sensors include but are not limited to gas sensors and air pressure sensors. Gas sensors include but are not limited to those for detecting CO 2 or H 2 or CO or CH 4 Sensor. In some application scenarios, the target sensor includes a gas sensor, and the battery data measured by the gas sensor may be the concentration of a specific gas in the battery. In some application scenarios, the target sensor is an air pressure sensor, and the battery data measured by the air pressure sensor may be the air pressure in the battery. The data in the target array is the derivative obtained by the derivation process. The derivation process includes but is not limited to sorting the battery data in the order of the acquisition time, and performing derivation on two adjacent battery data, or other methods such as using three adjacent or battery data for derivation. The method of evaluating the health state of the battery based on the target derivative array may be to obtain the change trend or change law of the derivative in the target derivative array, and determine the health state of the battery according to the change trend or law. The target evaluation result includes warning the health state of the battery or not warning the health state of the battery. Among them, warning the health state of the battery can be considered that the health state of the battery is not optimistic and needs to be warned. Not warning the health state of the battery can be considered that the health state of the battery is still optimistic and no warning is needed. For example, if it is determined that the battery has a risk of thermal runaway according to the target derivative array, the target evaluation result is to warn the health state of the battery.

[0039] In the above scheme, the battery data measured by the target sensor within a preset time period is obtained, and then the target derivative array is obtained by deriving the data of each battery. The data after derivation can reduce the error caused by sensor zero drift and other reasons compared to the data before derivation. Then, the health status of the battery is evaluated based on the target derivative array. Compared with directly using the battery data to evaluate the health status of the battery, the evaluation result obtained by this scheme is more accurate.

[0040] It is understandable that the number of times the battery data is derivatized can be once, or two or more. Exemplarily, the battery data is derivatized once to obtain a derivative array, and then the health state of the battery is evaluated based on the derivative array after the first derivation to obtain a target evaluation result. The method of one derivation is as described above and will not be repeated here. By deriving the battery data once, the interference caused by the zero drift of the sensor can be reduced. Compared with the method of directly using the battery data to evaluate the health state of the battery, the evaluation result obtained by this solution is more accurate.

[0041] In some embodiments, the above step S12 may include the following steps: performing a first derivative on the battery data to obtain an initial derivative array, and then performing a second derivative on the data in the initial derivative array to obtain a target derivative array.

[0042] If there are multiple target sensors, performing a derivation on the battery data once to obtain an initial derivative array can be performed by derivation of the battery data obtained by each target sensor once to obtain an initial derivative array corresponding to each target sensor, and performing a second derivation on the data in the initial derivative array to obtain a target derivative array. Specifically, performing a second derivation on each initial derivative array to obtain a target derivative array corresponding to each initial derivative array. The method of the first derivation and the second derivation can be the same. The initial derivative array obtained after the first derivation can represent the change trend of the battery data within a preset time period, and the target derivative array obtained by the second derivation can represent the change trend of the derivatives in the initial derivative array within a preset time period.

[0043] In the above scheme, the problem of sensor zero drift can be solved by taking a derivative of the sensor data once, and the data error caused by sensor aging and the like can be further reduced by taking a second derivative of the data in the initial derivative array.

[0044] In some embodiments, the above-mentioned method of evaluating the health status of the battery based on the target derivative array to obtain the target evaluation result may be: extracting the characteristics of the target derivative array, the characteristics including one or more of the maximum value, minimum value, average value, standard deviation, period, and difference between the maximum value and the minimum value. Then, sending the characteristics of the target derivative array to the health status evaluation model. And receiving the health status evaluation model to evaluate the health status of the battery based on the characteristics of the target derivative array to obtain the target evaluation result.

[0045] The maximum value refers to the largest derivative in the target derivative array, the minimum value refers to the smallest derivative in the target derivative array, the average value is the average value of each derivative in the target derivative array, the standard deviation refers to the standard deviation between each derivative in the target derivative array, and the period can be the change period of each derivative or the period can be the acquisition period of the battery data corresponding to the derivative. The health state assessment model can be a neural network model, a deep learning model, etc. The health state assessment model can be obtained by training in advance. Sending the features of the target derivative array to the health state assessment model can specifically be taking the features of the target derivative array as the input of the health state assessment model, and the output of the health state assessment model can be the target assessment result. Among them, the health state assessment model can be embedded in the health state assessment method execution device so that the execution device calls the health state assessment model to evaluate the health state of the battery based on the features of the target derivative array, or the health state assessment model can also be embedded in other devices that establish a communication connection with the execution device so that the features of the target derivative array are sent to other devices through the communication connection between the execution device and the other devices to change the health state assessment model embedded in the other device to evaluate the health state of the battery.

[0046] In the above scheme, by extracting the features from the target to the array and then using the health status assessment model to perform assessment based on the extracted features, the assessment error rate caused by subjective factors of human assessment can be reduced, the objectivity of the assessment can be improved, and the assessment efficiency can also be improved.

[0047] In some embodiments, the above-mentioned method of evaluating the health state of the battery based on the target derivative array to obtain the target evaluation result may be: obtaining the development trend of each derivative value in the target derivative array in time series, evaluating the health state of the battery according to the development trend, and obtaining the target evaluation result. Optionally, the acquired development trend is matched with multiple preset development trends, and the health state corresponding to the preset development trend whose matching result meets the matching requirement is used as the current health state of the battery. Among them, different preset development trends can be obtained by obtaining battery data collected from several batteries in different health states, and the health states corresponding to different preset development trends may be different. In some application scenarios, if the derivative value increases continuously, it is likely to indicate that the battery data in the battery changes continuously within the preset time period, and the health state of the battery is likely to have problems. The health state of the battery can also be judged by judging the development trend of each derivative value in the target derivative array in time series.

[0048] In some embodiments, the above method of sending the characteristics of the target derivative array to the health status assessment model may be: determine the target working state of the battery. Then determine the target health status assessment model corresponding to the target working state. Send the characteristics of the target derivative array to the target health status assessment model. The above method of receiving the target assessment result obtained by the health status assessment model based on the characteristics of the target derivative array to evaluate the battery may be: obtain the target assessment result of the health status of the battery by the target health status assessment model based on the characteristics of the target derivative array.

[0049] The number of health state assessment models may be at least two. The working states of the battery corresponding to different health state assessment models may be different. The working state of the battery includes two or three of a charging state, a discharging state, and an idle state. The idle state means that the battery is neither in a charging state nor in a discharging state. Among them, the method of determining the target working state of the battery may be to judge the state of the battery by the current (+, -, 0) in the battery. If the battery is in a charging state, the health state assessment model corresponding to the charging state is used to assess the health state of the battery based on the characteristics of the target derivative array to obtain a target assessment result. If the battery is in a discharging state, the health state assessment model corresponding to the discharging state is used to assess the health state of the battery based on the characteristics of the target derivative array to obtain a target assessment result. If the battery is in an idle state, the health state assessment model corresponding to the idle state is used to assess the health state of the battery based on the characteristics of the target derivative array to obtain a target assessment result.

[0050] In the above scheme, the working state of the battery is divided, and the extracted features are evaluated using the health state assessment model corresponding to the working state. Compared with using a single health state assessment model, this scheme can refer to the differences between the extracted features caused by the working state, thereby improving the accuracy of the evaluation results.

[0051] In some embodiments, the target health state assessment model is trained based on battery data collected from target working states of the battery in different health states, wherein the health state includes a normal state, an abnormal state, and an intermediate state between the normal state and the abnormal state.

[0052] The training data of the health state assessment model corresponding to the charging state may be collected when the battery is in the charging state. The training data of the health state assessment model corresponding to the discharging state may be collected when the battery is in the discharging state. The training data of the health state assessment model corresponding to the idle state may be collected when the battery is in the idle state. Abnormal states include but are not limited to thermal runaway, end of life, and diving. The normal state may be a state in which the working parameters of the battery meet the expected conditions, that is, a state in which the battery can work normally. The intermediate state between the normal state and the abnormal state may be a state in which the working parameters of the battery slightly deviate from the normal state and have not yet reached the abnormal state, for example, the intermediate state may be a state of n cycles before the abnormal state. The training battery data in each working state of the battery in different health states may specifically include the battery data collected when the battery is in the normal state, abnormal state, and intermediate state during the charging process as the training battery data, and the battery data collected when the battery is in the normal state, abnormal state, and intermediate state during the discharging process as the training battery data, and the battery data collected when the battery is in the normal state, abnormal state, and intermediate state during the idle process as the training battery data. For each health status assessment model, the health status assessment model is trained using the training battery data obtained in the working state corresponding to the health status assessment model. Exemplarily, for the health status model corresponding to the charging state, it is trained using the battery data collected in the charging state, for the health status model corresponding to the discharging state, it is trained using the battery data collected in the discharging state, and for the health status model corresponding to the idle state, it is trained using the battery data collected in the charging state. In some application scenarios, battery data is collected when the battery is in a normal state, an abnormal state, and an intermediate state, and the battery power SOC is 0, 20%, 50%, and 100%, respectively.

[0053] In the above scheme, the target health assessment model is trained by acquiring training battery data of the battery in various working states under different health states, so that the assessment results subsequently output by the trained health state assessment model are more accurate.

[0054] In some embodiments, the battery data includes gas concentration data and air pressure data. The above-mentioned method of performing derivative processing on the battery data to obtain the target derivative array may be: performing derivative processing on multiple gas concentration data and multiple air pressure data measured within a preset time period, respectively, to obtain a concentration derivative array and an air pressure derivative array. The above-mentioned method of evaluating the health state of the battery based on the target derivative array to obtain the target evaluation result may be: evaluating the health state of the battery based on the concentration derivative array to obtain a first evaluation result. And, evaluating the health state of the battery based on the air pressure derivative array to obtain a second evaluation result. Then, based on the first evaluation result and the second evaluation result, the target evaluation result is obtained.

[0055] The target sensor may include a gas sensor and a pressure sensor. The battery data measured by the gas sensor is the gas concentration data of the preset gas, and the battery data measured by the gas sensor includes pressure data. The preset gas may be CO 2 , H 2 , CO, CH 4 Etc. The multiple gas concentration data and multiple air pressure data measured within the preset time period are respectively derivatized to obtain the concentration derivative array and the pressure derivative array. Specifically, the gas concentration data collected by the gas sensor is derivatized to obtain the concentration derivative array, and the air pressure data collected by the pressure sensor is derivatized to obtain the pressure derivative array. The number of derivations can be one derivation or two derivations. That is, the health status of the battery is evaluated based on the concentration derivative array using the health status assessment model to obtain a first assessment result, and the health status of the battery is evaluated based on the pressure derivative array using the health status assessment model to obtain a second assessment result. Based on the first assessment result and the second assessment result, the target assessment result can be obtained by selecting one of the assessment results as the target assessment result, or by evaluating the first assessment result and the second assessment result to obtain the target assessment result.

[0056] In the above scheme, by acquiring the gas concentration data and the air pressure data within a preset time period, the corresponding first evaluation result and the second evaluation result are obtained, and then the final target evaluation result is obtained based on the first evaluation result and the second evaluation result. Compared with using a single type of sensor, the target evaluation result obtained by this scheme is more accurate.

[0057] In some embodiments, the method of obtaining the target evaluation result based on the first evaluation result and the second evaluation result may be: in response to the first evaluation result and the second evaluation result being the same, determining the target evaluation result to be the same as the first evaluation result and the second evaluation result. Or, in response to the first evaluation result and the second evaluation result being different and the target number of times being greater than or equal to a preset number of times, determining the target evaluation result includes issuing an early warning of the health status of the battery.

[0058] The target number is the number of times that an evaluation result that warns the health of the battery appears within a preset counting cycle. In response to the first evaluation result and the second evaluation result being the same, determining that the target evaluation result is the same as the first evaluation result and the second evaluation result may specifically be that if both the first evaluation result and the second evaluation result require an early warning of the health of the battery, the target evaluation result is an early warning of the health of the battery. If both the first evaluation result and the second evaluation result do not require an early warning of the health of the battery, the target evaluation result is not an early warning of the health of the battery. The first evaluation result and the second evaluation result may be different in that one of the first evaluation result and the second evaluation result is an early warning of the health of the battery, and the other is not an early warning of the health of the battery. The preset counting cycle may be a plurality of preset time periods, and the length of the specific preset counting cycle may be customized by the user. The number of target times may also be customized by the user. For example, the target number is 3, the preset counting cycle is 10 preset time periods, and if there are 3 times within 10 preset time periods when an early warning of the health of the battery is required, the target evaluation result is determined to be an early warning of the health of the battery. If the target number does not reach 3 times within 10 preset time periods, the target evaluation result is determined to be not an early warning of the health of the battery.

[0059] In the above scheme, if the two evaluation results are the same, then it can be determined that the two evaluation results are relatively accurate, and the first evaluation result and the second evaluation result can be directly used as the target evaluation result. If the evaluation results are different, but the number of evaluation results that require warnings is large, then in order to reduce the situation where battery problems are not discovered in time, this scheme will determine the target evaluation result as requiring a warning of battery health.

[0060] In some embodiments, the first evaluation result may include the confidence of the first evaluation result, and the second evaluation result may also include the confidence of the second evaluation result. Based on the first evaluation result and the second evaluation result, the target evaluation result may be obtained by weighted fusion of the first evaluation result and the second evaluation result to obtain the target evaluation result.

[0061] Exemplarily, the first evaluation result includes that an early warning of the battery's health state is required and the confidence of the first evaluation result is 0.3, and the second evaluation result is that an early warning of the battery's health state is not required and the confidence of the second evaluation result is 0.9. Assuming that the weighted weights of the first evaluation result and the second evaluation result are the same, the target evaluation result obtained after weighted fusion is that the battery's health state does not need to be evaluated. Optionally, the weighted weight corresponding to the evaluation result that requires an early warning of the battery's health state is greater than the weighted weight of the evaluation result that does not require an early warning of the battery's health state. For example, the first evaluation result includes that an early warning of the battery's health state is required and the confidence of the first evaluation result is 0.5, and the second evaluation result is that an early warning of the battery's health state is not required and the confidence of the second evaluation result is 0.5. The weighted weight of the evaluation result that requires an early warning of the battery's health state is 0.8, and the weighted weight of the evaluation result that does not require an early warning of the battery's health state is 0.2. Then the target evaluation result after weighted fusion is that an early warning of the battery's health state is required.

[0062] By setting corresponding weighted weights for the first evaluation result and the second evaluation result, and taking the evaluation result after weighted fusion as the target evaluation result, the evaluation result can be determined by referring to a variety of battery data, thereby improving the accuracy of the target evaluation result.

[0063] In other embodiments, the battery data includes gas concentration data and air pressure data. After the above-mentioned multiple gas concentration data and multiple air pressure data measured within the preset time period are respectively derivatized to obtain the concentration derivative array and the air pressure derivative array, the features of the concentration derivative array and the air pressure derivative array are extracted, and the features of the concentration derivative array and the air pressure derivative array are sent to the corresponding target health status assessment model. The target health status assessment model assesses the health status of the battery based on the features of the concentration derivative array and the features of the other derivative arrays. That is, the features of the two derivative arrays are simultaneously used as the input of the target health status assessment model to obtain the target assessment result.

[0064] In some embodiments, the training of the health status assessment model may refer to the following examples:

[0065] The data L' of the gas sensor built into the battery is twice derived to form new data L". The characteristics of the L" signal (maximum value, minimum value, average value, standard deviation, period, difference between maximum value and minimum value) are extracted as the input of the health status assessment model, so as to establish the assessment model of thermal runaway, end of life, diving, etc., with the gas sensor as the CO 2 Sensors as an example.

[0066] First, a database can be established to store the collected battery data. Then, the battery data under normal conditions (SOC = 0, 20%, 50%, 100%) is collected, including the CO during the charging process. 2 concentration, CO during discharge 2 concentration and CO during idle 2 Concentration. Collect the data of the 50 cycles before the abnormal state, including the CO 2 concentration, CO during discharge 2 concentration and CO during idle 2 Concentration. Collect the data of the 10 cycles before the abnormal state, including the CO 2 concentration, CO during discharge 2 concentration and CO during idle 2 Concentration. Collect data under abnormal conditions, including CO 2 concentration, CO during discharge 2 concentration and CO during idle 2 concentration.

[0067] Then, the derivatives of each group of collected data are derived to obtain the target derivative array, and then the features of the target derivative array are extracted. CO is collected in three states (charging, discharging and idle) 2 Battery data L0 (concentration). After deriving L0 twice, we get the data L0″, and extract the characteristic parameters of L0″, including the maximum value, minimum value, average value, standard deviation, cycle, the difference between the maximum value and the minimum value, and other parameters. The features of the target derivative arrays extracted during the charging process are used as input to train the health status assessment model corresponding to the charging process. The features of the target derivative arrays extracted during the discharge process are used as input to train the health status assessment model corresponding to the discharge process. The features of the target derivative arrays extracted during the idle process are used as input to train the health status assessment model corresponding to the idle process.

[0068] Specifically, the above battery data can be obtained through failure experiments. For example, by heating, pressurizing, overcharging, over-discharging, etc., the CO2 concentration values ​​of the three working states of the battery under charging, discharging and idle conditions are recorded until the battery fails. The CO2 concentration values ​​of the three states of charging, discharging and idle conditions are extracted in the case of failure (lithium battery abnormality, including thermal runaway, end of life, etc.), 10 charge and discharge cycles before failure, and 50 charge and discharge cycles before failure. 2Gas concentration data (L1, L2, L3), after taking the second derivative of the data, a data set (L1", L2", L3") is formed, and the aforementioned characteristic parameters are extracted from (L1", L2", L3"). The health status assessment model is trained using the data sets (L0", L1", L2", L3") in three states (charging, discharging and idle) respectively to obtain the health status assessment models corresponding to the three states (charging, discharging and idle), and the health status assessment model can be an ANN model.

[0069] During the application process, the state of the battery can be determined by detecting the current (+, -, 0). For example, the current parameter of the lithium battery is read by the BMS system to determine the current state of the lithium battery (charging, discharging and idle). Then the detected battery data is input into the corresponding health status assessment model to output the target assessment result. Among them, the target assessment result may also include the assessment results of multiple risks, such as the assessment results of thermal runaway, the diving assessment results and the life assessment results. The thermal runaway assessment result includes whether to perform a thermal runaway warning. For example, if it is determined that the battery has a thermal runaway risk, a thermal runaway warning is issued for the battery. The diving assessment result includes whether to perform a diving warning. For example, if it is determined that the capacity of the battery fluctuates more than the capacity in the previous preset time period, a diving warning is issued for the battery. The life assessment result includes whether to perform a life warning. For example, if it is determined that the available power of the battery drops to the preset available power, a life warning is issued. For example, if the available power of the battery is only 80% of the original power, a life warning is issued.

[0070] As mentioned above, gas sensors and pressure sensors can be used to collect data. 2 and air pressure sensors as examples, respectively for CO 2 The output data of the sensor and the air pressure sensor are processed to obtain corresponding concentration derivative arrays and pressure derivative arrays, and the first evaluation result and the second evaluation result are obtained based on the processing.

[0071] like Figure 3 As shown, read CO 2 The model warning value corresponding to the sensor and the air pressure sensor, that is, CO 2 The warning signal SC in the first evaluation result corresponding to the sensor, and the warning signal SP in the second evaluation result corresponding to the air pressure sensor. If both sensors are warned (SC=1&SP=1), the target evaluation result is determined to generate a warning signal, that is, to warn the health status of the battery. If the evaluation results obtained by both sensors are no warning (SC=0&SP=0), the target evaluation result is not to generate a warning signal, that is, not to warn the health status of the battery.

[0072] If the evaluation result corresponding to one sensor is a warning, and the evaluation result corresponding to the other sensor is no warning, the corresponding counter value is increased by 1, and this situation (one sensor warning, and the other sensor no warning) occurs three times in a row to generate a warning signal, that is, if it does not occur three times in a row, no warning will be issued. The three consecutive occurrences specifically refer to the evaluation results in three preset time periods being one warning and the other no warning. Optionally, if the situation occurs a preset number of times cumulatively, a warning may be issued even if it does not occur three times in a row.

[0073] See also Figure 4 ,like Figure 4 As shown, the health status assessment device 40 provided in this embodiment may include: a data acquisition module 41, a data processing module 42 and an assessment module 43. The data acquisition module 41 is used to acquire the battery data measured within a preset time period; the data processing module 42 is used to perform derivative processing on the battery data to obtain a target derivative array; the assessment module 43 is used to evaluate the health status of the battery based on the target derivative array to obtain a target assessment result, and the target assessment result includes warning the health status of the battery or not warning the health status of the battery.

[0074] In the above scheme, the battery data measured by the target sensor within a preset time period is obtained, and then the target derivative array is obtained by deriving the data of each battery. The data after derivation can reduce the error caused by sensor zero drift and other reasons compared to the data before derivation. Then, the health status of the battery is evaluated based on the target derivative array. Compared with directly using the battery data to evaluate the health status of the battery, the evaluation result obtained by this scheme is more accurate.

[0075] In some embodiments, the data processing module 42 performs derivative processing on the battery data to obtain a target derivative array, including: performing a first derivative processing on the battery data to obtain an initial derivative array; performing a second derivative processing on the data in the initial derivative array to obtain a target derivative array.

[0076] In the above scheme, the problem of sensor zero drift can be solved by taking a derivative of the sensor data once, and the data error caused by sensor aging and the like can be further reduced by taking a second derivative of the data in the initial derivative array.

[0077] In some embodiments, the evaluation module 43 evaluates the health status of the battery based on the target derivative array to obtain a target evaluation result, including: extracting features of the target derivative array, the features including one or more of the maximum value, minimum value, average value, standard deviation, period, and difference between the maximum value and the minimum value; sending the features of the target derivative array to the health status evaluation model; receiving the target evaluation result obtained by the health status evaluation model by evaluating the battery based on the features of the target derivative array.

[0078] In the above scheme, by extracting the features from the target to the array and then using the health status assessment model to perform assessment based on the extracted features, the assessment error rate caused by subjective factors of human assessment can be reduced, the objectivity of the assessment can be improved, and the assessment efficiency can also be improved.

[0079] In some embodiments, the evaluation module 43 sends the characteristics of the target derivative array to the health status evaluation model, including: determining the target working state of the battery; determining the target health status evaluation model corresponding to the target working state; sending the characteristics of the target derivative array to the target health status evaluation model; receiving the target evaluation result obtained by the health status evaluation model based on the characteristics of the target derivative array for evaluating the battery, including: obtaining the target evaluation result of the health status of the battery by the target health status evaluation model based on the characteristics of the target derivative array.

[0080] In the above scheme, the working state of the battery is divided, and the extracted features are evaluated using the health state assessment model corresponding to the working state. Compared with using a single health state assessment model, this scheme can refer to the differences between the extracted features caused by the working state, thereby improving the accuracy of the evaluation results.

[0081] In some embodiments, the target health status assessment model is trained based on battery data collected from target working states of the battery in different health states, wherein the health state includes a normal state, an abnormal state, and an intermediate state between the normal state and the abnormal state; for each health status assessment model, the health status assessment model is trained using the training battery data obtained in the working state corresponding to the health status assessment model.

[0082] In the above scheme, by obtaining training battery data of the battery in the target working state under different health states, and then training the target health state assessment model, the assessment results subsequently output by the trained health state assessment model are more accurate.

[0083] In some embodiments, the battery data includes gas concentration data and air pressure data. The data processing module 42 performs derivative processing on the battery data to obtain a target derivative array, including: performing derivative processing on multiple gas concentration data and multiple air pressure data measured within a preset time period to obtain a concentration derivative array and an air pressure derivative array; the evaluation module 43 evaluates the health state of the battery based on the target derivative array to obtain a target evaluation result, including: evaluating the health state of the battery based on the concentration derivative array to obtain a first evaluation result; and evaluating the health state of the battery based on the air pressure derivative array to obtain a second evaluation result; based on the first evaluation result and the second evaluation result, obtain a target evaluation result.

[0084] In the above scheme, by acquiring the gas concentration data and the air pressure data within a preset time period, the corresponding first evaluation result and the second evaluation result are obtained, and then the final target evaluation result is obtained based on the first evaluation result and the second evaluation result. Compared with using a single type of sensor, the target evaluation result obtained by this scheme is more accurate.

[0085] In some embodiments, the evaluation module 43 obtains a target evaluation result based on the first evaluation result and the second evaluation result, including: in response to the first evaluation result and the second evaluation result being the same, determining that the target evaluation result is the same as the first evaluation result and the second evaluation result; or, in response to the first evaluation result and the second evaluation result being different and the target number of times being greater than or equal to a preset number of times, determining that the target evaluation result includes a warning of the battery health status.

[0086] In the above scheme, if the two evaluation results are the same, then it can be determined that the two evaluation results are relatively accurate, and the first evaluation result and the second evaluation result can be directly used as the target evaluation result. If the evaluation results are different, but the number of evaluation results that require warnings is large, then in order to reduce the situation where battery problems are not discovered in time, this scheme will determine the target evaluation result as requiring a warning of battery health.

[0087] See also Figure 5 The device 50 provided in this embodiment includes a memory 51 and a processor 52. The processor 52 is used to execute the program instructions stored in the memory 51 to implement the steps in any of the above health status assessment method embodiments. In a specific implementation scenario, the electronic device 50 may include, but is not limited to: an electrical device, a vehicle, a monitoring device, a microcomputer, and a server. In addition, the electronic device 50 may also include a carrier device such as a laptop computer and a tablet computer, which is not limited here.

[0088] Specifically, the processor 52 is used to control itself and the memory 51 to implement the steps in any of the above-mentioned posture acquisition method embodiments. The processor 52 can also be called a CPU (Central Processing Unit). The processor 52 may be an integrated circuit chip with signal processing capabilities. The processor 52 can also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In addition, the processor 52 can be implemented by an integrated circuit chip.

[0089] In the above scheme, the battery data measured by the target sensor within a preset time period is obtained, and then the target derivative array is obtained by deriving the data of each battery. The data after derivation can reduce the error caused by sensor zero drift and other reasons compared to the data before derivation. Then, the health status of the battery is evaluated based on the target derivative array. Compared with directly using the battery data to evaluate the health status of the battery, the evaluation result obtained by this scheme is more accurate.

[0090] See also Figure 6 The computer-readable storage medium 60 provided in this embodiment stores program instructions 61 that can be executed by a processor, and the program instructions 61 are used to implement the steps in any of the above-mentioned health status assessment method embodiments.

[0091] In the above scheme, the battery data measured by the target sensor within a preset time period is obtained, and then the target derivative array is obtained by performing a derivative process on each battery data. The data after the derivative can reduce the error caused by sensor zero drift and other reasons compared to the data before the derivative. Then, the health status of the battery is evaluated based on the target derivative array. Compared with directly using the battery data to evaluate the health status of the battery, the evaluation result obtained by this scheme is more accurate. The above description of each embodiment tends to emphasize the differences between the embodiments. The same or similar aspects can be referenced to each other. For the sake of brevity, this article will not repeat them.

[0092] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as units or components can be combined or integrated into another subsystem, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0093] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

Claims

1. A health status assessment method, characterized in that: Obtaining battery data measured by a target sensor within a preset time period; Performing derivative processing on the battery data to obtain a target derivative array; The health state of the battery is evaluated based on the target derivative array to obtain a target evaluation result, wherein the target evaluation result includes issuing a warning for the health state of the battery or not issuing a warning for the health state of the battery.

2. The method according to claim 1, characterized in that: The step of performing a derivation process on the battery data to obtain a target derivative array includes: Deriving the battery data once to obtain an initial derivative array; The data in the initial derivative array is derivatized twice to obtain the target derivative array.

3. The method according to claim 1, characterized in that The step of evaluating the health status of the battery based on the target derivative array to obtain a target evaluation result includes: Extracting features of the target derivative array, wherein the features include one or more of a maximum value, a minimum value, an average value, a standard deviation, a period, and a difference between a maximum value and a minimum value; Sending the characteristics of the target derivative array to a health status assessment model; Receive the target assessment result obtained by the health status assessment model evaluating the battery based on the characteristics of the target derivative array.

4. The method according to claim 3, characterized in that The step of sending the characteristics of the target derivative array to the health status assessment model comprises: Determining a target operating state of the battery; Determining a target health status assessment model corresponding to the target working status; sending the characteristics of the target derivative array to the target health status assessment model; The receiving the target assessment result obtained by evaluating the battery by the health status assessment model based on the characteristics of the target derivative array includes: Obtain a target assessment result of the health state of the battery by the target health state assessment model based on the characteristics of the target derivative array.

5. The method according to claim 4, characterized in that The target health status assessment model is trained based on battery data collected from the target working state of the battery in different health states, wherein the health state includes a normal state, an abnormal state, and an intermediate state between the normal state and the abnormal state.

6. The method according to any one of claims 1 to 5, characterized in that: The battery data includes gas concentration data and gas pressure data, and the battery data is subjected to derivation processing to obtain a target derivative array, including: Performing derivative processing on the multiple gas concentration data and the multiple air pressure data measured within the preset time period respectively to obtain a concentration derivative array and an air pressure derivative array; The step of evaluating the health status of the battery based on the target derivative array to obtain a target evaluation result includes: Evaluate the health state of the battery based on the concentration derivative array to obtain a first evaluation result; and, evaluating the health status of the battery based on the air pressure derivative array to obtain a second evaluation result; The target evaluation result is obtained based on the first evaluation result and the second evaluation result.

7. The method according to claim 6, characterized in that The obtaining the target evaluation result based on the first evaluation result and the second evaluation result includes: In response to the first evaluation result and the second evaluation result being the same, determining that the target evaluation result is the same as the first evaluation result and the second evaluation result; or, In response to the first evaluation result and the second evaluation result being different and the target number of times being greater than or equal to a preset number of times, determining the target evaluation result includes issuing an early warning for the health status of the battery.

8. A health status assessment device, characterized in that: include: A data acquisition module, used to acquire battery data measured by a target sensor within a preset time period; A data processing module, used for performing derivative processing on the battery data to obtain a target derivative array; An evaluation module is used to evaluate the health status of the battery based on the target derivative array to obtain a target evaluation result, wherein the target evaluation result includes issuing a warning for the health status of the battery or not issuing a warning for the health status of the battery.

9. A device, characterized in that: The method comprises a memory and a processor, wherein the processor is used to execute program instructions stored in the memory to implement the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images