Novel battery cell matching method capable of improving matching efficiency

Abstract

The invention discloses a novel battery cell matching method capable of improving matching efficiency, and relates to the technical field of battery cell matching, and the method comprises the following steps: collecting attribute data and aging data of battery cells, and carrying out data cleaning and formatting processing; constructing an aging prediction model by using a deep learning technology, inputting the processed data into the aging prediction model, and predicting a performance change value of the battery cell; the voltage and temperature of the battery cell are remotely collected, and the state of the battery cell is monitored in real time; constructing an optimization objective function based on the predicted performance change value and the voltage and temperature of the battery cell, quickly solving a grouping optimization problem by using a quantum computing technology, searching an optimal grouping scheme, performing battery cell grouping by using automatic equipment according to the optimal grouping scheme, and performing performance test; according to the method, the aging prediction model is constructed through the deep learning technology, the matching optimization problem is rapidly solved by using the quantum computing technology, the optimal matching scheme can be found in an extremely short time, and the matching efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery cell grouping, and in particular to a novel battery cell grouping method capable of improving grouping efficiency. Background Art

[0002] With the rapid development of the new energy industry, especially the rapid expansion of electric vehicles and energy storage systems, the demand for high-performance batteries is growing. As the core component of the battery, the performance of the battery cell directly affects the efficiency and life of the entire battery system. Therefore, how to efficiently group the battery cells to ensure the overall performance of the battery system is optimal has become one of the focuses of the industry. Traditional battery cell grouping methods usually rely on manual experience and judgment, which is not only time-consuming and labor-intensive, but also difficult to ensure the consistency and accuracy of the grouping. In recent years, with the advancement of artificial intelligence technology and automation technology, there have been many attempts to improve the efficiency of grouping through intelligent means, such as using machine learning algorithms to predict the aging characteristics of battery cells and using sensors for remote monitoring. However, these methods still have certain limitations, especially in terms of computational efficiency and accuracy when processing large-scale data sets.

[0003] Although there are some relatively advanced battery cell matching technologies on the market, they still face some challenges in practical applications. First, in terms of data processing, traditional methods are often unable to effectively handle a large amount of complex battery cell property data and aging data, especially the limited processing capacity for unstructured data. Secondly, in the matching optimization process, traditional optimization algorithms often have slow calculation speeds, especially when dealing with large-scale data sets. It is difficult to find the optimal matching solution in a short time. In addition, existing matching methods usually rely on contact measurement technology when monitoring battery cell status. This method not only increases the complexity of the system, but also may cause measurement errors. Finally, most existing technologies lack a comprehensive and efficient testing method when conducting performance tests, making it difficult to accurately evaluate the overall performance of the battery pack after matching. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a novel battery cell grouping method that can improve the grouping efficiency to solve the problem that in terms of data processing, traditional methods are often unable to effectively process a large amount of complex battery cell attribute data and aging data, especially the processing capacity of unstructured data is limited. Secondly, in the grouping optimization process, traditional optimization algorithms often have slow calculation speeds, especially when processing large-scale data sets, it is difficult to find the optimal grouping solution in a short time. In addition, the existing grouping methods usually rely on contact measurement technology when monitoring the battery cell status. This method not only increases the complexity of the system, but also may cause measurement errors. Finally, most of the existing technologies lack a comprehensive and efficient testing method when performing performance testing, making it difficult to accurately evaluate the overall performance of the battery pack after grouping.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a novel battery cell grouping method that can improve grouping efficiency, which includes:

[0008] Collect the attribute data and aging data of the battery cells, and perform data cleaning and formatting;

[0009] Use deep learning technology to build an aging prediction model, input the processed data into the aging prediction model, and predict the performance change value of the battery cell;

[0010] Remotely collect the voltage and temperature of the battery cells and monitor the status of the battery cells in real time;

[0011] Based on the predicted performance change value and the voltage and temperature of the battery cell, an optimization objective function is constructed, and quantum computing technology is used to quickly solve the group optimization problem and find the optimal group solution;

[0012] According to the optimal matching plan, use automated equipment to match the cells and conduct performance tests.

[0013] As a preferred solution of the novel battery cell grouping method capable of improving grouping efficiency according to the present invention, the steps of collecting the attribute data and aging data of the battery cells, and performing data cleaning and formatting are as follows:

[0014] Use sensors to collect capacity, internal resistance, voltage, cycle count and usage time from the battery production line;

[0015] The collected data set is integrated into the attribute data set x;

[0016] Preprocess the attribute data set;

[0017] The standardized data is integrated into the data matrix M, which is expressed as:

[0018]

[0019] Among them, C′ n1 represents the standardized capacity of the n1th cell, R′ n2 Represents the standardized internal resistance of the n2th cell, V′ n3 represents the standardized voltage of the n3th cell, L′ n4 represents the normalized cycle number of the n4th cell, T′ n5 It represents the normalized usage time of the n5th battery cell, and n1, n2, n3, n4, and n5 represent the index variables of capacity, internal resistance, voltage, number of cycles, and usage time, respectively.

[0020] As a preferred solution of the novel battery cell grouping method that can improve grouping efficiency described in the present invention, wherein: the aging prediction model is constructed using deep learning technology, and the processed data is input into the aging prediction model to predict the performance change value of the battery cell. The specific steps are:

[0021] The multi-layer perceptron model MLP is used as the neural network architecture of the aging prediction model;

[0022] Perform feature extraction on the data matrix M;

[0023] The maximum value among the multi-type features of the battery cell used is selected as the battery cell feature, i.e. F;

[0024] The cell feature F is input into the aging prediction model. The number of input layer nodes is equal to the dimension of feature F, and the number of output layer nodes is 1. The performance change value of the cell is predicted.

[0025] As a preferred solution of the novel battery cell grouping method capable of improving grouping efficiency described in the present invention, the voltage and temperature of the battery cell are collected remotely, and the state of the battery cell is monitored in real time. The specific steps are as follows:

[0026] Collect the distance between the transmitting coil and the receiving coil and the angular frequency of the alternating magnetic field generated by the transmitting coil, and calculate the voltage to the battery cell;

[0027] The infrared radiation intensity on the surface of the battery cell is captured by the infrared sensor to calculate the temperature of the battery cell;

[0028] Perform time series analysis on the collected output voltage and output temperature to monitor the working status of the battery cell;

[0029] The transmitting coil generates an alternating magnetic field and eddy currents in the battery cell. The induced electromotive force detected by the receiving coil reflects the actual voltage of the battery cell.

[0030] The transmitting coil and the receiving coil work together to achieve remote contactless measurement of the battery cell voltage.

[0031] As a preferred solution of the novel battery cell grouping method for improving grouping efficiency described in the present invention, wherein: the use of quantum computing technology to quickly solve the grouping optimization problem and find the optimal grouping solution is specifically as follows:

[0032] The quantum annealing algorithm is used to calculate the optimization objective function g.

[0033] As a preferred solution of the novel battery cell grouping method for improving grouping efficiency described in the present invention, wherein: the battery cell grouping is performed using automated equipment according to the optimal grouping solution, including:

[0034] According to the optimal matching scheme parameter Q opt Determine the order of cells;

[0035] Use the robotic arm to grab the cells in order according to the order of the cells;

[0036] The visual system is used to monitor the position and status of the battery cells, so that the robot arm can place the battery cells;

[0037] After the assembly is completed, quality inspection is carried out.

[0038] As a preferred solution of the novel battery cell grouping method capable of improving grouping efficiency described in the present invention, the performance test refers to calculating the capacity performance of the battery cell based on the collected capacity.

[0039] In a second aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the new battery cell grouping method that can improve the grouping efficiency as described in the first aspect of the present invention is implemented.

[0040] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the new battery cell grouping method that can improve the grouping efficiency as described in the first aspect of the present invention is implemented.

[0041] The beneficial effects of the present invention are as follows: the present invention constructs an aging prediction model through deep learning technology, and uses quantum computing technology to quickly solve the matching optimization problem, so that the optimal matching solution can be found in a very short time, thereby improving the matching efficiency. Secondly, by remotely collecting the voltage and temperature of the battery cell and monitoring the status of the battery cell in real time, not only the error caused by contact measurement is reduced, but also the working condition of the battery cell can be fed back in time, which is helpful to prevent potential problems in advance. Finally, the present invention also includes a performance testing link, which ensures that the performance of the battery pack reaches the optimal level after matching through precise testing methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0043] Figure 1 This is a flow chart of a novel battery cell grouping method that can improve grouping efficiency in Example 1.

[0044] Figure 2 This is a flow chart of a novel battery cell grouping method that can improve grouping efficiency in Example 1. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0046] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0047] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0048] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a new battery cell grouping method that can improve grouping efficiency, comprising the following steps:

[0049] S1 collects the attribute data and aging data of the battery cells, and performs data cleaning and formatting;

[0050] Use sensors to collect capacity, internal resistance, voltage, cycle count and usage time from the battery production line;

[0051] The collected data set is integrated into the attribute data set x;

[0052] Preprocess the attribute data set, including data cleaning and formatting;

[0053] The standard deviation method SD is used to identify and remove outliers that are beyond the normal range. The expression is:

[0054]

[0055] Where N represents the number of samples, x p represents the pth sample data point, and p represents the index variable;

[0056] Use the standardization method to normalize the cleaned data to the [0,1] interval;

[0057] The standardized data is integrated into the data matrix M, which is expressed as:

[0058]

[0059] Among them, C′ n1 represents the standardized capacity of the n1th cell, R′ n2 Represents the standardized internal resistance of the n2th cell, V′ n3 represents the standardized voltage of the n3th cell, L′ n4 represents the normalized cycle number of the n4th cell, T′ n5 represents the normalized usage time of the n5th battery cell, and n1, n2, n3, n4, and n5 represent the index variables of capacity, internal resistance, voltage, number of cycles, and usage time, respectively;

[0060] Attribute data, including cell capacity, internal resistance, and voltage;

[0061] Aging data, including the number of cycles and usage time of the battery cell.

[0062] S2 uses deep learning technology to build an aging prediction model, inputs the processed data into the aging prediction model, and predicts the performance change value of the battery cell;

[0063] The multi-layer perceptron model MLP is used as the neural network architecture of the aging prediction model;

[0064] The feature extraction of data matrix M is expressed as:

[0065]

[0066] Among them, F i Indicates the characteristic combination value of battery cell i. The larger the value, the better the overall performance of the battery cell. i , R′ i , V′ i , L′ i and T′ i Respectively represent the standardized capacity, internal resistance, voltage, number of cycles and usage time of battery cell i, C′max , R' max , V' max , L' max and T' max respectively represent the maximum values of the standardized capacity, internal resistance, number of voltage cycles, and service time among all the battery cells;

[0067] Select the maximum value among multiple types of characteristics of the used battery cells as the battery cell characteristic, that is, F;

[0068] Input the battery cell characteristic F into the aging prediction model. The number of nodes in the input layer is equal to the dimension of the characteristic F, and the number of nodes in the output layer is 1 to predict the performance change value of the battery cell. The expression is:

[0069] P = f(F; θ);

[0070] where P represents the predicted performance change value, f represents the multi-layer perceptron model MLP, and θ represents the model parameters;

[0071] Use the mean square error as the loss function to measure the gap between the predicted performance change value and the actual performance change value. The expression is:

[0072]

[0073] where MSE represents the mean square error value, N represents the number of samples, P j represents the predicted performance change value of the jth battery cell, and P a,j represents the actual performance change value of the jth battery cell;

[0074] Set the threshold PY;

[0075] The value range of the mean square error value MSE is from 0 to positive infinity. The larger the value of MSE, the greater the difference between the predicted value and the actual value; the smaller the value of MSE, the smaller the difference between the predicted value and the actual value;

[0076] When 0 < MSE ≤ PY, it indicates that the performance of the aging prediction model is qualified;

[0077] When PY < MSE, it indicates that the performance is unqualified. Continuously collect data and input it into the aging prediction model for training and prediction until 0 < MSE ≤ PY.

[0078] S3 Remotely collect the voltage and temperature of the battery cell and monitor the status of the battery cell in real time;

[0079] Collect the distance between the transmitting coil and the receiving coil and the angular frequency of the alternating magnetic field generated by the transmitting coil, and calculate the voltage to the battery cell. The expression is:

[0080]

[0081] Among them, V out Represents the output voltage, V cell represents the voltage of the battery cell, d represents the distance between the transmitting coil and the receiving coil, ω represents the angular frequency of the alternating magnetic field, K v It represents the proportionality coefficient between the output voltage and the actual voltage, and π represents the circumference of a circle;

[0082] The infrared radiation intensity on the surface of the battery cell is captured by the infrared sensor, and the temperature of the battery cell is calculated. The expression is:

[0083] T out =K t (I rad e -αdI +C t );

[0084] Among them, T out Indicates the output temperature, I rad Indicates the infrared radiation intensity, dI indicates the distance between the infrared sensor and the surface of the battery cell, K t represents the gain coefficient in temperature conversion, α represents the attenuation coefficient, C t Indicates the temperature correction value;

[0085] Perform time series analysis on the collected output voltage and output temperature to monitor the working status of the battery cell. The expression is:

[0086]

[0087] Among them, R a (t) represents the time series analysis result, which is a vector containing the average voltage and average temperature in a given time interval, t f Indicates the end time of the time series analysis, t r Indicates the start time of the time series analysis, V out (t) represents the output voltage detected by the receiving coil at time t, T out (t) represents the output temperature detected by the receiving coil at time t, and dt represents the small time interval during the integration process;

[0088] The transmitting coil generates an alternating magnetic field and eddy currents in the battery cell. The induced electromotive force detected by the receiving coil reflects the actual voltage of the battery cell.

[0089] The transmitting coil and the receiving coil work together to achieve remote contactless measurement of the battery cell voltage;

[0090] The transmitting coil generates an alternating magnetic field by energizing the coil, which will generate eddy currents in the battery core close to the coil, thereby generating an induced electromotive force inside the battery core;

[0091] Receiving coil, when the alternating magnetic field generated by the transmitting coil passes through the receiving coil, an induced electromotive force is generated in the receiving coil. The magnitude of this electromotive force is related to the eddy current generated inside the battery cell, and thus to the voltage of the battery cell;

[0092] Output voltage V out The value range is from 0 to the maximum cell voltage value. out When it is equal to 0, it means that the battery has no voltage. out When it is equal to the maximum cell voltage value, it means that the cell voltage has reached the maximum value;

[0093] Output temperature T out The value range is from the minimum to the maximum value of the battery cell operating temperature, T out The larger the value, the higher the battery cell temperature, and vice versa.

[0094] S4 builds an optimization objective function based on the predicted performance change value and the voltage and temperature of the battery cell, and uses quantum computing technology to quickly solve the grouping optimization problem and find the optimal grouping solution;

[0095] The optimization objective function is constructed based on the predicted performance change value and the voltage and temperature of the battery cell. The expression is:

[0096]

[0097] Among them, g represents the value of the optimization objective function, Z represents the normalization factor, which is used to ensure the numerical stability of the optimization objective function and prevent numerical instability caused by large differences in the values ​​of each part, and P ref Indicates the preset value of the performance change value, V ref Indicates the preset value of voltage, T ref Indicates the preset value of temperature, Q p represents the penalty term;

[0098] The normalization factor Z is expressed as:

[0099]

[0100] Among them, w P 、w V and w T Weight coefficients representing performance variation values, voltage, and temperature, which can be set according to their importance;

[0101] Set the average operating temperature of the battery cell to not exceed the safety threshold T safe ;

[0102] Penalty term Q p , the expression is:

[0103]

[0104] Among them, T avg Indicates the average operating temperature of the battery pack after assembly, T safe Indicates the preset safety temperature threshold;

[0105] When the average temperature is lower than or equal to the safety threshold, the penalty term Q p =0, which means no penalty will be incurred due to temperature during the optimization process;

[0106] When the average temperature exceeds the safety threshold, the penalty term Q p It will increase in proportion to the degree of excess, causing the value of the optimization objective function g to decrease, thus making the matching method less optimal;

[0107] The quantum annealing algorithm is used to calculate the optimization objective function g, and the expression is:

[0108]

[0109] Among them, Q opt represents the optimal matching scheme parameters, It means finding the parameter Q that minimizes the objective function g, where Q represents the optimization parameter;

[0110] When Q opt When the corresponding objective function g reaches the minimum value, it means that the optimal matching solution has been found;

[0111] The larger the value of the objective function g is, the worse the quality of the matching scheme is.

[0112] S5 uses automated equipment to group cells according to the optimal grouping plan and conducts performance tests;

[0113] According to the optimal matching scheme parameter Q opt Determine the order of cells;

[0114] Use the robotic arm to grab the cells in order according to the order of the cells;

[0115] The visual system is used to monitor the position and status of the battery cells, so that the robot arm can place the battery cells;

[0116] After the assembly is completed, quality inspection is carried out;

[0117] Performance testing means calculating the capacity performance of the battery cell based on the collected capacity, and the expression is:

[0118]

[0119] Among them, η C Indicates the capacity retention rate of the battery cell, C initial represents the initial capacity, Cfinal Indicates the final capacity, C threshold Indicates the threshold capacity. When the cell capacity is lower than this value, it is considered no longer suitable for use. It can be set according to user needs.

[0120] η C The value range of is [0,1];

[0121] When η C When it is equal to 1, it means that the capacity of the battery cell is well maintained and there is almost no capacity degradation;

[0122] When η C When it is equal to 0, it means that the capacity of the battery cell is severely degraded.

[0123] This embodiment also provides a computer device, which is suitable for the case of a new battery cell grouping method that can improve the grouping efficiency, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the new battery cell grouping method that can improve the grouping efficiency as proposed in the above embodiment.

[0124] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0125] The present embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements a novel battery cell grouping method that can improve grouping efficiency as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0126] In summary, the present invention uses: deep learning technology to construct an aging prediction model, and uses quantum computing technology to quickly solve the matching optimization problem, so that the optimal matching solution can be found in a very short time, thereby improving the matching efficiency; secondly, by remotely collecting the voltage and temperature of the battery cell and monitoring the status of the battery cell in real time, not only the error caused by contact measurement is reduced, but also the working condition of the battery cell can be fed back in time, which is helpful to prevent potential problems in advance; finally, the present invention also includes a performance testing link, which ensures that the performance of the battery pack reaches the optimal level after matching through precise testing methods.

[0127] Example 2, referring to Table 1, is the second example of the present invention. To further verify the advancement of the present invention, experimental simulation data of a new cell grouping method that can improve grouping efficiency is provided.

[0128] In order to verify the effectiveness of the battery cell matching method proposed in the present invention, the following experiment was designed. Four battery cell samples were used for testing. These battery cells have different properties such as capacity, internal resistance, voltage, number of cycles and usage time. The experiment was divided into the following stages:

[0129] Use sensors to collect attribute data and aging data of battery cells from the battery cell production line, including capacity, internal resistance, voltage, number of cycles and usage time, integrate the collected data into an attribute data set x, preprocess the attribute data set, including data cleaning and formatting, use the standard deviation method SD to identify and remove outliers that exceed the normal range, use the standardization method to normalize the cleaned data to the [0,1] interval, and integrate the standardized data into a data matrix;

[0130] The multi-layer perceptron model MLP is used as the neural network architecture of the aging prediction model to extract features from the data matrix, construct cell features, input the cell features into the aging prediction model, and predict the performance change value of the cell. The mean square error is used as the loss function to measure the difference between the predicted performance change value and the actual performance change value, and the threshold is set;

[0131] Through the non-contact measurement technology between the transmitting coil and the receiving coil, the voltage and temperature of the battery cell are remotely collected, and the status of the battery cell is monitored in real time. The output voltage and output temperature of the battery cell are calculated, and the collected output voltage and output temperature are analyzed in time series to monitor the working status of the battery cell;

[0132] Based on the predicted performance change value and the voltage and temperature of the battery cell, an optimization objective function is constructed, and quantum computing technology is used to quickly solve the grouping optimization problem, find the optimal grouping solution, set the average operating temperature of the battery cell not to exceed the safety threshold, and define the penalty term;

[0133] According to the optimal matching scheme, use automated equipment to match the battery cells. After the matching is completed, perform performance tests and calculate the capacity retention rate of the battery cells.

[0134] The details are shown in Table 1:

[0135] Table 1 Experimental record table

[0136] Test subjects Initial capacity Final capacity Threshold capacity Capacity retention rate Battery 1 100 85 50 0.9 Battery 2 100 70 50 0.4 Battery 3 100 92 50 0.94 Battery 4 100 78 50 0.64

[0137] The analysis of the experimental data shows that the capacity retention rates of cells 1 and 3 are relatively high, which are 0.9 and 0.94 respectively, indicating that the capacity of these two cells is well maintained after a certain number of charge and discharge cycles, and there is almost no capacity degradation. In contrast, the capacity retention rates of cells 2 and 4 are relatively low, which are 0.4 and 0.64 respectively, indicating that the capacity degradation of these two cells is relatively serious.

[0138] Compared with the traditional manual grouping method, the battery cell grouping method proposed in the present invention shows obvious innovation and advantages in the following aspects:

[0139] Data processing capability: The present invention constructs an aging prediction model through deep learning technology, which can effectively process a large amount of battery cell attribute data and aging data, and improve the accuracy and efficiency of data processing;

[0140] Optimization algorithm: Using quantum computing technology to quickly solve the grouping optimization problem, it is possible to find the optimal grouping solution in a very short time, greatly improving the grouping efficiency;

[0141] Remote monitoring: Through non-contact measurement technology between the transmitting coil and the receiving coil, the status of the battery cell can be monitored in real time, reducing the error caused by contact measurement;

[0142] Performance test: By calculating the capacity retention rate η_C of the battery cell, the capacity degradation of the battery cell can be accurately evaluated to ensure that the overall performance of the battery pack is optimal after assembly;

[0143] Through the above analysis, it can be seen that the battery cell grouping method proposed in the present invention has significant advantages and innovations in improving grouping efficiency and ensuring battery pack performance, and provides new ideas and methods for the development of battery cell grouping technology.

[0144] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A novel battery cell assembly method capable of improving assembly efficiency, characterized in that: include, Collect the attribute data and aging data of the battery cells, and perform data cleaning and formatting; Use deep learning technology to build an aging prediction model, input the processed data into the aging prediction model, and predict the performance change value of the battery cell; Remotely collect the voltage and temperature of the battery cells and monitor the status of the battery cells in real time; Based on the predicted performance change value and the voltage and temperature of the battery cell, an optimization objective function is constructed, and quantum computing technology is used to quickly solve the group optimization problem and find the optimal group solution; According to the optimal matching plan, use automated equipment to match the cells and conduct performance tests.

2. The novel battery cell assembly method capable of improving assembly efficiency as claimed in claim 1, characterized in that: The method of collecting the property data and aging data of the battery cell and performing data cleaning and formatting processing comprises the following specific steps: Use sensors to collect capacity, internal resistance, voltage, cycle count and usage time from the battery production line; The collected data set is integrated into the attribute data set x; Preprocess the attribute data set; The standardized data is integrated into the data matrix M, which is expressed as: Among them, C′ n1 represents the standardized capacity of the n1th cell, R′ n2 Represents the standardized internal resistance of the n2th cell, V′ n3 represents the standardized voltage of the n3th cell, L′ n4 represents the normalized cycle number of the n4th cell, T′ n5 It represents the normalized usage time of the n5th battery cell, and n1, n2, n3, n4, and n5 represent the index variables of capacity, internal resistance, voltage, number of cycles, and usage time, respectively.

3. The novel battery cell assembly method capable of improving assembly efficiency as claimed in claim 2, characterized in that: The aging prediction model is constructed by using deep learning technology, and the processed data is input into the aging prediction model to predict the performance change value of the battery cell. The specific steps are: The multi-layer perceptron model MLP is used as the neural network architecture of the aging prediction model; Perform feature extraction on the data matrix M; The maximum value among the multi-type features of the battery cell used is selected as the battery cell feature, i.e. F; The cell feature F is input into the aging prediction model. The number of input layer nodes is equal to the dimension of feature F, and the number of output layer nodes is 1. The performance change value of the cell is predicted.

4. The novel battery cell assembly method capable of improving assembly efficiency as claimed in claim 3, characterized in that: The remote collection of the voltage and temperature of the battery cell and real-time monitoring of the status of the battery cell are specifically performed as follows: Collect the distance between the transmitting coil and the receiving coil and the angular frequency of the alternating magnetic field generated by the transmitting coil, and calculate the voltage to the battery cell; The infrared radiation intensity on the surface of the battery cell is captured by the infrared sensor to calculate the temperature of the battery cell; Perform time series analysis on the collected output voltage and output temperature to monitor the working status of the battery cell; The transmitting coil generates an alternating magnetic field and eddy currents in the battery cell. The induced electromotive force detected by the receiving coil reflects the actual voltage of the battery cell. The transmitting coil and the receiving coil work together to achieve remote contactless measurement of the battery cell voltage.

5. The novel battery cell assembly method capable of improving assembly efficiency as claimed in claim 4, characterized in that: The method of using quantum computing technology to quickly solve the grouping optimization problem and find the optimal grouping solution is specifically as follows: The quantum annealing algorithm is used to calculate the optimization objective function g.

6. The novel battery cell assembly method capable of improving assembly efficiency as claimed in claim 5, characterized in that: The method of using automated equipment to perform cell grouping according to the optimal grouping scheme includes: According to the optimal matching scheme parameter Q opt Determine the order of cells; Use the robotic arm to grab the cells in order according to the order of the cells; The visual system is used to monitor the position and status of the battery cells, so that the robot arm can place the battery cells; After the assembly is completed, quality inspection is carried out.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the novel battery cell grouping method capable of improving grouping efficiency as described in any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the novel battery cell grouping method capable of improving grouping efficiency as described in any one of claims 1 to 6 are implemented.

Images