A battery life prediction method and system based on data analysis

Abstract

The present application relates to the technical field of battery life prediction, and specifically to a battery life prediction method and system based on data analysis. The method includes: collecting data values ​​of remaining power, surface temperature, and different parameters; calculating the power consumption rate at each moment through the remaining power, and obtaining the load consumption according to the time difference between the maximum and maximum values ​​of the power consumption rate and the difference in power consumption rate; dividing the intervals according to the surface temperature, and obtaining the temperature difference coefficient according to the difference and duration between the corresponding temperature of each interval and the optimal working temperature; obtaining the battery damage coefficient based on the load consumption and temperature difference coefficient; obtaining the remaining capacity based on this combined parameter as a neural network input, and completing the prediction of the remaining battery life based on the comparison of the remaining capacity with the preset failure threshold. The present application improves the prediction accuracy of the remaining service life of the battery.

Description

Technical Field

[0001] The present application relates to the technical field of battery life prediction, and specifically to a battery life prediction method and system based on data analysis. Background Art

[0002] As human society pays more and more attention to energy conservation and environmental protection, new energy plays a more important role in life and industrial structure. Lithium batteries have gradually become the mainstream power source choice in electric vehicles, energy storage systems and other fields due to their high energy density, long cycle life, low self-discharge rate and no memory effect.

[0003] During the continuous charge and discharge process of lithium batteries, battery performance will deteriorate as capacity decreases and impedance increases, which can easily lead to equipment and system failures. Therefore, it is also very important to accurately estimate battery health and predict remaining life. With the development of the battery industry, battery life prediction technology is also constantly improving, and a variety of battery life prediction methods have been extended. For example, the data-driven method mines battery degradation information and the evolution of health status through historical data. However, the battery life prediction method based on data-driven energy storage systems usually only considers the battery operating parameters and the battery usage environment, but ignores the impact of the actual operating status of the battery on the remaining battery life, which makes the prediction accuracy of the remaining battery life in practical applications low. Summary of the invention

[0004] In order to solve the technical problem of low battery life prediction accuracy, the present application provides a battery life prediction method and system based on data analysis. The technical solutions adopted are as follows:

[0005] In a first aspect, the present application proposes a battery life prediction method based on data analysis, the method comprising the following steps:

[0006] Collecting the remaining power, surface temperature and data values ​​of different parameters at each moment in the charge and discharge cycle; the parameters include battery temperature, current and voltage;

[0007] All the remaining power in the charge and discharge cycle is formed into a sequence, and the power consumption rate is obtained according to the difference of the remaining power in the sequence, and a power consumption rate sequence is formed; the maximum value in the power consumption rate sequence is selected; the consumption intensity is obtained according to the difference between the power consumption rate in the charge and discharge cycle and the average power consumption rate; the load consumption of the battery in the charge and discharge cycle is obtained according to the consumption intensity in the charge and discharge cycle, the number of maximum values, and the time difference between adjacent maximum values;

[0008] The time corresponding to the surface temperature is divided into intervals according to the surface temperature, and the temperature of the interval is represented by a surface temperature in the interval, and the temperature difference coefficient of the battery in the charge and discharge cycle is obtained according to the difference between the temperature of the interval and the preset optimal working temperature and the time of the interval;

[0009] Obtaining a battery damage coefficient of the battery in the charge and discharge cycle according to the temperature difference coefficient and consumption intensity of the battery in the charge and discharge cycle;

[0010] The average values ​​of the battery's temperature, current, and voltage during the charge and discharge cycle and the battery damage coefficient are used as neural network inputs to obtain the remaining capacity of all subsequent charge and discharge cycles; the remaining battery life is predicted based on the comparison between the remaining capacity and the preset failure threshold.

[0011] In the above scheme, the present application collects battery operating parameter data obtained when the electronic device is running based on the remaining power when the electronic device is running, and constructs load consumption by analyzing the high-load behavior of the electronic device when it is working. The beneficial effect is that the load consumption takes into account the cumulative damage caused to the remaining battery life by frequent high-load behavior in the working process of the electronic device, and the negative impact of the high-load operation behavior of the electronic device on the battery state can be well reflected according to the load consumption; secondly, based on the change in the surface temperature of the electronic device during the operation of the electronic device and combined with the load consumption, a battery damage coefficient is constructed. The beneficial effect is that the battery damage coefficient fully considers the relationship between the chemical reaction rate in the battery and the ambient temperature, as well as the damage to the battery life caused by the high-load operation behavior of the electronic device under extreme ambient temperatures. The battery damage coefficient is used as a weight and combined with the temperature, current, and voltage data of the battery during operation in each charge and discharge cycle. It can more comprehensively capture the battery loss in actual applications, thereby improving the prediction accuracy of the remaining battery life.

[0012] In one embodiment, the method of obtaining the power consumption rate according to the difference of the remaining power in the sequence and forming the power consumption rate sequence is:

[0013] Record any moment as the target moment, obtain a sliding window of preset size with the target moment as the center, take the ratio of the change in the remaining power in the sliding window to the consumption time as the power consumption rate at the target moment, and sort all power consumption rates by time to obtain a power consumption rate sequence.

[0014] In one embodiment, the method for obtaining the consumption intensity according to the difference between the power consumption rate in the charge and discharge cycle and the average power consumption rate is:

[0015] The expression of consumption intensity is: , represents the power consumption rate at the rth moment, It represents the average rate of power consumption during the charge and discharge cycle. represents the number of collected moments, Indicates the intensity of battery consumption during the charge and discharge cycle.

[0016] In one embodiment, the method for obtaining the load consumption of the battery in the charge and discharge cycle according to the consumption intensity, the number of maximum values, and the time difference between adjacent maximum values ​​in the charge and discharge cycle is:

[0017] Load consumption is positively correlated with consumption intensity and the number of maxima, and negatively correlated with the time difference between adjacent maxima.

[0018] In one embodiment, the method of dividing the time corresponding to the surface temperature into intervals according to the surface temperature and representing the temperature of the interval by a surface temperature in the interval is:

[0019] Set the temperature range size. All surface temperatures are divided into intervals according to the temperature range size. All surface temperatures in each temperature range size constitute an interval. The interval is closed on the left and open on the right. A temperature in the interval is used to represent the temperature of the interval.

[0020] In one embodiment, the method for obtaining the temperature difference coefficient of the battery in the charge and discharge cycle according to the difference between the temperature of the interval and the preset optimal operating temperature and the time of the interval is:

[0021] The expression of temperature difference coefficient is: , represents the temperature of the qth interval, is the optimal operating temperature of the battery, represents the working time of the qth interval, represents the number of intervals, Indicates the temperature variation coefficient of the battery during the charge and discharge cycle.

[0022] In one embodiment, the method for obtaining the battery damage coefficient of the battery in the charge and discharge cycle according to the temperature difference coefficient and consumption intensity of the battery in the charge and discharge cycle is:

[0023] The temperature difference coefficient and the consumption intensity are linearly normalized respectively, and the average of the normalized temperature difference coefficient and the normalized consumption intensity within a charge and discharge cycle is taken as the battery damage coefficient of the battery within the charge and discharge cycle.

[0024] In one embodiment, the method for predicting the remaining life of the battery based on the comparison between the remaining capacity and the preset failure threshold is:

[0025] The remaining capacity of the current charge and discharge cycle and all subsequent charge and discharge cycles is obtained through a neural network. Among all the input energies, the first charge and discharge cycle with a remaining capacity less than 80% is selected as the end cycle, and the number of charge and discharge cycles from the current charge and discharge cycle to the end cycle is counted as the remaining battery life.

[0026] In one embodiment, the maximum value is the power consumption rate at a certain moment when the power consumption rate at that moment is greater than the power consumption rates at adjacent moments on both sides.

[0027] In the second aspect, an embodiment of the present application also provides a battery life prediction system based on data analysis, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned battery life prediction methods based on data analysis are implemented.

[0028] The beneficial effects of this application are:

[0029] This application collects battery operating parameter data obtained when the electronic device is running, based on the remaining power when the electronic device is running, and constructs load consumption by analyzing the high-load behavior of the electronic device when it is working. The beneficial effect is that the load consumption takes into account the cumulative damage caused to the remaining battery life by frequent high-load behavior in the working process of the electronic device, and the negative impact of the high-load operation behavior of the electronic device on the battery state can be well reflected according to the load consumption; secondly, based on the change in the surface temperature of the electronic device during the operation of the electronic device and combined with the load consumption, a battery damage coefficient is constructed. The beneficial effect is that the battery damage coefficient fully considers the relationship between the chemical reaction rate in the battery and the ambient temperature, as well as the damage to the battery life caused by the high-load operation behavior of the electronic device under extreme ambient temperatures. The battery damage coefficient is used as a weight and combined with the temperature, current, and voltage data of the battery during operation in each charge and discharge cycle. It can more comprehensively capture the battery loss in actual applications, thereby improving the prediction accuracy of the remaining battery life. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0031] Figure 1 A flow chart of a battery life prediction method based on data analysis provided in one embodiment of the present application;

[0032] Figure 2The present invention is a flowchart for implementing a battery life prediction method based on data analysis. DETAILED DESCRIPTION

[0033] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the battery life prediction method and system based on data analysis proposed in the present application, its specific implementation, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0034] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0035] An embodiment of a battery life prediction method based on data analysis:

[0036] The following is a detailed description of a battery life prediction method based on data analysis provided by the present application in conjunction with the accompanying drawings.

[0037] See also Figure 1 , which shows a flow chart of a battery life prediction method based on data analysis provided by an embodiment of the present application, the method comprising the following steps:

[0038] Step S001, collecting data values ​​of the remaining power, surface temperature and different parameters at each moment in the charge and discharge cycle.

[0039] The remaining battery life of an electronic device is not only related to the operating parameters of the battery itself, but also closely related to the operating status of the electronic device. This application uses temperature sensors, current sensors, and voltage sensors to obtain the operating parameters of the electronic device in each charge and discharge cycle, including the temperature, current, and voltage of the battery during operation, and obtains the remaining power of the electronic device during operation and the surface temperature data of the electronic device through the battery management system.

[0040] In the present application, the data sampling interval is t seconds, and the total amount of data collected for each parameter in each charge and discharge cycle is N. In this embodiment, the sampling interval t is 1 second, and the total amount of data collected for each parameter N is 10,000.

[0041] At this point, the data values ​​of different parameters at each moment, as well as the remaining power and surface temperature at each moment, are obtained.

[0042] Step S002, forming a sequence of all the remaining power in the charge and discharge cycle, obtaining the power consumption rate according to the difference in the remaining power in the sequence, and forming a power consumption rate sequence; selecting the maximum value in the power consumption rate sequence; obtaining the consumption intensity according to the difference between the power consumption rate in the charge and discharge cycle and the average power consumption rate; obtaining the load consumption of the battery in the charge and discharge cycle according to the consumption intensity in the charge and discharge cycle, the number of maximum values ​​and the time difference between adjacent maximum values.

[0043] During the operation of electronic devices, the high load behavior of electronic devices will accelerate the aging of batteries, which will have a significant impact on the remaining service life of electronic device batteries. However, previous studies usually only considered single-dimensional battery status data, while ignoring the complex relationships and interactions between these behaviors. Therefore, it is impossible to fully capture the loss of electronic device batteries under complex operating conditions, which makes the prediction accuracy of the remaining battery service life in practical applications low. Therefore, this application comprehensively analyzes the impact of electronic devices in complex operating states on the remaining battery service life to improve the prediction accuracy of the remaining service life of electronic device batteries.

[0044] High load behavior often occurs during the operation of electronic devices. The high load behavior of electronic devices will cause the battery inside the battery to experience rapid power loss in a short period of time, change the internal resistance of the battery, and accelerate battery aging. Although occasional high load behavior has limited impact on battery life, frequent high load operation will cause cumulative damage to the remaining battery life.

[0045] For a charge and discharge cycle, all the remaining power collected in chronological order constitute a sequence recorded as a power sequence. Any moment is recorded as the target moment. With the target moment as the center, a sliding window of length g is obtained. In this embodiment, the length g of the sliding window is 5. The ratio of the change in the remaining power in the sliding window to the consumption time is taken as the power consumption rate at the target moment. The sequence of the power consumption rate at each moment sorted in chronological order is recorded as a power consumption rate sequence.

[0046] In addition, considering the cumulative damage to the battery life caused by frequent high-load behavior of electronic devices during operation, the number of maximum values ​​and the time interval between two adjacent maximum values ​​are calculated for the calculated power consumption rate sequence of the electronic device.

[0047] For the power consumption rate sequence, if the power consumption rate at a certain moment is greater than the power consumption rate at the adjacent moments on both sides, the power consumption rate corresponding to this moment is recorded as the maximum value, and all the maximum values ​​of the power consumption rate sequence and the time corresponding to each maximum value are counted. The maximum value also indicates a large power consumption rate, and each maximum value represents a high load behavior.

[0048] Calculate the average of all power consumption rates in the charge-discharge cycle, and let the difference between the power consumption rate at all times and the average of the power consumption rate be the consumption intensity in the charge-discharge cycle. Then, obtain the load consumption in the charge-discharge cycle based on the consumption intensity, the number of maximum values, and the time difference between adjacent maximum values ​​in the charge-discharge cycle.

[0049] The load consumption is positively correlated with the consumption intensity and the number of maximum values, and negatively correlated with the time difference between adjacent maximum values.

[0050] It should be noted that positive correlation means that when one variable increases, the other variable also increases, and the two variables change in the same direction. When one variable changes from large to small or from small to large, the other variable also changes from large to small or from small to large. The specific relationship is determined by actual application and this application does not impose any special restrictions.

[0051] It should be noted that negative correlation means that when one variable increases, the other variable decreases accordingly, and the two variables change in opposite directions. When one variable changes from large to small or from small to large, the other variable also changes from small to large or from large to small. The specific relationship is determined by actual application and this application does not impose any special restrictions.

[0052] Preferably, in this embodiment, the expression of the consumption intensity of the battery in the charge and discharge cycle is: , represents the power consumption rate at the rth moment, It represents the average rate of power consumption during the charge and discharge cycle. represents the number of collected moments, Indicates the intensity of battery consumption during the charge and discharge cycle.

[0053] The expression for the load consumption of the battery during the charge and discharge cycle is: , represents the time of the jth high load behavior, represents the time of the j+1th high load behavior, Indicates the number of high load behaviors in the charge and discharge cycle, Indicates the battery's consumption intensity during the charge and discharge cycle. Indicates the load consumption of the battery during the charge and discharge cycle.

[0054] Preferably, in another embodiment of the present application, the expression for the consumption intensity of the battery in the charge and discharge cycle is: , represents the power consumption rate at the rth moment, It represents the average rate of power consumption during the charge and discharge cycle. represents the number of collected moments, Indicates the intensity of battery consumption during the charge and discharge cycle.

[0055] The expression for the load consumption of the battery during the charge and discharge cycle is: , represents the time of the jth high load behavior, represents the time of the j+1th high load behavior, Indicates the number of high load behaviors in the charge and discharge cycle, Indicates the battery's consumption intensity during the charge and discharge cycle. Indicates the load consumption of the battery during the charge and discharge cycle.

[0056] The greater the rate of power consumption at a certain moment in the battery's charge and discharge cycle, the greater the difference from the average power consumption rate, which causes the electronic device to have a higher load, the greater the consumption intensity, and the greater the impact on the battery life; the more times the high load occurs, the greater the impact on the battery life. At the same time, if the high load occurs continuously, the greater the load consumption, which means that the battery life is continuously affected. The shorter the interval, the greater the impact on the battery life.

[0057] At this point, the load consumption of the battery during the charge and discharge cycle is obtained.

[0058] Step S003, dividing the time corresponding to the surface temperature into intervals according to the surface temperature, and using a surface temperature in the interval to represent the temperature of the interval, and obtaining the temperature difference coefficient of the battery in the charge and discharge cycle according to the difference between the temperature of the interval and the preset optimal working temperature and the time of the interval.

[0059] The chemical reaction rate in the battery of electronic equipment is highly correlated with the ambient temperature. Low temperature will slow down the chemical reaction inside the battery, causing the battery's discharge capacity to decrease, while high temperature will accelerate the side reactions inside the battery, causing the battery material to age quickly and destroying the battery electrode material structure. Especially when electronic equipment is running under extreme temperature conditions, the impact of ambient temperature on battery performance is particularly significant. Too high temperature will accelerate the battery's aging process, while too low temperature will limit the battery's discharge efficiency. Especially in a continuous high temperature environment, the acceleration of the battery's internal chemical reaction will directly lead to a significant shortening of its service life.

[0060] Therefore, during the charge and discharge cycle, the surface temperature data of the electronic device when the electronic device is running is divided into intervals. Each interval is a temperature range. The electronic device has different working times at different temperatures. The working time of the electronic device at all surface temperatures in each interval is counted.

[0061] Preferably, in this embodiment, 1°C is used as the size of the temperature range to obtain multiple intervals, with 23°C ~ 24°C as one interval and 24°C ~ 25°C as another interval, and then the working time of the entire interval is counted, wherein the interval is closed on the left and open on the right, that is, for 23°C ~ 24°C, the working time of 23°C is counted, and the working time of 24°C is not counted; since the temperature difference within 1 degree is small, and for the convenience of calculation, all surface temperatures in the interval are recorded as the minimum temperature in the interval, that is, each interval corresponds to a surface temperature.

[0062] Preferably, in another embodiment of the present application, a temperature range of 1°C is used to obtain multiple intervals with each integer temperature as the center, for example, 23.5°C ~ 24.5°C is an interval, 24.5°C ~ 25.5°C is an interval, the intervals are closed on the left and open on the right, and the integer temperature at the center of each interval is used as the surface temperature of the entire interval.

[0063] Different battery models have different optimal operating temperatures. In this embodiment, the optimal operating temperature of the selected battery is 25 degrees. The temperature difference coefficient of the battery during the charge and discharge cycle is obtained based on the difference between the surface temperature and the optimal operating temperature and the duration of the surface temperature.

[0064] The temperature difference coefficient is positively correlated with the difference between the surface temperature and the optimal working temperature, and is positively correlated with the duration of the surface temperature.

[0065] Preferably, in this embodiment, the expression of the temperature difference coefficient is: , represents the temperature of the qth interval, is the optimal operating temperature of the battery, represents the working time of the qth interval, represents the number of intervals, Indicates the temperature variation coefficient of the battery during the charge and discharge cycle.

[0066] Preferably, in another embodiment of the present application, the expression of the temperature difference coefficient is: , represents the temperature of the qth interval, is the optimal operating temperature of the battery, represents the working time of the qth interval, represents the number of intervals, Indicates the temperature variation coefficient of the battery during the charge and discharge cycle.

[0067] Therefore, when the surface temperature of the electronic device during operation is far away from the optimal operating temperature of the battery, the more the surface temperature of the electronic device deviates from the optimal operating temperature and the longer the operating time, the greater the calculated temperature difference coefficient, and the more serious the impact on the battery life.

[0068] At this point, the temperature variation coefficient of the battery during the charge and discharge cycle is obtained.

[0069] Step S004, obtaining a battery damage coefficient of the battery in the charge and discharge cycle according to the temperature difference coefficient and consumption intensity of the battery in the charge and discharge cycle.

[0070] The battery damage coefficient of the battery in the charge and discharge cycle is calculated according to the temperature difference coefficient and load consumption of the battery in the charge and discharge cycle obtained above.

[0071] Preferably, in this embodiment, the expression of the battery damage coefficient is: , Indicates the battery's consumption intensity during the charge and discharge cycle. Indicates the temperature difference coefficient of the battery during the charge and discharge cycle, represents the linear normalization function, Indicates the battery damage factor during the charge and discharge cycle.

[0072] At the same time, when the surface temperature of the electronic device during operation is far away from the optimal operating temperature of the battery, the more high-load behaviors of the electronic device, that is, the greater the temperature difference coefficient and the greater the consumption intensity, which makes the battery damage coefficient in the final calculated charge and discharge cycle greater, indicating that the high-load behavior of the electronic device in the current charge and discharge cycle has a more serious impact on the remaining life of the battery.

[0073] At this point, the battery damage coefficient of the battery during the charge and discharge cycle is obtained.

[0074] Step S005, based on the average values ​​of the battery temperature, current, voltage and battery damage coefficient during the charge and discharge cycle as neural network input, the remaining capacity of all subsequent charge and discharge cycles is obtained; and the remaining life of the battery is predicted based on the comparison between the remaining capacity and the preset failure threshold.

[0075] Through the above calculation, the battery damage coefficient of the battery in the charge and discharge cycle is obtained. The battery damage coefficient reflects the impact of high-load operation behavior in each charge and discharge cycle on the remaining life of the battery. For each charge and discharge cycle, the average values ​​of the battery's temperature, current, and voltage in the charge and discharge cycle are calculated, and a state vector is formed with the battery damage coefficient.

[0076] When the remaining capacity of the battery is less than 80%, it is considered that the battery has reached its life limit. The state vector of each charge and discharge cycle of each battery in the history is obtained, and the remaining capacity of all subsequent charge and discharge cycles of the vector is obtained. The state vector is used as input and the remaining capacity after all subsequent charge and discharge cycles is used as the label for training through a neural network.

[0077] Neural network selects recurrent neural network for battery remaining capacity prediction, and can use network structures such as LSTM and GRU. The network structure consists of a recurrent layer and a fully connected layer. The recurrent layer can capture the long-term dependency between battery status data and battery capacity in the time series. The fully connected layer converts the output of the recurrent layer into a specific battery capacity prediction value. The activation function uses the tanh activation function, and the loss function uses the mean square error.

[0078] After completing the neural network training, for the current battery, the state vector of the current charge and discharge cycle is obtained, input into the neural network, and output as the remaining capacity of each subsequent charge and discharge cycle. Among all the remaining capacities output, the first charge and discharge cycle with a remaining capacity less than 80% is selected as the end cycle, and the number of charge and discharge cycles between the current charge and discharge cycle and the end cycle is counted, which is the remaining number of battery uses, i.e., the remaining battery life. The specific implementation flow chart for completing the battery life prediction is as follows: Figure 2 shown.

[0079] At this point, the remaining battery life has been predicted based on the battery capacity.

[0080] Based on the same inventive concept as the above method, an embodiment of the present invention also provides a battery life prediction system based on data analysis, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, it implements the steps of any one of the above-mentioned battery life prediction methods based on data analysis.

[0081] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0082] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A battery life prediction method based on data analysis, characterized in that: The method comprises the following steps: Collecting the remaining power at each moment in the charge and discharge cycle, the surface temperature of the electronic device, and data values ​​of different parameters; the parameters include the temperature, current, and voltage of the battery; All the remaining power in the charge and discharge cycle is formed into a sequence, and the power consumption rate is obtained according to the difference of the remaining power in the sequence, and a power consumption rate sequence is formed; the maximum value in the power consumption rate sequence is selected; the consumption intensity is obtained according to the difference between the power consumption rate in the charge and discharge cycle and the average power consumption rate; the load consumption of the battery in the charge and discharge cycle is obtained according to the consumption intensity in the charge and discharge cycle, the number of maximum values, and the time difference between adjacent maximum values; The time corresponding to the surface temperature is divided into intervals according to the surface temperature, and a surface temperature in the interval represents the temperature of the interval, and the temperature difference coefficient of the battery in the charge and discharge cycle is obtained according to the difference between the temperature of the interval and the preset optimal working temperature and the time of the interval; Obtaining a battery damage coefficient of the battery in the charge and discharge cycle according to the temperature difference coefficient and consumption intensity of the battery in the charge and discharge cycle; The average values ​​of the battery's temperature, current, and voltage during the charge and discharge cycle and the battery damage coefficient are used as neural network inputs to obtain the remaining capacity of all subsequent charge and discharge cycles; the remaining battery life is predicted based on the comparison between the remaining capacity and the preset failure threshold.

2. A battery life prediction method based on data analysis as claimed in claim 1, characterized in that: The method for obtaining the power consumption rate according to the difference of the remaining power in the sequence and forming the power consumption rate sequence is: Record any moment as the target moment, obtain a sliding window of preset size with the target moment as the center, take the ratio of the change in the remaining power in the sliding window to the consumption time as the power consumption rate at the target moment, and sort all the power consumption rates by time to obtain a power consumption rate sequence.

3. The battery life prediction method based on data analysis according to claim 1, characterized in that: The method for obtaining the consumption intensity according to the difference between the power consumption rate in the charge and discharge cycle and the average power consumption rate is: The expression of consumption intensity is: , represents the power consumption rate at the rth moment, It represents the average rate of power consumption during the charge and discharge cycle. represents the number of collected moments, Indicates the intensity of battery consumption during the charge and discharge cycle.

4. The battery life prediction method based on data analysis according to claim 1, characterized in that: The method for obtaining the load consumption of the battery in the charge and discharge cycle according to the consumption intensity in the charge and discharge cycle, the number of maximum values, and the time difference between adjacent maximum values ​​is: Load consumption is positively correlated with consumption intensity and the number of maxima, and negatively correlated with the time difference between adjacent maxima.

5. The battery life prediction method based on data analysis according to claim 1, characterized in that: The method of dividing the time corresponding to the surface temperature into intervals according to the surface temperature and using a surface temperature in the interval to represent the temperature of the interval is: Set the temperature range size. All surface temperatures are divided into intervals according to the temperature range size. All surface temperatures in each temperature range size constitute an interval. The interval is closed on the left and open on the right. A temperature in the interval is used to represent the temperature of the interval.

6. The battery life prediction method based on data analysis according to claim 1, characterized in that: The method for obtaining the temperature difference coefficient of the battery in the charge and discharge cycle according to the difference between the temperature of the interval and the preset optimal working temperature and the time of the interval is: The expression of temperature difference coefficient is: , represents the temperature of the qth interval, is the optimal operating temperature of the battery, represents the working time of the qth interval, represents the number of intervals, Indicates the temperature variation coefficient of the battery during the charge and discharge cycle.

7. The battery life prediction method based on data analysis according to claim 1, characterized in that: The method for obtaining the battery damage coefficient of the battery in the charge and discharge cycle according to the temperature difference coefficient and consumption intensity of the battery in the charge and discharge cycle is: The temperature difference coefficient and the consumption intensity are linearly normalized respectively, and the average of the normalized temperature difference coefficient and the normalized consumption intensity within a charge and discharge cycle is taken as the battery damage coefficient of the battery within the charge and discharge cycle.

8. The battery life prediction method based on data analysis according to claim 1, characterized in that: The method for predicting the remaining life of the battery based on the comparison between the remaining capacity and the preset failure threshold is as follows: The remaining capacity of the current charge and discharge cycle and all subsequent charge and discharge cycles is obtained through a neural network. Among all the input energies, the first charge and discharge cycle with a remaining capacity less than 80% is selected as the end cycle, and the number of charge and discharge cycles from the current charge and discharge cycle to the end cycle is counted as the remaining battery life.

9. The method for predicting battery life based on data analysis according to claim 1, characterized in that: The maximum value is the power consumption rate at a certain moment when the power consumption rate at that moment is greater than the power consumption rates at adjacent moments on both sides.

10. A battery life prediction system based on data analysis, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of a battery life prediction method based on data analysis as described in any one of claims 1 to 9 are implemented.

Images