Battery life prediction method, device and equipment based on intelligent analysis of battery data

Abstract

The present invention relates to a battery life prediction method, device and equipment based on intelligent analysis of battery data, the method comprising obtaining the capacity retention rate of the battery at the end of the previous day; obtaining the battery usage record of the day; inputting the usage record of the day into the capacity decay learning model, outputting the capacity loss rate of the battery on the day, and adding it to the last record in the capacity loss rate form; according to the capacity retention rate at the end of the previous day and the capacity loss rate of the day, obtaining the capacity retention rate at the end of the day, and adding it to the last record in the capacity retention rate form; extracting a preset number of capacity loss rate records from the last record in the capacity loss rate form, and calculating the average capacity loss rate of the stage, and then correcting the estimated remaining life of the battery. The present invention realizes the real-time collection and accurate analysis of battery data, and can continuously and accurately update the estimated remaining life during the use of the battery.

Description

Technical Field

[0001] The present invention relates to the field of new energy technologies, and in particular to a battery life prediction method, device and equipment based on intelligent analysis of battery data. Background Art

[0002] With the widespread application of portable electronic devices, electric vehicles and energy storage systems, batteries, as their core power source, have become the focus of industry attention in terms of performance stability and life prediction. Battery life is directly related to the operating cost, safety and user experience of the equipment. However, traditional battery life prediction methods often rely on empirical formulas or simple statistical models, which have many shortcomings and are difficult to accurately reflect the performance degradation law of batteries under actual complex usage conditions.

[0003] In view of this, it is urgent to develop a new battery life prediction method. Summary of the invention

[0004] The main purpose of the present invention is to provide a battery life prediction method, device and equipment based on intelligent analysis of battery data to solve the problem of low accuracy of battery life prediction in the prior art.

[0005] To achieve the above object, a first aspect of the present invention provides a battery life prediction method based on intelligent analysis of battery data, the method comprising:

[0006] Obtain the capacity retention rate of the battery at the end of the previous day from the capacity retention rate table;

[0007] Obtain the battery usage record for the day from the usage record form;

[0008] Input the usage record of the day into a pre-trained capacity decay learning model, output the capacity loss rate of the battery on the day, and add the capacity loss rate of the day to the last record in the capacity loss rate table;

[0009] According to the capacity retention rate at the end of the previous day and the capacity loss rate on the current day, the capacity retention rate at the end of the current day is obtained, and the capacity retention rate at the end of the current day is added after the last record in the capacity retention rate table;

[0010] Extracting a preset number of capacity loss rate records from the last record in the capacity loss rate table, and calculating the average capacity loss rate of the stage corresponding to the preset number of capacity loss rate records;

[0011] The estimated remaining life of the battery is corrected according to the average capacity loss rate in the stage and the last capacity retention rate in the capacity retention rate table.

[0012] Furthermore, before the step of obtaining the battery usage record of the day from the usage record form, the method further comprises:

[0013] Recording the charging process, discharging process, static process and average ambient temperature of the battery on the day, and storing them in the usage record form;

[0014] Among them, the charging process includes the charging start voltage, charging end voltage, charging current, and charging time; the discharging process includes the discharging start voltage, discharging end voltage, discharging current, and discharging time; the static process includes the static start time, static end time, and static open circuit voltage.

[0015] Furthermore, the capacity decay learning model includes a charging process sub-model, a discharging process sub-model, a static process sub-model and an integration layer, and the step of inputting the usage record of the day into the pre-trained capacity decay learning model and outputting the capacity loss rate of the battery on the day includes:

[0016] Extracting charging process records, discharging process records and static process records from the usage records of the day respectively;

[0017] The data corresponding to the charging process record, the discharging process record and the static process record are cleaned and normalized respectively to obtain pre-processed charging data, pre-processed discharging data and pre-processed static data;

[0018] Inputting the preprocessed charging data into the trained charging process sub-model to obtain a first capacity loss rate corresponding to the charging process; inputting the preprocessed discharging data into the trained discharging process sub-model to obtain a second capacity loss rate corresponding to the discharging process; and inputting the preprocessed static data into the trained static process sub-model to obtain a third capacity loss rate corresponding to the static process;

[0019] The first capacity loss rate, the second capacity loss rate and the third capacity loss rate are input into an integration layer, and the first capacity loss rate, the second capacity loss rate and the third capacity loss rate are added by the integration layer to output the capacity loss rate of the battery on that day.

[0020] Further, the step of obtaining the capacity retention rate at the end of the day according to the capacity retention rate at the end of the previous day and the capacity loss rate on the day, and adding the capacity retention rate at the end of the day to the step after the last record in the capacity retention rate form includes:

[0021] Searching from the usage record of the day whether there is a full discharge record, wherein the full discharge record is a discharge process from a charging cut-off voltage corresponding to full charge to a preset discharge cut-off voltage;

[0022] If yes, extracting the first full discharge amount corresponding to the full discharge record;

[0023] Calculating a first actual capacity retention rate of the battery according to the first full discharge amount and a preset rated capacity;

[0024] The last capacity retention rate record in the capacity retention rate table is replaced with the first actual capacity retention rate.

[0025] Further, the step of obtaining the capacity retention rate at the end of the day according to the capacity retention rate at the end of the previous day and the capacity loss rate on the day, and adding the capacity retention rate at the end of the day to the last record in the capacity retention rate form, comprises:

[0026] Searching from the usage records of the day to see whether there is a first record of discharging to a preset discharge cut-off voltage;

[0027] If yes, searching the current day usage record and the historical usage record for the charging end record closest to the first record;

[0028] Determining whether the charging cut-off record is a record of charging to a preset charging cut-off voltage;

[0029] If yes, then obtaining the total duration from the charging cut-off record to the first record, and the total battery resting duration from the charging cut-off record to the first record;

[0030] Calculating the ratio between the total static time of the battery and the total time;

[0031] Determining whether the ratio is less than a preset ratio threshold;

[0032] If yes, the sum of the multiple discharge capacities within the total time is taken as the second full discharge capacity;

[0033] Calculating a second actual capacity retention rate of the battery according to the second full discharge amount and the preset rated capacity;

[0034] The last capacity retention rate record in the capacity retention rate table is replaced with the second actual capacity retention rate.

[0035] Furthermore, after the step of determining whether the ratio is less than a preset ratio threshold, the method further includes:

[0036] If it is greater than the preset ratio threshold, it is determined whether the battery rest time of each section is less than the preset rest time;

[0037] If yes, then continue to execute the step of adding the sum of the multiple discharge capacities within the total time as the second full discharge capacity.

[0038] Furthermore, the step of correcting the estimated remaining life of the battery according to the stage average capacity loss rate and the last capacity retention rate in the capacity retention rate table includes:

[0039] The average capacity loss rate in the stage is used as the estimated capacity loss rate of the battery;

[0040] Calculating the difference between the last capacity retention rate in the capacity retention rate table and a preset capacity remaining rate, wherein the preset capacity remaining rate is a ratio of a preset minimum capacity that can be normally used by the battery to a preset rated capacity;

[0041] The ratio of the difference to the estimated capacity loss rate is calculated, and the ratio is used as the estimated remaining life of the battery.

[0042] Furthermore, after the step of correcting the estimated remaining life of the battery according to the stage average capacity loss rate and the last capacity retention rate in the capacity retention rate table, the step further includes:

[0043] Calculating an estimated expiration date of the battery based on the estimated remaining life of the battery;

[0044] The expiration date is displayed on a preset display device.

[0045] A second aspect of the present invention provides a battery life prediction device based on intelligent analysis of battery data, the device comprising:

[0046] A first acquisition module is used to acquire the capacity retention rate of the battery at the end of the previous day from the capacity retention rate table;

[0047] The second acquisition module is used to obtain the battery usage record of the day from the usage record form;

[0048] A capacity loss rate calculation module, used to input the usage record of the day into a pre-trained capacity decay learning model, output the capacity loss rate of the battery on the day, and add the capacity loss rate of the day to the last record in the capacity loss rate table;

[0049] A capacity retention rate calculation module, used to obtain the capacity retention rate at the end of the current day according to the capacity retention rate at the end of the previous day and the capacity loss rate of the current day, and add the capacity retention rate at the end of the current day to the end of the last record in the capacity retention rate form;

[0050] an average capacity loss rate calculation module, configured to extract a preset number of capacity loss rate records from the last record in the capacity loss rate table, and calculate the average capacity loss rate of the stage corresponding to the preset number of capacity loss rate records;

[0051] The battery life estimation module is used to correct the estimated remaining life of the battery according to the average capacity loss rate in the stage and the last capacity retention rate in the capacity retention rate table.

[0052] A third aspect of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned battery life prediction method based on intelligent analysis of battery data when executing the computer program.

[0053] The battery life prediction method, device and equipment based on intelligent analysis of battery data provided by the present invention input the battery's daily usage record into a pre-trained capacity decay learning model to accurately evaluate the battery's daily capacity loss rate according to the actual usage and individual differences of the battery; and further update the capacity retention rate of the day on a daily basis according to the capacity retention rate at the end of the previous day and the capacity loss rate of the day, and significantly improve the accuracy of battery life prediction and reduce prediction errors through a dynamic analysis method based on actual data; and then, in combination with the user's recent usage habits, extract a preset number of capacity loss rate records from the last record in the capacity loss rate form, and then calculate the stage average capacity loss rate, and correct the estimated remaining life of the battery according to the stage average capacity loss rate and the last capacity retention rate in the capacity retention rate form, thereby realizing real-time collection and accurate analysis of battery data, and being able to continuously and accurately update its capacity retention rate and estimated remaining life during battery use. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a flow chart of a battery life prediction method based on intelligent analysis of battery data in one embodiment of the present invention;

[0055] Figure 2 is a schematic block diagram of the structure of a battery life prediction device based on intelligent analysis of battery data in one embodiment of the present invention;

[0056] Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0057] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] The battery of the present invention can be applied to various terminal devices, such as mobile phones, tablets, electric vehicles, electric toys, etc. The terminal device is provided with a processor and a memory, and various data forms are stored in the memory. When executing the method, the processor calls the data in the memory and executes each step of the battery life method.

[0060] Reference Figure 1 , an embodiment of the present invention provides a battery life prediction method based on intelligent analysis of battery data, the method comprising:

[0061] S1. Obtain the capacity retention rate of the battery at the end of the previous day from the capacity retention rate table;

[0062] S2. Obtaining the battery usage record for the day from the usage record form;

[0063] S3, inputting the usage record of the day into a pre-trained capacity decay learning model, outputting the capacity loss rate of the battery on the day, and adding the capacity loss rate of the day to the last record in the capacity loss rate table;

[0064] S4. Obtain the capacity retention rate at the end of the current day according to the capacity retention rate at the end of the previous day and the capacity loss rate on the current day, and add the capacity retention rate at the end of the current day to the end of the last record in the capacity retention rate table;

[0065] S5. Extracting a preset number of capacity loss rate records from the last record in the capacity loss rate table, and calculating the average capacity loss rate of the stage corresponding to the preset number of capacity loss rate records;

[0066] S6. Correct the estimated remaining life of the battery according to the average capacity loss rate in the stage and the last capacity retention rate in the capacity retention rate table.

[0067] In this embodiment, in the above step S1, the capacity retention rate form is a record table that stores the capacity retention rate data of the battery at different time points. Specifically, the capacity retention rate of the day can be generated and recorded at a preset time point of each day (such as 11:50 p.m., etc.). The capacity retention rate is the percentage of the remaining capacity of the battery relative to the initial capacity. The capacity retention rate form can be stored in the memory of the terminal device, and the capacity retention rate corresponding to the previous day (i.e., the day before the current date) can be found and read through the date index.

[0068] In the above step S2, the above usage record form records the usage of the battery in one day, including but not limited to the charging process, discharging process, static process and average ambient temperature of the day, wherein the charging process includes the charging start voltage, the charging end voltage, the charging current and the charging time; the discharging process includes the discharge start voltage, the discharge end voltage, the discharge current and the discharge time; the static process includes the static start time, the static end time and the static open circuit voltage. These data are used in the subsequent steps to evaluate the capacity loss rate of the battery on that day. Similarly, the usage record form can be accessed through the date index to extract all usage records under the current date.

[0069] In the above step S3, the capacity decay learning model is a machine learning model trained based on a large amount of historical data, which can predict the capacity loss rate of the battery on that day based on the battery usage record. The capacity loss rate form is used to record the capacity loss rate of the battery on different dates. This can achieve real-time update of the capacity loss rate every day, thereby improving the accuracy of battery life prediction.

[0070] In the above step S4, the capacity retention rate at the end of the previous day is subtracted from the capacity loss rate of the day to obtain the capacity retention rate at the end of the day, and the capacity retention rate at the end of the day is added to the end of the capacity retention rate table. The capacity retention rate is updated in real time, so that the actual changes in battery performance can be reflected.

[0071] In the above step S5, in order to more accurately predict the long-term performance of the battery, the capacity loss trend in the recent specified time period is combined. The stage average capacity loss rate represents the average level of battery capacity loss in the specified time period. The specified time period starts from the end of the capacity loss rate form, and traces back a preset number of days of capacity loss rate records (such as the last 30 days), calculates the average capacity loss rate of these records, and obtains the stage average capacity loss rate. By combining the capacity loss trend in the recent specified time, the stability and reliability of estimating the remaining life are improved.

[0072] The average capacity loss rate in a stage = (Σ(capacity loss rate per day) / number of days in a specified time period) × 100%.

[0073] Among them, Σ(daily capacity loss rate) represents the sum of daily capacity loss rates in the specified time period; the number of days in the specified time period refers to the number of days starting from the end of the capacity loss rate table and tracing back to the preset number of capacity loss rate records.

[0074] In the above step S6, based on the current capacity retention rate and the average capacity loss rate of the stage, the estimated remaining life of the battery is dynamically adjusted to reflect the actual changes in battery performance. It should be noted that the last capacity retention rate in the above capacity retention rate form is not completely equivalent to the capacity retention rate at the end of the day obtained in the above step S4. Under normal circumstances, if the capacity retention rate at the end of the day obtained in step S4 is not corrected, the last capacity retention rate in the above capacity retention rate form is the capacity retention rate at the end of the day obtained in step S4; if the capacity retention rate at the end of the day obtained in step S4 is corrected, the last capacity retention rate in the above capacity retention rate form is the corrected value of the capacity retention rate at the end of the day.

[0075] This embodiment inputs the battery's daily usage record into a pre-trained capacity decay learning model to accurately evaluate the battery's daily capacity loss rate based on the actual battery usage and individual differences; and further updates the capacity retention rate of the day on a daily basis based on the capacity retention rate at the end of the previous day and the capacity loss rate of the day. Through a dynamic analysis method based on actual data, the accuracy of battery life prediction is significantly improved and the prediction error is reduced. Combined with the user's recent usage habits, a preset number of capacity loss rate records are extracted from the last record in the capacity loss rate form, and then the stage average capacity loss rate is calculated. According to the stage average capacity loss rate and the last capacity retention rate in the capacity retention rate form, the estimated remaining life of the battery is corrected, thereby realizing real-time collection and precise analysis of battery data, and being able to continuously and accurately update its capacity retention rate and estimated remaining life during battery use.

[0076] In a specific embodiment, before the step S2 of obtaining the battery usage record of the day from the usage record form, the following steps are included:

[0077] S01, recording the charging process, discharging process, static process and average ambient temperature of the battery on the day, and storing them in the usage record form;

[0078] Among them, the charging process includes the charging start voltage, charging end voltage, charging current, and charging time; the discharging process includes the discharging start voltage, discharging end voltage, discharging current, and discharging time; the static process includes the static start time, static end time, and static open circuit voltage.

[0079] In this embodiment, the data of the charging process, discharging process, static process and average ambient temperature of the battery can be acquired and recorded in real time through the data interface and sensor of the battery.

[0080] The above-mentioned charging start voltage is the battery voltage at the beginning of a charging process, the above-mentioned charging end voltage is the battery voltage at the end of a charging process, the above-mentioned charging current is the current size during the charging process, and the above-mentioned charging time is the total time from the beginning of charging to the end of charging.

[0081] The above-mentioned discharge start voltage is the battery voltage at the beginning of a discharge process, the above-mentioned discharge end voltage is the battery voltage at the end of a discharge process, the above-mentioned discharge current is the current size during the discharge process, and the above-mentioned discharge time is the total time from the start of discharge to the end of discharge.

[0082] The static open circuit voltage is recorded at preset time intervals, such as several seconds, tens of seconds, etc.

[0083] The above-mentioned average ambient temperature can be measured by an ambient temperature sensor.

[0084] All the recorded data above are organized into a structured format and stored in a usage record form. The form can be a table in a database or a file in a file system, and the specific form depends on how the system is implemented. When storing, corresponding process labels (such as charging, discharging, and standing) are added to the charging process, the discharging process, and the static process to distinguish them. This is because different usage processes cause different capacity losses to the battery. The above three battery usage processes are recorded separately for separate evaluation of the capacity loss rate in subsequent steps, thereby improving the prediction accuracy of the battery's daily capacity loss rate.

[0085] In a specific embodiment, the capacity decay learning model includes a charging process sub-model, a discharging process sub-model, a static process sub-model and an integration layer. The step S3 of inputting the usage record of the day into the pre-trained capacity decay learning model and outputting the capacity loss rate of the battery on the day includes:

[0086] S301, extracting charging process records, discharging process records and static process records from the usage records of the day respectively;

[0087] S302, performing data cleaning and normalization processing on the data corresponding to the charging process record, the discharging process record and the static process record, respectively, to obtain pre-processed charging data, pre-processed discharging data and pre-processed static data;

[0088] S303, inputting the pre-processed charging data into the trained charging process sub-model to obtain a first capacity loss rate corresponding to the charging process; inputting the pre-processed discharging data into the trained discharging process sub-model to obtain a second capacity loss rate corresponding to the discharging process; and inputting the pre-processed static data into the trained static process sub-model to obtain a third capacity loss rate corresponding to the static process;

[0089] S304. Input the first capacity loss rate, the second capacity loss rate and the third capacity loss rate into an integration layer, add the first capacity loss rate, the second capacity loss rate and the third capacity loss rate through the integration layer, and output the capacity loss rate of the battery on that day.

[0090] In this embodiment, in the above step S301, the record is divided into three parts: charging, discharging and standing according to the labels in the usage record form, and the respective process records are extracted respectively.

[0091] In the above step S302, outliers, missing values, etc. in the records are removed or corrected by data cleaning to ensure the integrity and accuracy of the data. The cleaned data is converted to the same numerical range (such as between 0 and 1) by normalization.

[0092] In the above step S303, the influence of the charging, discharging and resting processes on the battery capacity is calculated respectively by using the pre-trained sub-models for different processes.

[0093] The pre-training process of the charging process sub-model includes training a preset neural network model in a supervised learning manner according to a plurality of charging process sample data to obtain the charging process sub-model. Specifically, the charging process sample data includes the normalized charging start voltage, charging end voltage, charging current, charging time and the corresponding capacity loss rate data, as well as the average ambient temperature.

[0094] The training process includes dividing a plurality of charging process sample data into a plurality of training data and a plurality of verification data according to a preset ratio; inputting the plurality of training data into a preset neural network model (such as a recurrent neural network, a long short-term memory network, a support vector model, etc.), setting the initial parameters of the neural network model, and using a supervised learning method to perform training processing to obtain a charging process sub-model to be verified; using a plurality of verification data to perform verification processing on the charging process sub-model to be verified to obtain a verification result; judging whether the verification result is passed; if the verification result is passed, recording the charging process sub-model to be verified as the charging process sub-model.

[0095] The pre-training process of the above-mentioned discharge process sub-model and the above-mentioned static process sub-model is similar to the pre-training process of the above-mentioned charging process sub-model, except that the training sample data used by the discharge process sub-model is the discharge process sample data, including the normalized discharge start voltage, discharge end voltage, discharge current, discharge time and the corresponding capacity loss rate data, and the average ambient temperature; the training sample data used by the static process sub-model is the static process sample data, including the normalized static start time, static end time, static open circuit voltage and the corresponding capacity loss rate data, and the average ambient temperature.

[0096] In the above step S304, the output results of each sub-model are combined through the integration layer to obtain the total capacity loss rate of the battery on that day. The integration layer receives the capacity loss rate outputs (first, second, and third capacity loss rates) from the three sub-models, performs a simple addition operation on the three capacity loss rates, obtains the total capacity loss rate of the battery on that day, and serves as the final output of step S3.

[0097] This embodiment takes into account the different capacity attenuation of the battery under different usage conditions, subdivides the battery usage process into a charging process, a discharging process, and a static process, and evaluates the capacity loss rate of each process respectively, and then adds the three together to obtain the capacity loss rate of the day, thereby improving the accuracy and reliability of the prediction of the capacity loss rate of the day.

[0098] In a specific embodiment, the step S4 of obtaining the capacity retention rate at the end of the day according to the capacity retention rate at the end of the previous day and the capacity loss rate of the day, and adding the capacity retention rate at the end of the day to the last record in the capacity retention rate form includes:

[0099] S401, searching from the usage record of the day to see whether there is a full discharge record, wherein the full discharge record is a discharge process from a charging cut-off voltage corresponding to full charge to a preset discharge cut-off voltage;

[0100] S402: If yes, extract the first full discharge amount corresponding to the full discharge record;

[0101] S403, calculating a first actual capacity retention rate of the battery according to the first full discharge amount and a preset rated capacity;

[0102] S404: Replace the last capacity retention rate record in the capacity retention rate table with the first actual capacity retention rate.

[0103] In this embodiment, in the above step S401, the above usage records for the day are traversed to find out whether there is a full discharge record. The full discharge record here specifically refers to the complete discharge process of the battery starting from a fully charged state (i.e. reaching the charge cut-off voltage) and continuously discharging until reaching the preset discharge cut-off voltage. The above charge cut-off voltage and discharge cut-off voltage are both preset values ​​and can be set according to the type of battery.

[0104] In the above step S402, if there is a full discharge record in the usage record of the day, the total discharge amount corresponding to the record is extracted and recorded as the first full discharge amount. The first full discharge amount reflects the actual remaining capacity of the battery.

[0105] In the above step S403, the first actual capacity retention rate can be specifically calculated by the following formula:

[0106] First actual capacity retention rate = first full discharge capacity / preset rated capacity × 100%

[0107] In the above step S404, during the use of the battery, if a full discharge occurs on the same day, the capacity retention rate at the end of the day output in step S4 is corrected by the method of this embodiment. Since the first full discharge reflects the actual remaining capacity of the battery, the capacity retention rate calculated by the actual measured remaining capacity is more accurate, and the first actual capacity retention rate can be used to replace the last record in the capacity retention rate table.

[0108] Through the method of this embodiment, the capacity loss rate calculated based on the capacity decay learning model after a period of time can be corrected based on the actual use of the battery from time to time, taking into account the universality of big data prediction and the particularity of individual batteries, thereby further improving the accuracy of battery life prediction.

[0109] In a specific embodiment, the step S4 of obtaining the capacity retention rate at the end of the day according to the capacity retention rate at the end of the previous day and the capacity loss rate of the day, and adding the capacity retention rate at the end of the day to the last record in the capacity retention rate form includes:

[0110] S411, searching from the usage record of the day to see whether there is a first record of discharging to a preset discharge cut-off voltage;

[0111] S412: If yes, searching the current day usage record and the historical usage record for a charging end record closest to the first record;

[0112] S413, determining whether the charging cut-off record is a record of charging to a preset charging cut-off voltage;

[0113] S414: If yes, obtain the total duration from the charging cut-off record to the first record, and the total battery resting duration from the charging cut-off record to the first record;

[0114] S415, calculating the ratio between the total static time of the battery and the total time;

[0115] S416, determining whether the ratio is less than a preset ratio threshold;

[0116] S417: If yes, the sum of the multiple discharge capacities within the total time is taken as the second full discharge capacity;

[0117] S418, calculating a second actual capacity retention rate of the battery according to the second full discharge amount and the preset rated capacity;

[0118] S419: Replace the last capacity retention rate record in the capacity retention rate table with the second actual capacity retention rate.

[0119] This embodiment is used to correct the capacity loss rate calculated by the capacity decay learning model according to the actual use of the battery.

[0120] In the above step S411, the usage records of the day are traversed to find whether there is at least one record showing that the battery has been discharged to a preset discharge cut-off voltage. The above preset discharge cut-off voltage is a preset value and can be set according to the type of battery.

[0121] In the above step S412, if there is a first record of discharging to the discharge cut-off voltage, the charge cut-off record closest to the first record in time is searched in the current day and historical usage records. Between the latest charge cut-off record and the first record is the discharge process of the battery, which may be continuous or may be multi-stage (including at least one static state).

[0122] In the above steps S413-S414, if the charge cutoff record is that the battery is charged to a preset charge cutoff voltage, then the period from the charge cutoff record to the discharge to discharge cutoff voltage is a complete discharge process, and the total duration from the charge cutoff record to the first record of the discharge to discharge cutoff voltage (i.e., the duration of the entire charge and discharge cycle) is calculated, as well as the total duration that the battery is in a static state during the cycle.

[0123] In the above steps S415-S416, the ratio between the total static time of the battery and the total time of the entire charge and discharge cycle is calculated, and the ratio is compared with a preset ratio threshold. The above preset ratio threshold is manually preset (for example, 0.1, 0.05, etc.). If the ratio is less than the preset threshold, it means that the static time is short, and the fluctuation of the battery capacity caused by the static process can be ignored. The data of this discharge process can be used to be equivalent to a full discharge amount, and then the battery capacity retention rate is calculated.

[0124] In the above step S417, the discharge capacities of the above multiple discharge records are added together to obtain the second full discharge capacity. The "multiple discharge" here refers to multiple partial discharge events that occur within a single discharge process of this embodiment (from the above charge cutoff record to the first record of the discharge to discharge cutoff voltage), which together constitute a complete test of the battery capacity.

[0125] In the above step S418, the second actual capacity retention rate can be specifically calculated by the following formula:

[0126] The second actual capacity retention rate = the second full discharge capacity / preset rated capacity × 100%

[0127] In the above step S419, since the second full discharge capacity basically reflects the actual remaining capacity of the battery, the capacity retention rate calculated by the actually measured remaining capacity is more accurate, and the second actual capacity retention rate can be used to replace the last record in the capacity retention rate form.

[0128] Through the method of this embodiment, the capacity loss rate calculated based on the capacity decay learning model after a period of time can be corrected based on the actual use of the battery from time to time, taking into account the universality of big data prediction and the particularity of individual batteries, thereby further improving the accuracy of battery life prediction.

[0129] In a specific embodiment, after the step S416 of determining whether the ratio is less than a preset ratio threshold, the method further includes:

[0130] S4161, if it is greater than the preset ratio threshold, determine whether the battery rest time of each section is less than the preset rest time;

[0131] S4162: If yes, continue to execute the step of adding the sum of the multiple discharge capacities within the total time as the second full discharge capacity.

[0132] In this embodiment, if it is determined that the ratio of the total battery rest time to the total time is greater than the preset ratio threshold, that is, the rest time is relatively long, which may affect the accurate assessment of the battery capacity retention rate. At this time, the data of the charge and discharge cycle is not immediately excluded, but the rest time of each battery segment in the cycle is further analyzed. The entire charge and discharge cycle is traversed, and the rest time of each battery segment is checked and recorded, and compared with a preset rest time threshold. The preset rest time threshold is pre-set according to the battery characteristics and test requirements, and is used to determine whether the rest time is too long to affect the validity of the discharge data.

[0133] If the judgment result in step S4161 is that the rest time of all batteries in the charge and discharge cycle is less than the preset rest time threshold, it means that although the overall rest time is long, the rest time of each segment is not long enough to significantly affect the validity of the discharge data. Therefore, the data of the charge and discharge cycle can continue to be used to evaluate the battery capacity retention rate. At this time, continue to perform the same operation as step S417 in the above embodiment, that is, add the discharge capacity of all discharge segments in the discharge cycle to obtain the second full discharge capacity. It is only further used to update the capacity retention rate.

[0134] This embodiment helps to more flexibly utilize individual battery data of an actual discharge process to optimize the evaluation result of the battery capacity retention rate calculated by the capacity decay learning model.

[0135] In a specific embodiment, the step S6 of correcting the estimated remaining life of the battery according to the stage average capacity loss rate and the last capacity retention rate in the capacity retention rate table includes:

[0136] S601, taking the average capacity loss rate in the stage as the estimated capacity loss rate of the battery;

[0137] S602, calculating the difference between the capacity retention rate at the end of the day and a preset capacity remaining rate, wherein the preset capacity remaining rate is a ratio of a preset minimum capacity that can be normally used by the battery to a preset rated capacity;

[0138] S603: Calculate a ratio of the difference to the estimated capacity loss rate, and use the ratio as the estimated remaining life of the battery.

[0139] In this embodiment, the above-mentioned stage average capacity loss rate reflects the average capacity decay rate of the battery under the current conditions of use. The above-mentioned difference is divided by the estimated capacity loss rate, and the result obtained is the estimated remaining life of the battery, which can be expressed in days, months or years. This embodiment dynamically corrects the estimated remaining life of the battery based on the latest stage average capacity loss data and capacity retention rate records, thereby showing the user the estimated service life of the battery under continuous use under the current conditions of use. When the user sees that the estimated service life increases or decreases, he will also understand whether the current conditions of use are suitable or not, which will also serve to constrain and remind the user's usage habits.

[0140] In a specific embodiment, after the step S6 of correcting the estimated remaining life of the battery according to the stage average capacity loss rate and the last capacity retention rate in the capacity retention rate table, the step further includes:

[0141] S7. Calculating an estimated expiration date of the battery based on the estimated remaining life of the battery;

[0142] S8. Display the expiration date on a preset display device.

[0143] In this embodiment, after the estimated remaining life of the battery is corrected, in order to further facilitate the user to understand the battery usage, the step of calculating the estimated expiration date of the battery and displaying it on a preset display device is performed based on the current date (i.e., the date currently recorded by the system) and the estimated remaining life (expressed in days, months, or years), and the current date is added to the estimated remaining life to obtain the estimated expiration date of the battery.

[0144] The expiration date is further displayed on a preset display device. The preset display device can be a computer screen, a smart phone, a tablet computer or any other device with a display function connected to the battery management system. Through the user interface of the display device, the user can intuitively see the expected expiration date of the battery, so as to prepare in advance, such as replacing the battery or adjusting the usage plan.

[0145] This embodiment not only provides the user with detailed information about the estimated remaining battery life, but also further enhances the user's usage experience and convenience by displaying the estimated expiration date.

[0146] Reference Figure 2 The embodiment of the present invention provides a battery life prediction device based on intelligent analysis of battery data, the device comprising:

[0147] A first acquisition module 10 is used to acquire the capacity retention rate of the battery at the end of the previous day from the capacity retention rate table;

[0148] The second acquisition module 20 is used to obtain the battery usage record of the day from the usage record form;

[0149] The capacity loss rate calculation module 30 is used to input the usage record of the day into the pre-trained capacity decay learning model, output the capacity loss rate of the battery on the day, and add the capacity loss rate of the day to the last record in the capacity loss rate table;

[0150] A capacity retention rate calculation module 40, configured to obtain the capacity retention rate at the end of the current day according to the capacity retention rate at the end of the previous day and the capacity loss rate of the current day, and add the capacity retention rate at the end of the current day to the end of the last record in the capacity retention rate table;

[0151] The average capacity loss rate calculation module 50 is used to extract a preset number of capacity loss rate records from the last record in the capacity loss rate table, and calculate the average capacity loss rate of the stage corresponding to the preset number of capacity loss rate records;

[0152] The battery life estimation module 60 is used to correct the estimated remaining life of the battery according to the average capacity loss rate in the stage and the last capacity retention rate in the capacity retention rate table.

[0153] Furthermore, the device also includes:

[0154] A recording module, used to record the charging process, discharging process, static process and average ambient temperature of the battery on the day, and store them in the usage record form;

[0155] Among them, the charging process includes the charging start voltage, charging end voltage, charging current, and charging time; the discharging process includes the discharging start voltage, discharging end voltage, discharging current, and discharging time; the static process includes the static start time, static end time, and static open circuit voltage.

[0156] Furthermore, the capacity decay learning model includes a charging process sub-model, a discharging process sub-model, a static process sub-model and an integration layer, and the capacity loss rate calculation module 30 includes:

[0157] A record extraction unit, used to extract charging process records, discharging process records and static process records from the daily usage records;

[0158] A data preprocessing unit, used for performing data cleaning and normalization processing on the data corresponding to the charging process record, the discharging process record and the static process record, respectively, to obtain preprocessed charging data, preprocessed discharging data and preprocessed static data;

[0159] a capacity loss rate distribution calculation unit, configured to input the pre-processed charging data into a trained charging process sub-model to obtain a first capacity loss rate corresponding to the charging process; input the pre-processed discharging data into a trained discharging process sub-model to obtain a second capacity loss rate corresponding to the discharging process; and input the pre-processed static data into a trained static process sub-model to obtain a third capacity loss rate corresponding to the static process;

[0160] The capacity loss rate addition calculation unit is used to input the first capacity loss rate, the second capacity loss rate and the third capacity loss rate into an integration layer, add the first capacity loss rate, the second capacity loss rate and the third capacity loss rate through the integration layer, and output the capacity loss rate of the battery on that day.

[0161] Furthermore, the device also includes:

[0162] A first search module is used to search from the usage record of the day to see whether there is a full discharge record, wherein the full discharge record is a discharge process from a charging cut-off voltage corresponding to full charge to a preset discharge cut-off voltage;

[0163] A first extraction module, configured to extract a first full-discharge amount corresponding to a full-discharge record if there is one;

[0164] A first calculation module, used for calculating a first actual capacity retention rate of the battery according to the first full discharge amount and a preset rated capacity;

[0165] The first replacement module is used to replace the last capacity retention rate record in the capacity retention rate table with the first actual capacity retention rate.

[0166] Furthermore, the device also includes:

[0167] A second search module is used to search from the usage record of the day to see whether there is a first record of discharging to a preset discharge cut-off voltage;

[0168] A third search module, configured to search for a charging cut-off record closest to the first record from the usage record of the day and the historical usage record if there is a first record of discharging to a preset discharge cut-off voltage;

[0169] A first judgment module, used to judge whether the charging cut-off record is a record of charging to a preset charging cut-off voltage;

[0170] A duration acquisition module, for acquiring, if it is a record of charging to a preset charging cut-off voltage, a total duration from the charging cut-off record to the first record, and a total battery resting duration from the charging cut-off record to the first record;

[0171] A second calculation module is used to calculate the ratio between the total static time of the battery and the total time;

[0172] A second judgment module is used to judge whether the ratio is less than a preset ratio threshold;

[0173] A third calculation module is used to calculate the sum of the multiple discharge capacities within the total time as the second full discharge capacity if it is less than a preset ratio threshold;

[0174] a fourth calculation module, configured to calculate a second actual capacity retention rate of the battery according to the second full discharge amount and a preset rated capacity;

[0175] The second replacement module is used to replace the last capacity retention rate record in the capacity retention rate table with the second actual capacity retention rate.

[0176] Furthermore, the device also includes:

[0177] The third judgment module is used to judge whether the static time of each battery section is less than the preset static time if it is greater than the preset ratio threshold;

[0178] If both are less than the preset static time, the third judgment module is connected to the third calculation module to continue to execute the step of adding the sum of the multiple discharge capacities within the total time as the second full discharge capacity.

[0179] Furthermore, the battery life estimation module 60 includes:

[0180] A setting unit, used for taking the average capacity loss rate in the stage as the estimated capacity loss rate of the battery;

[0181] a difference calculation unit, used to calculate the difference between the last capacity retention rate in the capacity retention rate form and a preset capacity remaining rate, wherein the preset capacity remaining rate is a ratio of a preset minimum capacity that can be normally used by the battery to a preset rated capacity;

[0182] The life estimation unit is used to calculate the ratio of the difference to the estimated capacity loss rate, and use the ratio as the estimated remaining life of the battery.

[0183] Furthermore, the device also includes:

[0184] A date calculation module, used to calculate the expected expiration date of the battery according to the estimated remaining life of the battery;

[0185] The date display module is used to display the expiration date on a preset display device.

[0186] For the specific implementation of each module or unit in the above-mentioned device embodiments, please refer to the above-mentioned method embodiments, which will not be repeated here.

[0187] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected via a system bus. Among them, the processor designed by the computer 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, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above-mentioned battery tab defect visual recognition method is implemented.

[0188] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0189] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0190] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0191] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0192] The embodiments of the present invention can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0193] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0194] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A battery life prediction method based on intelligent analysis of battery data, characterized in that: The method comprises: Obtain the capacity retention rate of the battery at the end of the previous day from the capacity retention rate table; Obtain the battery usage record for the day from the usage record form; Input the usage record of the day into a pre-trained capacity decay learning model, output the capacity loss rate of the battery on the day, and add the capacity loss rate of the day to the last record in the capacity loss rate table; According to the capacity retention rate at the end of the previous day and the capacity loss rate on the current day, the capacity retention rate at the end of the current day is obtained, and the capacity retention rate at the end of the current day is added after the last record in the capacity retention rate table; Extracting a preset number of capacity loss rate records from the last record in the capacity loss rate table, and calculating the average capacity loss rate of the stage corresponding to the preset number of capacity loss rate records; Correcting the estimated remaining life of the battery according to the average capacity loss rate in the stage and the last capacity retention rate in the capacity retention rate table; Average capacity loss rate in the said stage = (Σ(capacity loss rate per day) / number of days in the specified time period) × 100%; Among them, Σ(Daily Capacity Loss Rate) represents the sum of daily capacity loss rates in the specified time period; the number of days in the specified time period refers to the number of days starting from the end of the capacity loss rate table and tracing back to the preset number of capacity loss rate records; The step of correcting the estimated remaining life of the battery according to the stage average capacity loss rate and the last capacity retention rate in the capacity retention rate table comprises: The average capacity loss rate in the stage is used as the estimated capacity loss rate of the battery; Calculating the difference between the last capacity retention rate in the capacity retention rate table and a preset capacity remaining rate, wherein the preset capacity remaining rate is a ratio of a preset minimum capacity that can be normally used by the battery to a preset rated capacity; The ratio of the difference to the estimated capacity loss rate is calculated, and the ratio is used as the estimated remaining life of the battery.

2. The battery life prediction method based on battery data intelligent analysis according to claim 1 is characterized in that: Before the step of obtaining the battery usage record of the day from the usage record form, the method includes: Recording the charging process, discharging process, static process and average ambient temperature of the battery on the day, and storing them in the usage record form; Among them, the charging process includes the charging start voltage, charging end voltage, charging current, and charging time; the discharging process includes the discharging start voltage, discharging end voltage, discharging current, and discharging time; the static process includes the static start time, static end time, and static open circuit voltage.

3. The battery life prediction method based on battery data intelligent analysis according to claim 2 is characterized in that: The capacity decay learning model includes a charging process sub-model, a discharging process sub-model, a static process sub-model and an integration layer. The step of inputting the usage record of the day into the pre-trained capacity decay learning model and outputting the capacity loss rate of the battery on the day includes: Extracting charging process records, discharging process records and static process records from the usage records of the day respectively; The data corresponding to the charging process record, the discharging process record and the static process record are cleaned and normalized respectively to obtain pre-processed charging data, pre-processed discharging data and pre-processed static data; Inputting the preprocessed charging data into the trained charging process sub-model to obtain a first capacity loss rate corresponding to the charging process; inputting the preprocessed discharging data into the trained discharging process sub-model to obtain a second capacity loss rate corresponding to the discharging process; and inputting the preprocessed static data into the trained static process sub-model to obtain a third capacity loss rate corresponding to the static process; The first capacity loss rate, the second capacity loss rate and the third capacity loss rate are input into an integration layer, and the first capacity loss rate, the second capacity loss rate and the third capacity loss rate are added together through the integration layer to output the capacity loss rate of the battery on that day.

4. The battery life prediction method based on battery data intelligent analysis according to claim 1, characterized in that: The step of obtaining the capacity retention rate at the end of the current day according to the capacity retention rate at the end of the previous day and the capacity loss rate on the current day, and adding the capacity retention rate at the end of the current day to the step after the last record in the capacity retention rate form includes: Searching from the usage record of the day whether there is a full discharge record, wherein the full discharge record is a discharge process from a charging cut-off voltage corresponding to full charge to a preset discharge cut-off voltage; If yes, extracting the first full discharge amount corresponding to the full discharge record; Calculating a first actual capacity retention rate of the battery according to the first full discharge amount and a preset rated capacity; The last capacity retention rate record in the capacity retention rate table is replaced with the first actual capacity retention rate.

5. The battery life prediction method based on battery data intelligent analysis according to claim 1, characterized in that: The step of obtaining the capacity retention rate at the end of the current day according to the capacity retention rate at the end of the previous day and the capacity loss rate on the current day, and adding the capacity retention rate at the end of the current day to the step after the last record in the capacity retention rate form includes: Searching from the usage records of the day to see whether there is a first record of discharging to a preset discharge cut-off voltage; If yes, searching the current day usage record and the historical usage record for the charging end record closest to the first record; Determining whether the charging cut-off record is a record of charging to a preset charging cut-off voltage; If yes, then obtaining the total duration from the charging cut-off record to the first record, and the total battery resting duration from the charging cut-off record to the first record; Calculating the ratio between the total static time of the battery and the total time; Determining whether the ratio is less than a preset ratio threshold; If yes, the sum of the multiple discharge capacities within the total time is taken as the second full discharge capacity; Calculating a second actual capacity retention rate of the battery according to the second full discharge amount and the preset rated capacity; The last capacity retention rate record in the capacity retention rate table is replaced with the second actual capacity retention rate.

6. The battery life prediction method based on battery data intelligent analysis according to claim 5, characterized in that: After the step of determining whether the ratio is less than a preset ratio threshold, the method further includes: If it is greater than the preset ratio threshold, it is determined whether the battery rest time of each section is less than the preset rest time; If yes, then continue to execute the step of adding the sum of the multiple discharge capacities within the total time as the second full discharge capacity.

7. The battery life prediction method based on battery data intelligent analysis according to claim 1, characterized in that: After the step of correcting the estimated remaining life of the battery according to the stage average capacity loss rate and the last capacity retention rate in the capacity retention rate table, the method further includes: Calculating an estimated expiration date of the battery based on the estimated remaining life of the battery; The expiration date is displayed on a preset display device.

8. A battery life prediction device based on intelligent analysis of battery data, characterized in that: The device comprises: A first acquisition module is used to acquire the capacity retention rate of the battery at the end of the previous day from the capacity retention rate table; The second acquisition module is used to obtain the battery usage record of the day from the usage record form; A capacity loss rate calculation module, used to input the usage record of the day into a pre-trained capacity decay learning model, output the capacity loss rate of the battery on the day, and add the capacity loss rate of the day to the last record in the capacity loss rate table; A capacity retention rate calculation module, used to obtain the capacity retention rate at the end of the current day according to the capacity retention rate at the end of the previous day and the capacity loss rate of the current day, and add the capacity retention rate at the end of the current day to the end of the last record in the capacity retention rate form; an average capacity loss rate calculation module, configured to extract a preset number of capacity loss rate records from the last record in the capacity loss rate table, and calculate the average capacity loss rate of the stage corresponding to the preset number of capacity loss rate records; A battery life estimation module, used for correcting the estimated remaining life of the battery according to the average capacity loss rate in the stage and the last capacity retention rate in the capacity retention rate table; The battery life estimation module includes: A setting unit, used for taking the average capacity loss rate in the stage as the estimated capacity loss rate of the battery; a difference calculation unit, used to calculate the difference between the last capacity retention rate in the capacity retention rate form and a preset capacity remaining rate, wherein the preset capacity remaining rate is a ratio of a preset minimum capacity that can be normally used by the battery to a preset rated capacity; a life estimating unit, configured to calculate a ratio of the difference to the estimated capacity loss rate, and use the ratio as an estimated remaining life of the battery; Average capacity loss rate in the said stage = (Σ(capacity loss rate per day) / number of days in the specified time period) × 100%; Among them, Σ(daily capacity loss rate) represents the sum of daily capacity loss rates in the specified time period; the number of days in the specified time period refers to the number of days starting from the end of the capacity loss rate table and tracing back to the preset number of capacity loss rate records.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the battery life prediction method based on intelligent analysis of battery data described in any one of claims 1 to 7 are implemented.

Images