Battery surface temperature monitoring method, server, battery swap station and storage medium
Through the LSTM model based on attention mechanism and non-contact temperature sensor, combined with multi-dimensional information to monitor the surface temperature of the power battery, the problem of thermal runaway in the existing technology is solved, and timely and accurate monitoring and alarm of the surface temperature of the battery is achieved. Reduced operational risks.
Patent Information
- Application Number
- CN202411046441.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-19
AI Technical Summary
The existing battery management system cannot timely and accurately monitor whether the surface temperature of the power battery is abnormal, resulting in the inability to effectively avoid the battery thermal runaway, increasing the operating risk of the battery swap station.
The LSTM model based on attention mechanism is used to monitor the battery's multi-dimensional information (including surface temperature, state information and environmental information). The temperature map is collected through non-contact temperature sensors and clustered to obtain temperature characteristics and use pre-pre- Suppose the knowledge base determines the temperature status and alerts when an abnormality is detected.
It realizes timely and accurate monitoring of the battery surface temperature, reduces manual intervention, adapts to unmanned scenarios, has high compatibility and convenient upgrades, and can promptly avoid thermal runaway from the battery.
Smart Images

Figure CN118980946A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of battery monitoring technology, and specifically to a battery surface temperature monitoring method, server, battery swap station and storage medium. Background Art
[0002] With the continuous development of power battery technology, more and more people are using electric vehicles. If the power battery has external damage, internal circuit failure, performance aging, improper use and maintenance, etc., the internal temperature of the power battery may be too high during charging and storage, which may cause abnormal surface temperature of the power battery and even fire or explosion.
[0003] A battery swap station is a facility that provides power battery replacement services for electric vehicles. The battery swap station integrates battery charging, storage, maintenance and other functions. For the convenience of users, it is widely deployed in transportation hubs such as roads, bus stops, subway stations, as well as commercial, residential, office and other areas that are close to actual needs. In order to ensure the safety of the battery swap station, it is necessary to accurately monitor the surface temperature of the batteries in the battery swap station so that safety measures can be taken in time when abnormal surface temperature is detected to prevent thermal runaway of the battery. At present, the monitoring of battery status is mainly carried out by the Battery Management System (BMS), but the battery management system cannot timely and accurately monitor whether the surface temperature of the battery is abnormal, so it is impossible to effectively avoid thermal runaway of the battery, which increases the operating risk of the battery swap station.
[0004] Accordingly, the art needs a new technical solution to solve the above problems. Summary of the invention
[0005] In order to overcome the above-mentioned defects, the present application is proposed to solve or at least partially solve the technical problem of timely and accurately monitoring whether the battery surface temperature is abnormal.
[0006] In a first aspect, a method for monitoring a battery surface temperature is provided, comprising:
[0007] Acquire multi-dimensional information of a first target battery in a battery compartment, where the first target battery is any battery in the battery compartment;
[0008] Adopting an LSTM model based on an attention mechanism and determining a surface temperature state of the first target battery according to the multi-dimensional information of the first target battery, wherein the surface temperature state is a normal surface temperature or an abnormal surface temperature;
[0009] in,
[0010] The multi-dimensional information includes surface temperature information and status information of the battery, and the status information includes power status and working status;
[0011] The LSTM model based on the attention mechanism is trained according to a first data sample and a second data sample, wherein the first data sample includes the multi-dimensional information of the battery when the surface temperature is normal, and the second data sample includes the multi-dimensional information of the battery when the surface temperature is abnormal.
[0012] In a technical solution of the above-mentioned method for monitoring the battery surface temperature, the multi-dimensional information includes environmental information, and the environmental information includes the temperature and humidity of the battery compartment.
[0013] In a technical solution of the above-mentioned method for monitoring the battery surface temperature, the multi-dimensional information includes surface temperature information of a second target battery, and the second target battery is a battery remaining in the battery compartment other than the first target battery.
[0014] In a technical solution of the above-mentioned method for monitoring the battery surface temperature, a non-contact temperature sensor is provided in the battery compartment, and the surface temperature information is the actual temperature value of the battery surface. The method includes obtaining the actual temperature value by:
[0015] Using the non-contact temperature sensor to collect a temperature map of the battery in the battery compartment;
[0016] The pixel value of each pixel in the temperature map represents the actual temperature value at different positions on the battery surface.
[0017] In a technical solution of the above-mentioned method for monitoring the battery surface temperature, the surface temperature information is a temperature characteristic of the battery surface, and the method includes obtaining the temperature characteristic by:
[0018] Clustering the pixels in the temperature map according to the pixel values of the pixels in the temperature map to obtain a plurality of pixel sets;
[0019] According to the pixel values of the pixels in each pixel set, respectively obtain the first temperature characteristics of each pixel set;
[0020] A second temperature characteristic of the surface temperature of the first target battery is acquired according to the first temperature characteristic of all pixel sets.
[0021] In a technical solution of the above-mentioned battery surface temperature monitoring method, the first temperature feature of each pixel set is obtained according to the pixel value of each pixel in the pixel set, including:
[0022] According to the pixel value of each pixel in the pixel set, m1 maximum pixel values are obtained, wherein the m1 maximum pixel values are pixel values of pixels arranged from the 1st to the m1th pixels after the pixels are arranged in descending order according to the pixel values, and m1>1;
[0023] According to the deviation of each pixel in the pixel set, m2 maximum deviations are obtained, wherein the m2 maximum deviations are deviations of pixels arranged from the first to the m2th pixels after the pixels are arranged in descending order according to the deviations, and m2>1;
[0024] According to the temperature rise rate of each pixel in the pixel set, m3 maximum temperature rise rates are obtained, wherein the m3 maximum temperature rise rates are the temperature rise rates of the pixels arranged from the first to the m3th pixels after the pixels are arranged in descending order according to the temperature rise rates, and m3>1;
[0025] acquiring the first temperature feature according to the m1 largest pixel values, the m2 largest deviations and the m3 largest temperature rise rates;
[0026] in,
[0027] The deviation is the difference between the pixel value of the pixel and the pixel mean, and the pixel mean is the average value of the pixel values of all pixels in the pixel set;
[0028] The temperature rise rate is a rate of change of the pixel value of the pixel within a preset time.
[0029] In one technical solution of the above-mentioned method for monitoring the battery surface temperature, the method comprises:
[0030] Determine whether the multi-dimensional information is in a normal state according to a preset knowledge base; if it is in a normal state, adopt the LSTM model based on the attention mechanism, and determine the surface temperature state of the first target battery according to the multi-dimensional information of the first target battery;
[0031] The knowledge base includes normal features and abnormal features, wherein the normal features represent features when the multi-dimensional information is in a normal state, and the abnormal features represent features when the multi-dimensional information is in an abnormal state.
[0032] In a technical solution of the above-mentioned method for monitoring the surface temperature of a battery, the method includes: if the surface temperature state of the first target battery is abnormal, an alarm is issued.
[0033] In a second aspect, a server is provided, comprising at least one processor; and a memory communicatively connected to the at least one processor; wherein a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the method described in any one of the technical solutions of the above-mentioned method for monitoring the surface temperature of the battery is implemented.
[0034] In a third aspect, a battery swap station is provided, comprising a battery compartment and the server provided in the second aspect.
[0035] In a technical solution of the above-mentioned battery swap station, a non-contact temperature sensor is provided in the battery compartment, and the non-contact temperature sensor is used to collect a temperature map of the battery in the battery compartment, and the pixel value of each pixel in the temperature map respectively represents the actual temperature value at different positions on the battery surface.
[0036] In a fourth aspect, a computer-readable storage medium is provided, wherein a plurality of program codes are stored therein, wherein the program codes are suitable for being loaded and run by a processor to execute the method described in any one of the technical solutions of the above-mentioned method for monitoring the surface temperature of a battery.
[0037] Solution 1. A method for monitoring the surface temperature of a battery, characterized in that the method comprises:
[0038] Acquire multi-dimensional information of a first target battery in a battery compartment, where the first target battery is any battery in the battery compartment;
[0039] Adopting an LSTM model based on an attention mechanism and determining a surface temperature state of the first target battery according to the multi-dimensional information of the first target battery, wherein the surface temperature state is a normal surface temperature or an abnormal surface temperature;
[0040] in,
[0041] The multi-dimensional information includes surface temperature information and status information of the battery, and the status information includes power status and working status;
[0042] The LSTM model based on the attention mechanism is trained according to a first data sample and a second data sample, wherein the first data sample includes the multi-dimensional information of the battery when the surface temperature is normal, and the second data sample includes the multi-dimensional information of the battery when the surface temperature is abnormal.
[0043] Solution 2. The method according to Solution 1 is characterized in that the multi-dimensional information includes environmental information, and the environmental information includes the temperature and humidity of the battery compartment.
[0044] Solution 3. The method according to Solution 1 or 2 is characterized in that the multi-dimensional information includes surface temperature information of a second target battery, and the second target battery is the battery remaining in the battery compartment other than the first target battery.
[0045] Solution 4. The method according to Solution 1 is characterized in that a non-contact temperature sensor is provided in the battery compartment, the surface temperature information is the actual temperature value of the battery surface, and the method comprises obtaining the actual temperature value by:
[0046] Using the non-contact temperature sensor to collect a temperature map of the battery in the battery compartment;
[0047] The pixel value of each pixel in the temperature map represents the actual temperature value at different positions on the battery surface.
[0048] Solution 5. The method according to Solution 4 is characterized in that the surface temperature information is a temperature characteristic of the battery surface, and the method comprises obtaining the temperature characteristic by:
[0049] Clustering the pixels in the temperature map according to the pixel values of the pixels in the temperature map to obtain a plurality of pixel sets;
[0050] According to the pixel values of the pixels in each pixel set, respectively obtain the first temperature characteristics of each pixel set;
[0051] A second temperature characteristic of the surface temperature of the first target battery is acquired according to the first temperature characteristic of all pixel sets.
[0052] Solution 6. The method according to Solution 5 is characterized in that the step of obtaining the first temperature feature of each pixel set according to the pixel value of the pixel in each pixel set comprises:
[0053] According to the pixel value of each pixel in the pixel set, m1 maximum pixel values are obtained, wherein the m1 maximum pixel values are pixel values of pixels arranged from the 1st to the m1th pixels after the pixels are arranged in descending order according to the pixel values, and m1>1;
[0054] According to the deviation of each pixel in the pixel set, m2 maximum deviations are obtained, wherein the m2 maximum deviations are deviations of pixels arranged from the first to the m2th pixels after the pixels are arranged in descending order according to the deviations, and m2>1;
[0055] According to the temperature rise rate of each pixel in the pixel set, m3 maximum temperature rise rates are obtained, wherein the m3 maximum temperature rise rates are the temperature rise rates of the pixels arranged from the first to the m3th pixels after the pixels are arranged in descending order according to the temperature rise rates, and m3>1;
[0056] acquiring the first temperature feature according to the m1 largest pixel values, the m2 largest deviations and the m3 largest temperature rise rates;
[0057] in,
[0058] The deviation is the difference between the pixel value of the pixel and the pixel mean, and the pixel mean is the average value of the pixel values of all pixels in the pixel set;
[0059] The temperature rise rate is a rate of change of the pixel value of the pixel within a preset time.
[0060] Solution 7. The method according to Solution 1, characterized in that the method comprises:
[0061] Determine whether the multi-dimensional information is in a normal state according to a preset knowledge base; if it is in a normal state, adopt the LSTM model based on the attention mechanism, and determine the surface temperature state of the first target battery according to the multi-dimensional information of the first target battery;
[0062] The knowledge base includes normal features and abnormal features, wherein the normal features represent features when the multi-dimensional information is in a normal state, and the abnormal features represent features when the multi-dimensional information is in an abnormal state.
[0063] Solution 8. The method according to Solution 1 is characterized in that the method includes: if the surface temperature state of the first target battery is abnormal, an alarm is issued.
[0064] Solution 9. A server, comprising:
[0065] at least one processor;
[0066] and, a memory communicatively coupled to the at least one processor;
[0067] Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the method for monitoring the battery surface temperature described in any one of Schemes 1 to 8 is implemented.
[0068] Solution 10. A battery swap station, comprising a battery compartment, characterized in that the battery swap station comprises the server described in Solution 9.
[0069] Solution 11. The battery swap station according to Solution 10, characterized in that:
[0070] A non-contact temperature sensor is arranged in the battery compartment, and the non-contact temperature sensor is used to collect a temperature map of the battery in the battery compartment, and the pixel value of each pixel in the temperature map represents the actual temperature value at different positions on the battery surface.
[0071] Solution 12. A computer-readable storage medium storing a plurality of program codes, wherein the program codes are suitable for being loaded and run by a processor to execute the battery surface temperature monitoring method described in any one of Solutions 1 to 8.
[0072] The above one or more technical solutions of the present application have at least one or more of the following beneficial effects:
[0073] In a technical solution for implementing the battery surface temperature monitoring method provided in the present application, multi-dimensional information of a first target battery in a battery compartment can be obtained, where the first target battery is any battery in the battery compartment; an LSTM model based on an attention mechanism is used, and based on the multi-dimensional information of the first target battery, the surface temperature state of the first target battery is determined, where the surface temperature state is normal or abnormal; wherein the multi-dimensional information includes the surface temperature information and state information of the battery, and the state information includes the charge state and the working state; the LSTM model based on the attention mechanism is trained based on a first data sample and a second data sample, where the first data sample includes multi-dimensional information of the battery when the surface temperature is normal, and the second data sample includes multi-dimensional information of the battery when the surface temperature is abnormal.
[0074] Through the above implementation scheme, the above LSTM model can be used to utilize the multi-dimensional information of the battery to timely and accurately monitor whether the battery surface temperature is normal, so that timely measures can be taken when the battery surface temperature is abnormal to avoid thermal runaway of the battery. In addition, the above implementation scheme can reduce manual intervention and is more suitable for unmanned scenarios. The above implementation scheme has no restrictions on battery types and has high compatibility. At the same time, the above method can be upgraded by iteratively upgrading the LSTM model, so it has high upgrade convenience. BRIEF DESCRIPTION OF THE DRAWINGS
[0075] The disclosure of the present application will become easier to understand with reference to the accompanying drawings. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present application. Among them:
[0076] Figure 1 It is a schematic flow chart of the main steps of a method for monitoring the surface temperature of a battery according to an embodiment of the present application;
[0077] Figure 2 is a schematic flow chart of the main steps of obtaining the temperature characteristics of the battery surface according to an embodiment of the present application;
[0078] Figure 3 is a schematic diagram of an application scenario of a method for monitoring battery surface temperature according to an embodiment of the present application;
[0079] Figure 4 is a schematic diagram of the structure of an LSTM model based on an attention mechanism according to an embodiment of the present application;
[0080] Figure 5 It is a schematic diagram of the main structure of a server according to an embodiment of the present application.
[0081] Reference numerals:
[0082] 11: memory; 12: processor. DETAILED DESCRIPTION
[0083] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the protection scope of the present application.
[0084] In the description of this application, "processor" may include hardware, software or a combination of the two. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, hardware or a combination of the two. Computer-readable storage media include any suitable medium that can store program code, such as a disk, a hard disk, an optical disk, a flash memory, a read-only memory, a random access memory, etc.
[0085] The relevant user personal information that may be involved in the various embodiments of this application is strictly in accordance with the requirements of laws and regulations, following the principles of legality, legitimacy and necessity, based on the reasonable purposes of business scenarios, to process the personal information that users actively provide during the use of products / services or generated due to the use of products / services, as well as the personal information obtained with the user's authorization.
[0086] The user personal information processed by this application will vary depending on the specific product / service scenario, and shall be based on the specific scenario in which the user uses the product / service, and may involve the user's account information, device information, driving information, vehicle information or other related information. This application will treat the user's personal information and its processing with a high degree of diligence.
[0087] This application attaches great importance to the security of user personal information and has taken reasonable and feasible security protection measures that meet industry standards to protect user information and prevent personal information from being accessed, disclosed, used, modified, damaged or lost without authorization.
[0088] The following is an example of a method for monitoring the battery surface temperature provided by the present application. Figure 1 , Figure 1 FIG. 1 is a flow chart of the main steps of a method for monitoring the battery surface temperature according to an embodiment of the present application. Figure 1 As shown, the method for monitoring the battery surface temperature in the embodiment of the present application mainly includes the following steps S101 to S102.
[0089] Step S101: obtaining multi-dimensional information of a first target battery in a battery compartment, where the first target battery is any battery in the battery compartment.
[0090] The battery compartment can be used to store one or more batteries, and the battery can be a power battery.
[0091] The multi-dimensional information may include the surface temperature information and status information of the battery. The status information includes the charge status and the working status. The charge status is used to indicate the remaining charge of the battery. The working status may include the charging status and the static status. The charging status indicates that the battery is being charged. The static status indicates that the battery is static and is neither charged nor discharged.
[0092] Step S102: using an LSTM (Long Short-Term Memory) model based on an attention mechanism and according to the multi-dimensional information of the first target battery, determining the surface temperature state of the first target battery, where the surface temperature state is normal or abnormal.
[0093] In some embodiments, if the surface temperature state of the first target battery is abnormal, an alarm may be issued. Based on the alarm information, the user can quickly identify the battery with abnormal conditions and take timely measures to avoid battery thermal runaway.
[0094] The LSTM model based on the attention mechanism can be trained based on the first data sample and the second data sample, wherein the first data sample includes multi-dimensional information of the battery when the surface temperature is normal, and the second data sample includes multi-dimensional information of the battery when the surface temperature is abnormal. In this embodiment, a conventional model training method can be used to classify and train the above model using the first and second data samples, so that the trained model can predict the type of the surface temperature state based on the multi-dimensional information, whether the surface temperature is normal or abnormal.
[0095] When the above model is used to determine the surface temperature state of the first target battery, the multi-dimensional information of the first target battery can be input into the LSTM model based on the attention mechanism. The model can classify and predict the surface temperature state of the multi-dimensional information to obtain the type of the surface temperature state, which can be normal surface temperature or abnormal surface temperature.
[0096] The LSTM model based on the attention mechanism may include an input layer, an LSTM layer, a hidden layer, an attention layer, a fully connected layer, and an output layer. The input layer can be used to receive multi-dimensional information and input the multi-dimensional information into the LSTM layer; the LSTM layer can be used to obtain the hidden layer state corresponding to each dimensional information according to each dimensional information in the multi-dimensional information; the attention layer can be used to obtain the attention weight of the hidden layer state corresponding to each dimensional information; the fully connected layer can be used to perform a fully connected operation on the hidden layer state corresponding to all dimensional information according to the attention weight of the hidden layer state corresponding to each dimensional information to obtain a state fusion result; the output layer can predict the type of battery surface temperature state according to the state fusion result. By introducing the attention mechanism, the above attention weight can be used to dynamically adjust the attention to each dimensional information.
[0097] Based on the method described in steps S101 to S102 above, the LSTM model based on the attention mechanism and the multi-dimensional information of the battery can be used to timely and accurately monitor whether the battery surface temperature is normal, so that measures can be taken in time when the battery surface temperature is abnormal to avoid thermal runaway of the battery.
[0098] The above step S101 is described below.
[0099] In some implementations of the above step S101, the multi-dimensional information may also include environmental information, and the environmental information may include the temperature and humidity of the battery compartment. A temperature sensor and a humidity sensor may be provided in the battery compartment, and the temperature and humidity of the battery compartment are collected by the sensors. Based on this implementation, the surface temperature state of the battery can be comprehensively determined by using the information of the battery itself (indicating temperature information and state information) and the information of the environment in which the battery is located, which can further improve the accuracy of the determination result.
[0100] In some embodiments of the above step S101, the multi-dimensional information may also include surface temperature information of a second target battery, where the second target battery is the battery remaining in the battery compartment outside the first target battery. The method for acquiring the surface temperature information of the first and second target batteries is the same. Based on this embodiment, the surface temperature state of the battery can be comprehensively determined by using the information of the battery itself and the information of other batteries in the battery compartment, which can further improve the accuracy of the determination result. In some embodiments, the multi-dimensional information may simultaneously include the surface temperature information, state information, and environmental information of the first target battery and the surface temperature information of the second target battery, that is, the surface temperature state of the battery is determined by simultaneously integrating information of four dimensions.
[0101] In some embodiments of the above step S101, a non-contact temperature sensor is provided in the battery compartment, and the surface temperature information of the battery is the actual temperature value of the battery surface. In this embodiment, a non-contact temperature sensor can be used to collect a temperature map of the battery in the battery compartment, and the pixel value of each pixel in the temperature map represents the actual temperature value at different positions on the battery surface.
[0102] A non-contact temperature sensor is a sensor that can collect the temperature of an object without contacting the object. In this embodiment, the object is a battery. Since there is no contact with the object, the collected temperature can be understood as the surface temperature of the object. In this embodiment, a conventional non-contact temperature sensor can be used, for example, an infrared temperature sensor can be used.
[0103] In this embodiment, a non-contact temperature sensor with fast response and wide operating temperature range can be selected. Fast response is the ability to quickly sense changes in the temperature of objects within the detection range. The wide operating temperature range includes stable operation in an operating environment of -40 to 60 degrees Celsius, and the ability to accurately collect the surface temperature of objects from -50 to 300 degrees Celsius. In this embodiment, a non-contact temperature sensor is installed in the battery compartment, and it is necessary to ensure that the non-contact temperature sensor does not interfere with or affect the movement of the battery, and at the same time, it is necessary to ensure that the detection range covers the surface of the battery.
[0104] In some embodiments of the above step S101, the surface temperature information of the battery is the temperature characteristics of the battery surface. Figure 2 The following steps S1011 to S1013 are shown to obtain the temperature characteristics of the battery surface.
[0105] Step S1011: clustering the pixels in the temperature map according to their pixel values to obtain a plurality of pixel sets. Specifically, a conventional clustering method may be used to cluster the pixels according to their pixel values. The pixel set includes at least one pixel.
[0106] Step S1012: acquiring first temperature characteristics of each pixel set according to the pixel values of the pixels in each pixel set.
[0107] The pixel value represents the actual temperature value of the battery surface, and the first temperature feature of a pixel set can be obtained according to the distribution of the pixel values in the pixel set. For example, the maximum pixel value in the pixel set can be obtained, and the first temperature feature can be obtained according to the maximum pixel value.
[0108] Step S1013: Acquire a second temperature characteristic of the surface temperature of the first target battery according to the first temperature characteristics of all pixel sets. Specifically, the first temperature characteristics of all pixel sets are combined together as the second temperature characteristic.
[0109] Based on the method described in steps S1011 to S1013 above, the temperature characteristics of the battery surface can be obtained conveniently and accurately using the temperature map collected by the non-contact temperature sensor.
[0110] The above step S1012 is explained below.
[0111] In some implementations of the above step S1012, m1 maximum pixel values, m2 maximum deviations, and m3 maximum temperature rise rates can be obtained based on the pixels in the pixel set, and the first temperature feature can be obtained based on these data, and m1, m2, and m3 are all greater than 1. For example, the m1 maximum pixel values, the m2 maximum deviations, and the m3 maximum temperature rise rates are combined together as the first temperature feature. The method for obtaining the m1 maximum pixel values, the m2 maximum deviations, and the m3 maximum temperature rise rates is described below.
[0112] 1. Method for obtaining the m1 largest pixel values.
[0113] In this embodiment, m1 maximum pixel values can be obtained according to the pixel value of each pixel in the pixel set, and the m1 maximum pixel value is the pixel value arranged from the 1st to the m1th pixels after the pixels are arranged in descending order of pixel value.
[0114] Specifically, the m1 largest pixel values can be obtained by the following formula (1).
[0115]
[0116] The meanings of the parameters in formula (1) are as follows:
[0117] k represents the serial number of the pixel set, k = 1, ..., K, K represents the total number of pixel sets; t represents the acquisition time of the temperature map, D k,t represents the kth pixel set obtained from the temperature map obtained at the tth acquisition time, Dk,t ={T i,j,t}(k=1,2,...,K), i and j represent the row number and column number of the pixel in the temperature map respectively, T i,j,t represents the pixel in the i-th row and j-th column of the temperature map obtained at the t-th acquisition time; i 1 The serial number representing the maximum pixel value; Represents the i-th pixel obtained from the k-th pixel set 1 The maximum pixel value, Maximum means taking the maximum value function.
[0118] 2. Method for obtaining the m2 largest deviations.
[0119] In this embodiment, m2 maximum deviations can be obtained according to the deviations of each pixel in the pixel set, and the m2 maximum deviations are the deviations of the pixels arranged from the first to the m2th pixels after the pixels are arranged in descending order of deviation. The deviation is the difference between the pixel value of the pixel and the pixel mean, and the pixel mean is the average value of the pixel values of all pixels in the pixel set.
[0120] Specifically, the m2 largest deviations can be obtained by the following formula (2).
[0121]
[0122] The meanings of the parameters in formula (2) are as follows:
[0123] i 2 Indicates the number of the largest deviation, According to the k-th pixel set D k,t The i-th 2 The largest deviation; Devi k,t Denotes the pixel set D k,t The set of deviations of all pixels in k,t Compared with D in formula (1) k,t Same meaning.
[0124] Devi k,t ={Devi i,j,t} and Devi i,j,t =(T i,j,t -Tmean k,t ), T i,j,t Compared with T in formula (1) i,j,t Same meaning, Tmean k,t Indicates D k,t The average value of all pixels in Devi i,j,t Indicates T i,j,t Deviation, Devi k,t Indicates D k,tThe deviation set of all pixels in .
[0125] 3. The method of obtaining the maximum temperature rise rate of m3.
[0126] In this embodiment, m3 maximum temperature rise rates are obtained according to the temperature rise rate of each pixel in the pixel set, and the m3 maximum temperature rise rates are the temperature rise rates of the pixels arranged from the first to the m3th pixels after the pixels are arranged in descending order of temperature rise rate. The temperature rise rate is the rate of change of the pixel value of the pixel within a preset time.
[0127] Specifically, the m3 maximum temperature rise rates can be obtained by the following formula (3).
[0128]
[0129] The meanings of the parameters in formula (3) are as follows:
[0130] i 3 The serial number indicating the maximum rate of temperature rise, According to the k-th pixel set D k,t The i-th 3 The maximum temperature rise rate; Slope k,t Denotes the pixel set D k,t The collection of temperature rise rates of all pixels in D k,t Compared with D in formula (1) k,t Same meaning.
[0131] Slope k,t = {Slope i,j,t}and T i,j,t Compared with T in formula (1) i,j,t Same meaning, T i,j,t-Δt represents the pixel in the i-th row and j-th column of the temperature map obtained at the t-Δt-th acquisition time, where Δt represents the preset time; Slope i,j,t Indicates T i,j,t Temperature rise rate, Slope k,t Denotes the pixel set D k,t The shorter the preset time is, the finer the temperature rise rate is. When setting a specific value of the preset time, those skilled in the art can set the value of the preset time according to the degree of fineness of the temperature rise rate to be obtained.
[0132] The following is a description of an embodiment of the battery surface temperature monitoring method provided by the present application. In some embodiments according to the present application, it is possible to first determine whether the multi-dimensional information of the first target battery is in a normal state based on a preset knowledge base; if it is in a normal state, an LSTM model based on an attention mechanism is adopted, and the surface temperature state of the first target battery is determined based on the multi-dimensional information of the first target battery (i.e., executing step S102). If it is not in a normal state or is in an abnormal state, the above-mentioned multi-dimensional information can be output through the human-computer interaction interface so that the user can determine the surface temperature state of the first target battery based on these multi-dimensional information. In addition, the multi-dimensional information can also be re-acquired, and the surface temperature state of the first target battery can be determined based on the above-mentioned model and based on the re-acquired multi-dimensional information.
[0133] The knowledge base may include normal features and abnormal features. Normal features represent features when the multi-dimensional information is in a normal state, and abnormal features represent features when the multi-dimensional information is in an abnormal state. Taking ambient temperature as an example, the normal feature may be the temperature range when the ambient temperature is in a normal state, and the abnormal feature may be the temperature range when the ambient temperature is in an abnormal state. If the multi-dimensional information is in a normal state, it indicates that the credibility of the multi-dimensional information is relatively high, so the surface temperature state of the battery can be determined based on the above model and the multi-dimensional information; if the multi-dimensional information is in an abnormal state, it indicates that the credibility of the multi-dimensional information is relatively low. At this time, the surface temperature state of the battery can be determined manually based on the multi-dimensional information, or the multi-dimensional information can be re-acquired, and then the surface temperature state of the battery can be determined based on the above model and the multi-dimensional information. For example, when the battery compartment is set in different geographical areas, the temperature and humidity of the battery compartment will be quite different. If the environmental information is judged to be in an abnormal state according to the knowledge base, the user can first judge whether the environmental information is normal or abnormal according to the geographical area, and then comprehensively determine the surface temperature state of the battery based on the judgment result and combined with other dimensional information. In addition, the normal features and abnormal features in the knowledge base can be updated regularly to ensure the reliability of determining the surface temperature state of the battery based on the above model.
[0134] The following is combined with Figure 3 and attached Figure 4 , an embodiment of the method for monitoring the battery surface temperature provided by the present application is described. The method for monitoring the battery surface temperature according to the embodiment of the present application is applied to a battery swap station, which includes a battery compartment, which is used to store batteries. The batteries can also be charged in the battery compartment, and the batteries can be power batteries. In addition, a non-contact temperature sensor (i.e. Figure 3 The temperature sensor shown in the figure), the battery surface temperature monitoring method provided in the present application can be executed by a processor.
[0135] like Figure 3 As shown, the processor can communicate with the battery end and obtain the status information of the first target battery from the battery end; the processor can communicate with the cloud and obtain environmental information (including the temperature and humidity of the battery compartment) from the cloud; the temperature sensor can periodically collect the surface temperature of each battery in the battery compartment, and the processor can obtain the temperature characteristics of the first and second target batteries according to the collection results of the temperature sensor, and combine the above status information, environmental information, and the temperature characteristics of the first and second target batteries to form multi-dimensional information, and input the multi-dimensional information into the LSTM model based on the attention mechanism for classification and prediction to obtain the surface temperature state of the first target battery. Then, it is determined whether the surface temperature of the first target battery is abnormal according to the surface temperature state; if abnormal, an alarm signal is output; if normal, the above process is repeated. Among them, the battery end can be a battery management system of the battery, the cloud can be a cloud server, the battery swap station and the cloud server communicate with each other, the battery swap station can send its various status information to the cloud server, and the cloud server can store this information. The knowledge base has the same meaning and function as the knowledge base in the aforementioned method embodiment, and will not be repeated here.
[0136] like Figure 4 As shown, the LSTM model based on the attention mechanism may include an input layer, an LSTM layer, a hidden layer, an attention layer, a fully connected layer, and an output layer. Tmax1 represents the maximum pixel value in the second temperature feature, TDevi1 represents the maximum deviation in the second temperature feature, TSlope1 represents the maximum temperature rise rate in the second temperature feature, the environmental information includes the temperature and humidity of the battery compartment, and the status information includes the power status and working status of the first target battery.
[0137] LSTM1 is the LSTM layer based on the first dimension information ( Figure 4 is Tmax1), obtain the first dimension information corresponding to the hidden layer state h1; LSTMr is the LSTM layer according to the rth dimension information ( Figure 4 is state information), obtain the r-th dimension information corresponding to the hidden layer state hr; LSTM2, LSTM3, LSTM(r-1) are similar and will not be repeated. a1 is the attention weight of the first dimension information corresponding to the hidden layer state h1, ar is the attention weight of the r-th dimension information corresponding to the hidden layer state hr, a2, a3, a(r-1) have similar meanings and will not be repeated.
[0138] The above LSTM model based on the attention mechanism makes full use of multivariate information while ensuring the generalization ability of the model, so that it can be more reliably and accurately applied to complex scenarios such as battery swap stations to monitor the surface temperature status of the battery in a timely and accurate manner. In addition, the above model has the advantages of simple structure and lightweight, which is conducive to the deployment and operation of the model.
[0139] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art can understand that in order to achieve the effect of the present application, different steps do not have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders. These adjusted schemes are equivalent to the technical schemes described in this application, and therefore will also fall within the scope of protection of this application.
[0140] It is understood by those skilled in the art that all or part of the processes in the method for implementing the above-mentioned embodiment of the present application can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.
[0141] Another aspect of the present application also provides a computer-readable storage medium.
[0142] In an embodiment of a computer-readable storage medium according to the present application, the computer-readable storage medium can be configured to store a program for executing the battery surface temperature monitoring method of the above method embodiment, and the program can be loaded and run by the processor to implement the above battery surface temperature monitoring method. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present application. The computer-readable storage medium can be a storage device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-temporary computer-readable storage medium.
[0143] Another aspect of the present application also provides a server.
[0144] In an embodiment of a server according to the present application, the server may include at least one processor; and a memory connected to the at least one processor in communication; wherein a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the method described in any of the above embodiments is implemented. Figure 5 , Figure 5 FIG. 4 exemplarily shows that the memory 11 and the processor 12 are communicatively connected via a bus.
[0145] Another aspect of the present application provides a battery swap station.
[0146] In an embodiment of a battery swap station according to the present application, the battery swap station may include a battery compartment and the server described in the above embodiment. The battery compartment can be used to store batteries, and the batteries can also be charged in the battery compartment. The surface temperature state of each battery in the battery compartment can be effectively monitored by the above server. In some embodiments, a non-contact temperature sensor is provided in the battery compartment, and the non-contact temperature sensor is used to collect a temperature map of the batteries in the battery compartment, and the pixel value of each pixel in the temperature map represents the actual temperature value at different positions on the battery surface. For ease of explanation, only the parts related to the embodiments of the present application are shown. For specific technical details not disclosed, please refer to other embodiments of the present application.
[0147] So far, the technical solution of the present application has been described in conjunction with an embodiment shown in the accompanying drawings, but it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present application.
Claims
1. A method for monitoring battery surface temperature, characterized in that: The method comprises: Acquire multi-dimensional information of a first target battery in a battery compartment, where the first target battery is any battery in the battery compartment; Adopting an LSTM model based on an attention mechanism and determining a surface temperature state of the first target battery according to the multi-dimensional information of the first target battery, wherein the surface temperature state is a normal surface temperature or an abnormal surface temperature; in, The multi-dimensional information includes surface temperature information and status information of the battery, and the status information includes power status and working status; The LSTM model based on the attention mechanism is trained according to a first data sample and a second data sample, wherein the first data sample includes the multi-dimensional information of the battery when the surface temperature is normal, and the second data sample includes the multi-dimensional information of the battery when the surface temperature is abnormal.
2. The method according to claim 1, characterized in that The multi-dimensional information includes environmental information, and the environmental information includes the temperature and humidity of the battery compartment.
3. The method according to claim 1 or 2, characterized in that: The multi-dimensional information includes surface temperature information of a second target battery, where the second target battery is a battery remaining in the battery compartment other than the first target battery.
4. The method according to claim 1, characterized in that: A non-contact temperature sensor is provided in the battery compartment, the surface temperature information is the actual temperature value of the battery surface, and the method comprises obtaining the actual temperature value by: Using the non-contact temperature sensor to collect a temperature map of the battery in the battery compartment; The pixel value of each pixel in the temperature map represents the actual temperature value at different positions on the battery surface.
5. The method according to claim 4, characterized in that The surface temperature information is a temperature characteristic of the battery surface, and the method includes obtaining the temperature characteristic by: Clustering the pixels in the temperature map according to the pixel values of the pixels in the temperature map to obtain a plurality of pixel sets; According to the pixel values of the pixels in each pixel set, respectively obtain the first temperature characteristics of each pixel set; A second temperature characteristic of the surface temperature of the first target battery is acquired according to the first temperature characteristic of all pixel sets.
6. The method according to claim 5, characterized in that The obtaining the first temperature characteristics of each pixel set according to the pixel values of the pixels in each pixel set respectively includes: According to the pixel value of each pixel in the pixel set, m1 maximum pixel values are obtained, wherein the m1 maximum pixel values are pixel values of pixels arranged from the 1st to the m1th pixels after the pixels are arranged in descending order according to the pixel values, and m1>1; According to the deviation of each pixel in the pixel set, m2 maximum deviations are obtained, wherein the m2 maximum deviations are deviations of pixels arranged from the first to the m2th pixels after the pixels are arranged in descending order according to the deviations, and m2>1; According to the temperature rise rate of each pixel in the pixel set, m3 maximum temperature rise rates are obtained, wherein the m3 maximum temperature rise rates are the temperature rise rates of the pixels arranged from the first to the m3th pixels after the pixels are arranged in descending order according to the temperature rise rates, and m3>1; acquiring the first temperature feature according to the m1 largest pixel values, the m2 largest deviations and the m3 largest temperature rise rates; in, The deviation is the difference between the pixel value of the pixel and the pixel mean, and the pixel mean is the average value of the pixel values of all pixels in the pixel set; The temperature rise rate is a rate of change of the pixel value of the pixel within a preset time.
7. The method according to claim 1, characterized in that The method comprises: Determine whether the multi-dimensional information is in a normal state according to a preset knowledge base; if it is in a normal state, adopt the LSTM model based on the attention mechanism, and determine the surface temperature state of the first target battery according to the multi-dimensional information of the first target battery; The knowledge base includes normal features and abnormal features, wherein the normal features represent features when the multi-dimensional information is in a normal state, and the abnormal features represent features when the multi-dimensional information is in an abnormal state.
8. The method according to claim 1, characterized in that The method comprises: If the surface temperature state of the first target battery is abnormal, an alarm is issued.
9. A server, characterized in that: include: at least one processor; and, a memory communicatively coupled to the at least one processor; Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the method for monitoring the battery surface temperature according to any one of claims 1 to 8 is implemented.
10. A battery swap station, comprising a battery compartment, characterized in that: The battery swap station includes the server described in claim 9.
Citation Information
Patent Citations
Battery energy storage integrated system and method
CN112615070A
Power battery monitoring method and device, server and storage medium
CN113787914A
Thermal runaway early warning method and system for battery of battery swap station
CN114487856A
High-voltage battery thermal runaway early warning method and alarm method
CN115122921A
Battery temperature rise condition detection method and device and electronic equipment
CN115219916A
Cited By
Multi-dimensional temperature measurement method of organic electroluminescence display panel
CN119290198A
