CN116203427A - Method, device, equipment and storage medium for predicting vehicle-mounted battery health state - Google Patents
Method, device, equipment and storage medium for predicting vehicle-mounted battery health state Download PDFInfo
- Publication number
- CN116203427A CN116203427A CN202310012734.0A CN202310012734A CN116203427A CN 116203427 A CN116203427 A CN 116203427A CN 202310012734 A CN202310012734 A CN 202310012734A CN 116203427 A CN116203427 A CN 116203427A
- Authority
- CN
- China
- Prior art keywords
- discharge
- sample vehicle
- vehicle
- battery
- depth
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L58/00—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles
- B60L58/10—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries
- B60L58/16—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries responding to battery ageing, e.g. to the number of charging cycles or the state of health [SoH]
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L2240/00—Control parameters of input or output; Target parameters
- B60L2240/40—Drive Train control parameters
- B60L2240/54—Drive Train control parameters related to batteries
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Abstract
The application relates to the technical field of battery health state prediction, and provides a vehicle-mounted battery health state prediction method.
Description
Technical Field
The present disclosure relates to the field of battery state of health prediction, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for predicting a vehicle-mounted battery state of health.
Background
The vehicle battery has stable charge and discharge performance and sufficient stored energy, and the vehicle battery health status is an important standard for measuring whether the battery has stable charge and discharge performance and sufficient stored energy, so that calculation and prediction of the vehicle battery health status are very necessary, but the calculation and prediction method of the battery health status is not accurate enough at present.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a method, an apparatus, a computer device, a storage medium, and a computer program product for predicting a state of health of an in-vehicle battery.
The application provides a prediction method of a vehicle-mounted battery state of health, which comprises the following steps:
acquiring the travel frequency of a sample vehicle;
if the travel frequency of the sample vehicle is high, selecting a discharge process with high discharge depth from a plurality of discharge processes of the sample vehicle, and calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the discharge process with high discharge depth;
If the travel frequency of the sample vehicle is low, selecting a discharge process with low discharge depth and high charge amplitude from a plurality of discharge processes of the sample vehicle, and calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to data of charge processes adjacent to the discharge process with low discharge depth and high charge amplitude;
according to the data of the discharging process with high discharging depth and/or the data of the discharging process with low discharging depth and high charging amplitude, an input sequence is obtained;
obtaining an output sequence according to the maximum discharge capacity corresponding to the discharge process with high discharge depth and/or the maximum discharge capacity corresponding to the discharge process with low discharge depth and high charge amplitude;
training a target model according to the input sequence and the output sequence; and predicting the health state of the vehicle-mounted battery according to the maximum discharge capacity predicted by the target model.
In one embodiment, calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the discharge process with high discharge depth includes:
according to the current and the voltage in the data of the discharging process with high discharging depth, calculating the energy released by corresponding SOC consumed by a sample vehicle-mounted battery of the sample vehicle;
And according to the proportional relation between the corresponding SOC consumed by the sample vehicle-mounted battery of the sample vehicle and the released energy, obtaining the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle.
In one embodiment, after calculating the energy released by the corresponding SOC of the sample on-board battery of the sample vehicle, the method further comprises:
and calculating the current health state of the sample vehicle-mounted battery of the sample vehicle according to a fitting slope and a correlation coefficient by a linear regression algorithm, wherein the fitting slope and the correlation coefficient are the fitting slope and the correlation coefficient between the corresponding SOC consumed by the sample vehicle-mounted batteries of a plurality of groups of sample vehicles and the released energy.
In one embodiment, calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the charge process adjacent to the discharge process with low discharge depth and high charge amplitude includes:
calculating the maximum charge capacity and the battery coulomb efficiency coefficient of a sample vehicle-mounted battery of the sample vehicle according to the data of the charge process adjacent to the discharge process with low discharge depth and high charge amplitude;
and multiplying the maximum charge capacity by a battery coulomb efficiency coefficient to obtain the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle.
In one embodiment, calculating the maximum charge capacity of the sample vehicle-mounted battery of the sample vehicle according to the data of the charge process adjacent to the discharge process with low discharge depth and high charge amplitude includes:
for the data of the adjacent charging process of the discharging process with low discharging depth and high charging amplitude, a sliding window method is adopted, and the maximum charging capacity corresponding to a single window is calculated according to the data in the single window; the window is a time window;
after sliding the windows for a plurality of times, obtaining the maximum charging capacity corresponding to each window, and eliminating abnormal values in the maximum charging capacity corresponding to each window;
and taking an average value of the plurality of maximum charging capacities after the abnormal values are removed, and obtaining the maximum charging capacity of the sample vehicle-mounted battery of the sample vehicle.
In one embodiment, calculating the battery coulomb efficiency coefficient of the sample vehicle-mounted battery of the sample vehicle according to the data of the charging process adjacent to the discharging process with low discharging depth and high charging amplitude comprises:
when the variation range of the SOC of the sample vehicle-mounted battery of the sample vehicle is the same, according to the discharge process data with low discharge depth and high charge amplitude and the data of the charge processes adjacent to the discharge processes with low discharge depth and high charge amplitude;
The ratio of the released charge amount to the absorbed charge amount is calculated to obtain the battery coulomb efficiency coefficient.
In one embodiment, the discharging process data and the charging process data include:
sampling time, sample vehicle-mounted battery SOC of the sample vehicle, voltage of the sample vehicle-mounted battery of the sample vehicle, current of the sample vehicle-mounted battery of the sample vehicle, temperature of the sample vehicle-mounted battery of the sample vehicle, and driving range of the sample vehicle.
The application provides a prediction device of on-vehicle battery state of health, the device includes:
the frequency acquisition module is used for acquiring the travel frequency of the sample vehicle;
the calculation module is used for selecting a discharge process with high discharge depth from a plurality of discharge processes of the sample vehicle if the travel frequency of the sample vehicle is high, and calculating the maximum discharge capacity of the vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the discharge process with high discharge depth;
the calculation module is further used for selecting a discharging process with low discharging depth and high charging amplitude from a plurality of discharging processes of the sample vehicle if the travel frequency of the sample vehicle is low, and calculating the maximum discharging capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharging process according to the data of the charging processes adjacent to the discharging process with low discharging depth and high charging amplitude;
The training module is used for obtaining an input sequence according to the data of the discharge process with high discharge depth and/or the data of the discharge process with low discharge depth and high charge amplitude;
the training module is also used for obtaining an output sequence according to the maximum discharge capacity corresponding to the discharge process with high discharge depth and/or the maximum discharge capacity corresponding to the discharge process with low discharge depth and high charge amplitude;
the prediction module is used for training a target model according to the input sequence and the output sequence; and predicting the health state of the vehicle-mounted battery according to the maximum discharge capacity predicted by the target model.
The present application provides a computer device comprising a memory storing a computer program and a processor executing the above method.
The present application provides a computer readable storage medium having stored thereon a computer program for execution by a processor of the above method.
The present application provides a computer program product having a computer program stored thereon, the computer program being executed by a processor to perform the above method.
According to the prediction method, the prediction device, the computer equipment, the storage medium and the computer program product of the vehicle-mounted battery health state, the travel frequency of the sample vehicle is mainly obtained, if the travel frequency of the sample vehicle is high, the discharge process with high discharge depth is selected from a plurality of discharge processes of the sample vehicle, the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process is calculated according to the data of the discharge process with high discharge depth, if the travel frequency of the sample vehicle is low, the discharge process with low discharge depth and high charge amplitude is selected from the plurality of discharge processes of the sample vehicle, and the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process is calculated according to the data of the adjacent discharge processes with low discharge depth and high charge amplitude, so that the accurate calculation of the current maximum discharge capacity of the vehicle-mounted battery is realized.
Then, according to the data of the discharging process with high discharging depth and/or the data of the discharging process with low discharging depth and high charging amplitude, an input sequence is obtained, according to the maximum discharging capacity corresponding to the discharging process with high discharging depth and/or the maximum discharging capacity corresponding to the discharging process with low discharging depth and high charging amplitude, an output sequence is obtained, and according to the input sequence and the output sequence, a target model is trained to predict the health state of the vehicle-mounted battery according to the maximum discharging capacity predicted by the target model, so that the accurate prediction of the health state of the vehicle-mounted battery is realized.
Drawings
FIG. 1 is a flow chart of a method for predicting a state of health of an on-board battery according to one embodiment;
FIG. 2 is a schematic diagram showing a comparison of charge and discharge process data in one embodiment;
FIG. 3 is a flow chart of a method for predicting a state of health of an on-board battery according to one embodiment;
FIG. 4 is a flow diagram of a sliding window method in one embodiment;
FIG. 5 is a flow chart of a method for predicting a state of health of an on-board battery according to one embodiment;
FIG. 6 is a diagram of long and short term memory network nodes in a method of predicting the state of health of an on-board battery according to one embodiment;
FIG. 7 is a schematic diagram showing the effect of a method for predicting the state of health of an on-board battery according to one embodiment;
FIG. 8 is a block diagram of an apparatus for predicting a state of health of an on-board battery in one embodiment;
fig. 9 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.
In one embodiment, as shown in fig. 1, there is provided a method for predicting a state of health of a vehicle-mounted battery, including the steps of:
Step S101, obtaining a travel frequency of the sample vehicle.
The travel frequency refers to the number of times of travel of the sample vehicle in a certain time, for example, the travel frequency of the sample vehicle is 5 times per week.
Step S102, if the travel frequency of the sample vehicle is high, selecting a discharge process with a high discharge depth from a plurality of discharge processes of the sample vehicle, and calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the discharge process with the high discharge depth.
The above travel frequency is higher than a set travel frequency, for example, the set travel frequency is 5 times per week, then when the travel frequency of the sample vehicle is higher than 5 times per week, the travel frequency of the sample vehicle is higher, when the travel frequency of the sample vehicle is lower than 5 times per week, the travel frequency of the sample vehicle is lower, the above depth of discharge is higher than the set depth of discharge, for example, the set depth of discharge is 60%, then when the depth of discharge exceeds 60%, that is, the SOC (state of charge) value decreases by more than 60%, the depth of discharge of the sample vehicle battery is higher, and when the depth of discharge does not exceed 60%, that is, the SOC value decreases by not more than 60%.
Specifically, sample vehicle charging and discharging process data including sampling time, sample vehicle-mounted battery SOC of the sample vehicle, voltage of sample vehicle-mounted battery of the sample vehicle, current of sample vehicle-mounted battery of the sample vehicle, temperature of sample vehicle-mounted battery of the sample vehicle and driving mileage of the sample vehicle can be read from a database, then whether the travel frequency of the sample vehicle is higher than a set frequency or not is judged according to the sampling time and the sample vehicle-mounted battery SOC of the sample vehicle, namely, whether the travel frequency exceeds 5 times per week, for vehicles with high travel frequency, a plurality of discharging processes with high discharging depth, namely, discharging processes with discharging depth exceeding 60%, are selected, the maximum discharging capacity of the corresponding discharging process is calculated according to the discharging process data with each discharging depth exceeding 60%, and the obtained maximum amplifying capacity is compared with the rated capacity of the vehicle-mounted battery, so that the current vehicle health state can be obtained.
Further, for a sample vehicle with high travel frequency, a discharge process with high discharge depth is selected to calculate the maximum discharge capacity corresponding to the discharge process with high discharge depth, because for a sample vehicle with high travel frequency and discharge depth, the discharge process data of the battery is sufficient, and the calculation result is more accurate, wherein the calculation method is that the energy released by the corresponding SOC is calculated by the sample vehicle-mounted battery according to the current and voltage in the data of the discharge process with high discharge depth, the maximum discharge capacity of the sample vehicle-mounted battery is calculated according to the proportional relation between the corresponding SOC of the sample vehicle-mounted battery and the released energy, for example, for one discharge process, the battery SOC display value is reduced by 10%, the proportional relation between the actual energy already consumed and the corresponding SOC is calculated by using the integral calculation of the discharge power and the time, for example, the energy released by 10% SOC is A, the corresponding energy released by 100% SOC is 10A, namely the maximum discharge capacity is calculated by the following formula:
Step S103, if the travel frequency of the sample vehicle is low, selecting a discharge process with low discharge depth and high charge amplitude from a plurality of discharge processes of the sample vehicle, and calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the charge processes adjacent to the discharge process with low discharge depth and high charge amplitude.
The discharging process with low discharging depth and high charging amplitude refers to a charging process with high charging amplitude performed on the sample vehicle before a discharging process with low discharging depth is performed, and the charging process is a charging process adjacent to the discharging process with low discharging depth and high charging amplitude; the high charging range refers to a charging process in which the SOC of the vehicle-mounted battery increases by more than a set value, for example, a charging process in which the SOC of the vehicle-mounted battery increases by more than 60% is a charging process in which the charging range is high.
For a sample vehicle with low travel frequency, if the discharge process data of the vehicle-mounted battery is directly used to calculate the maximum discharge capacity, the accuracy of calculation is reduced, so for a sample vehicle with low travel frequency, the charge process data of the discharge process with low discharge depth and high charge amplitude is selected to calculate the maximum discharge capacity of the corresponding discharge process, wherein at this time, the charge and discharge data pair of the vehicle-mounted battery is as shown in fig. 2.
Step S104, according to the data of the discharging process with high discharging depth and/or the data of the discharging process with low discharging depth and high charging amplitude, an input sequence is obtained.
According to the method, different calculation modes are selected according to different discharging depths and trip frequencies of the vehicle-mounted battery, the maximum discharging capacity corresponding to the discharging process with high discharging depth is calculated and obtained, and/or the maximum discharging capacity corresponding to the discharging process with low discharging depth and high charging amplitude is used as an output sequence of a CNN (Convolutional Neural Network ) -seq2seq (sequence-to-sequence) model, and data of the discharging process with high discharging depth corresponding to the calculated maximum discharging capacity, and/or the data of the discharging process with low discharging depth and high charging amplitude is used as an input sequence of a CNN-seq2seq model to carry out model training, wherein the CNN-seq2seq model is a training target model.
Specifically, the CNN learns the characteristic corresponding relation between various input data and the maximum discharge capacity of the battery, learns the sequence relation of nodes with different sampling times through seq2seq, encodes the input battery state quantity, adjusts the weight of the hidden node, and decodes to obtain the maximum discharge capacity sequence of the battery. The CNN-seq2seq model is constructed through a layering strategy and is divided into a model input layer, a CNN convolution layer, a pooling layer, a data coding layer, a data decoding layer, a full connection layer and an output layer according to the sequence of data operation.
For example, after the maximum discharge capacity of the vehicle-mounted battery is calculated by the method, charging and discharging processes with SOC value change exceeding 70% can be screened, namely, discharging processes with discharge depth exceeding 70% are selected for data of discharging processes with high discharge depth, discharging processes with discharge amplitude exceeding 70% are selected for data of discharging processes with low discharge depth and high charge amplitude, namely, discharging process data with discharge depth being low and charge amplitude being up to 70% are selected as input sequences, and the CNN-seq2seq model is trained according to corresponding maximum discharge capacities obtained by the discharging processes according to different methods as output sequences. Each discharging process can select a plurality of state quantities as a group of input sequences, for example, each process can select 2640 state quantities as a group of input sequences, the state quantities can include voltage, current, temperature and mileage, meanwhile, a plurality of vehicle-mounted batteries, for example, 120 vehicle-mounted batteries, can be used as output sequences to train a CNN-seq2seq model according to the maximum discharge capacity calculated by different methods, in addition, the state quantities in discharging process data corresponding to the SOC value change exceeding 70% in one quarter in the past period can be used as input sequences, and the corresponding maximum discharge capacities are used as output sequences to train the CNN-seq2seq model.
Step S105, obtaining an output sequence according to the maximum discharge capacity corresponding to the discharge process with high discharge depth and/or the maximum discharge capacity corresponding to the discharge process with low discharge depth and high charge amplitude.
Step S106, training a target model according to the input sequence and the output sequence; and predicting the health state of the vehicle-mounted battery according to the maximum discharge capacity predicted by the target model.
According to the method for predicting the state of health of the vehicle-mounted battery, the travel frequency of the sample vehicle is mainly obtained, if the travel frequency of the sample vehicle is high, the discharge process with high discharge depth is selected from a plurality of discharge processes of the sample vehicle, the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process is calculated according to the data of the discharge process with high discharge depth, if the travel frequency of the sample vehicle is low, the discharge process with low discharge depth and high charge amplitude is selected from a plurality of discharge processes of the sample vehicle, and the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process is calculated according to the data of the charge processes adjacent to the discharge process with low discharge depth and high charge amplitude, so that the accurate calculation of the current maximum discharge capacity of the vehicle-mounted battery is realized.
Then, according to the data of the discharging process with high discharging depth and/or the data of the discharging process with low discharging depth and high charging amplitude, an input sequence is obtained, according to the maximum discharging capacity corresponding to the discharging process with high discharging depth and/or the maximum discharging capacity corresponding to the discharging process with low discharging depth and high charging amplitude, an output sequence is obtained, according to the input sequence and the output sequence, a target model is trained and obtained, so that the state of health of the vehicle-mounted battery is predicted according to the maximum discharging capacity predicted by the target model, and the accurate prediction of the state of health of the vehicle-mounted battery is realized.
In one embodiment, calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the discharge process with high discharge depth includes:
according to the current and the voltage in the data of the discharging process with high discharging depth, calculating the energy released by the corresponding SOC consumed by the sample vehicle-mounted battery of the sample vehicle; and obtaining the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle according to the proportional relation between the corresponding SOC consumed by the sample vehicle-mounted battery of the sample vehicle and the released energy.
In one embodiment, after calculating the energy released by the corresponding SOC of the sample vehicle's sample on-board battery, the method further includes:
and calculating the current health state of the sample vehicle-mounted battery of the sample vehicle according to the fitting slope and the correlation coefficient by a linear regression algorithm, wherein the fitting slope and the correlation coefficient are the fitting slope and the correlation coefficient between the corresponding SOC consumed by the sample vehicle-mounted batteries of a plurality of groups of sample vehicles and the released energy.
The current vehicle-mounted battery state of health is calculated by comparing the calculated maximum discharge capacity with the rated capacity, and the vehicle-mounted battery of the sample vehicle with high travel frequency and high discharge depth can be used for calculating the vehicle-mounted battery state of health of the sample vehicle based on a linear regression mode after corresponding SOC and released energy are consumed by the sample vehicle-mounted batteries of a plurality of groups of sample vehicles are calculated, mainly selecting a plurality of pieces of vehicle-mounted battery discharge process data to perform linear fitting, calculating the maximum capacity of the battery, and calculating the slope and R between the consumed SOC and the consumed energy in the driving process 2 Values to fit battery capacity, slope needs to be multiplied by normalized R 2 Finally, normalizing to finally calculate the health state of the battery, wherein slope is a fitting slope, R 2 Is a correlation coefficient.
Specifically, the description with reference to fig. 3 includes step S201: calculating slope and R between consumed SOC and consumed energy in running process calculated by vehicle 2 . Step S202: r is R 2 The column is divided by the maximum value of the column for normalization, and the calculation formula is as follows:
step S203: the slope is combined with R 2 The column multiplication results in SOH (battery state of health) columns, the calculation formula is: soh=r 2 Slope_n. Step S204: the SOH column is normalized by dividing it by the maximum value of the column. Step S205: and finally calculating to obtain a final SOH value SOH_result of the battery, wherein the calculation formula is as follows: />
Wherein, before calculation, data processing is needed first, the driving process is reclassified, if the time interval between the last data and the next data is more than 300s, the two data are consideredStatistics are for individual driving courses, corresponding to both driving courses.
Therefore, for vehicles with higher travel frequency and vehicle-mounted battery discharge depth, the current vehicle-mounted battery health state can be directly and accurately calculated through the method.
In one embodiment, calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the charge process adjacent to the discharge process with low discharge depth and high charge amplitude includes:
Calculating the maximum charge capacity and the battery coulomb efficiency coefficient of a sample vehicle-mounted battery of the sample vehicle according to the data of the charge process adjacent to the discharge process with low discharge depth and high charge amplitude; and multiplying the maximum charge capacity by a battery coulomb efficiency coefficient to obtain the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle.
Further, calculating a battery coulomb efficiency coefficient of the sample vehicle-mounted battery of the sample vehicle from data of charging processes adjacent to the discharging process having the low discharging depth and the high charging amplitude, includes:
when the variation range of the SOC of the sample vehicle-mounted battery of the sample vehicle is the same, according to the discharge process data with low discharge depth and high charge amplitude and the data of the charge processes adjacent to the discharge processes with low discharge depth and high charge amplitude; the ratio of the released charge amount to the absorbed charge amount is calculated to obtain the battery coulomb efficiency coefficient.
Specifically, the coulomb efficiency refers to the percentage of the charge amount of the battery that is discharged by the vehicle-mounted battery in the adjacent charge-discharge period of one battery, reflecting the content of active lithium ions available for charge transfer in the adjacent charge process and discharge process, the most visual representation of the battery capacity decay is the decrease of the discharge amount of the battery, and the ratio of the discharge amount to the absorption amount of the battery energy represents the utilization efficiency of the battery under the condition that the SOC variation range is the same, and the calculation formula for the coulomb efficiency is as follows:
In one embodiment, calculating the maximum charge capacity of the sample vehicle-mounted battery of the sample vehicle according to the data of the charge process adjacent to the discharge process with low discharge depth and high charge amplitude includes:
for the data of the adjacent charging processes of the discharging process with low discharging depth and high charging amplitude, a sliding window method is adopted, and the maximum charging capacity corresponding to a single window is calculated according to the data in the single window; the window is a time window; after sliding the windows for a plurality of times, obtaining the maximum charging capacity corresponding to each window, and eliminating abnormal values in the maximum charging capacity corresponding to each window; and taking an average value of the plurality of maximum charging capacities after the abnormal values are removed, and obtaining the maximum charging capacity of the sample vehicle-mounted battery of the sample vehicle.
Specifically, when the maximum charge capacity of the current vehicle-mounted battery is calculated by using the vehicle-mounted battery charging process data, data with the SOC less than 30% can be abandoned, and when the SOC is too low, the battery performance is unstable, so that the calculation result is influenced; discarding data with SOC greater than 90%, when SOC is close to 100%, the battery is in constant voltage charging or trickle charging mode, and current is gradually attenuated, so that calculation results are affected; the time interval between two adjacent sampling points is <100s.
For calculating the maximum charge capacity of the vehicle-mounted battery, referring to fig. 4, in order to improve the calculation accuracy, a sliding window method may be used to fully utilize data and eliminate the influence of abnormal values, and for each charging process, the full charge energy of the battery is calculated by using the data in the time window every time the charging time window with a fixed length is moved once; after the time frame reaches the end of the charging process, detecting abnormal values in the calculation result by using a box graph and removing the abnormal values; after eliminating abnormal energy values, taking an average value as full charge energy of the battery in the charging process, namely maximum charge capacity, and calculating the following formula:
therefore, the maximum discharge capacity of the vehicle-mounted battery can be obtained by calculating the battery coulomb efficiency coefficient and the maximum charge capacity of the battery and multiplying the battery maximum charge capacity and the battery coulomb efficiency coefficient, and the current state of health of the vehicle-mounted battery can be obtained by comparing the maximum discharge capacity with the battery rated capacity. By the method, the maximum discharge capacity and the vehicle-mounted battery health state obtained by calculation of the vehicle-mounted battery of the vehicle with low travel frequency and low discharge depth are more accurate. Wherein the maximum charge capacity is also referred to as full charge energy.
In one embodiment, the discharging process data and the charging process data include:
sampling time, sample vehicle-mounted battery SOC of the sample vehicle, voltage of the sample vehicle-mounted battery of the sample vehicle, current of the sample vehicle-mounted battery of the sample vehicle, temperature of the sample vehicle-mounted battery of the sample vehicle, and driving range of the sample vehicle.
In order to better understand the above method, an application example of the method for predicting the state of health of the vehicle-mounted battery of the present application is described in detail below with reference to fig. 5.
The method can realize the functions of two parts, wherein the first part can realize the function of obtaining the maximum capacity of the battery of the vehicle in a certain past period through integration or based on full charge energy and coulomb efficiency based on historical operation data of the vehicle, comparing the maximum capacity with the initial capacity of the battery, and calculating the state of health of the battery; and constructing a CNN-seq2seq network model, selecting the vehicle driving mileage and voltage, current and temperature which are greatly influenced by the battery capacity attenuation as model inputs, outputting the real-time maximum capacity of the battery, and predicting the battery health state.
When the historical state of health of the battery is calculated, different calculation modes are adopted for different battery charge and discharge depths and vehicle travel frequencies: by integration, the battery maximum capacity and the battery state of health are calculated from the ratio relationship between the consumed SOC and the released energy and based on the battery full charge energy. Meanwhile, the health state of the vehicle with high travel frequency and high discharge depth can be directly calculated through a linear regression algorithm.
Specifically, the charge and discharge process data of the sample vehicle are read from the database, including sampling time, sample vehicle-mounted battery SOC of the sample vehicle, voltage of sample vehicle-mounted battery of the sample vehicle, current of sample vehicle-mounted battery of the sample vehicle, temperature of sample vehicle-mounted battery of the sample vehicle and driving mileage of the sample vehicle, then whether travel frequency of the sample vehicle is higher than a set frequency or not is judged according to the sampling time and the sample vehicle-mounted battery SOC (residual capacity) of the sample vehicle, namely, whether the travel frequency exceeds 5 times per week, for vehicles with high travel frequency, a plurality of discharge processes with high discharge depth are selected, namely, the discharge process with the discharge depth exceeding 60%, and the maximum discharge capacity of the corresponding discharge process is calculated according to the discharge process data with each discharge depth exceeding 60%.
The method for calculating the maximum discharge capacity of the sample vehicle battery according to the ratio of the corresponding SOC consumed by the sample vehicle battery to the released energy is calculated, for example, for a discharge process, the battery SOC display value is reduced by 10%, the ratio of the actual energy consumed by the calculation of the integral of the discharge power to the time is used for calculating the maximum discharge capacity of the discharge process, for example, the energy released by the 10% SOC is consumed, the corresponding energy released by the 100% SOC is consumed by the 10A, namely, the maximum discharge capacity is calculated according to the ratio of the corresponding SOC consumed by the sample vehicle battery to the released energy, for example, the ratio of the actual energy consumed by the integral of the discharge power to the time is calculated to be A, and the corresponding energy released by the 100% SOC is consumed by the 10A, and the calculation formula is as follows:
Meanwhile, for the vehicle-mounted battery of the vehicle with high travel frequency and high discharge depth, after corresponding SOC and released energy consumption of the sample vehicle-mounted battery of a plurality of groups of sample vehicles are calculated, the vehicle-mounted battery health state can be calculated based on a linear regression mode, and mainly multi-section vehicle-mounted battery discharge process data are selected for linear fitting, the maximum capacity of the battery is calculated, and the SOC and the consumption are consumed according to the driving processSlope and R between energies 2 Values to fit battery capacity, slope needs to be multiplied by normalized R 2 And finally, carrying out normalization so as to finally calculate the state of health of the battery. Specifically, the description with reference to fig. 3 includes step S201: calculating slope and R between consumed SOC and consumed energy in running process calculated by vehicle 2 . Step S202: r is R 2 The column is divided by the maximum value of the column for normalization, and the calculation formula is as follows:
step S203: the slope is combined with R 2 The column multiplication results in SOH (battery state of health) columns, the calculation formula is: soh=r 2 Slope_n. Step S204: the SOH column is normalized by dividing it by the maximum value of the column. Step S205: and finally calculating to obtain a final SOH value SOH_result of the battery, wherein the calculation formula is as follows: />
Before calculation, data processing is needed first to re-divide the running process, and if the time interval between the previous data and the next data is greater than 300s, the two data are considered to correspond to two running processes, and statistics are all directed to the independent running processes.
For a sample vehicle with low travel frequency, if the discharge process data of the vehicle-mounted battery is directly used to calculate the maximum discharge capacity, the accuracy of calculation is reduced, so for a sample vehicle with low travel frequency, the charge process data of the discharge process with low discharge depth and high charge amplitude is selected to calculate the maximum discharge capacity of the corresponding discharge process. Calculating the maximum charge capacity and the battery coulomb efficiency coefficient of a sample vehicle-mounted battery of the sample vehicle according to the data of the adjacent charge processes of the discharge processes with low discharge depth and high charge amplitude; and multiplying the maximum charge capacity by a battery coulomb efficiency coefficient to obtain the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle, wherein the calculation of the maximum charge capacity is as follows: when the maximum charge capacity of the current vehicle-mounted battery is calculated by using the vehicle-mounted battery charging process data, data with the SOC less than 30% can be abandoned, and when the SOC is too low, the battery performance is unstable, so that the calculation result is influenced; discarding data with SOC greater than 90%, when SOC is close to 100%, the battery is in constant voltage charging or trickle charging mode, and current is gradually attenuated, so that calculation results are affected; the time interval between two adjacent sampling points is <100s. For the purpose of calculating the maximum charge capacity of the vehicle battery, a sliding window method may be used to fully utilize data and eliminate the influence of abnormal values, and for each charging process, the full charge energy of the battery is calculated by using the data in the time window every time the charging time window with a fixed length is moved once for each movement, as described with reference to fig. 4; after the time frame reaches the end of the charging process, detecting abnormal values in the calculation result by using a box graph and removing the abnormal values; after eliminating abnormal energy values, taking an average value as full charge energy of the battery in the charging process, namely maximum charge capacity, and calculating the following formula:
for coulomb efficiency calculation, the ratio of the released charge amount and the absorbed charge amount represents the battery energy utilization efficiency in the case where the SOC variation range is the same, and the calculation formula for coulomb efficiency is as follows:
After the maximum discharge capacity of the vehicle-mounted battery is calculated by the method, charging and discharging processes with the SOC value change exceeding 70% can be screened, namely, discharging processes with the discharging depth exceeding 70% are selected for data of discharging processes with the discharging depth being high, discharging processes with the discharging amplitude exceeding 70% are selected for data of discharging processes with the discharging depth being low and the charging amplitude being high, namely, discharging process data with the discharging depth being low and the charging amplitude reaching 70% are selected as input sequences, and the CNN-seq2seq model is trained according to the corresponding maximum discharge capacities obtained by the discharging processes according to different methods as output sequences. Each process can select 2640 state quantities as a group of input sequences, the state quantities can comprise voltage, current, temperature and mileage, meanwhile, a plurality of vehicle-mounted batteries, such as 120 vehicle-mounted batteries, can be used as output sequences to train the CNN-seq2seq model according to maximum discharge capacities calculated by different methods, furthermore, state quantities in discharge process data corresponding to charge and discharge processes, such as a quarter of SOC value change exceeding 70%, can be used as input sequences, and corresponding maximum discharge capacities can be used as output sequences to train the CNN-seq2seq model.
The construction of the CNN-seq2seq model is realized in a layered manner, and is divided into a model CNN convolution layer and a pooling layer, an encoder and a decoder, a full connection layer and an input and output layer according to the sequence of data flow. The convolution layer extracts battery data characteristics, reduces the dimension of original data, improves the operation efficiency of a network model, and a mathematical model of the convolution layer can be shown by the following formula:
wherein (1)>
Is a model input sequence, x i Representing 660 groups of battery state quantity in a single charge or discharge process, and the length of an input sequence is the number of charge and discharge processes screened in the past quarter. />
Is a 3*3 convolution matrix of layer l, +.>
Is the vector offset constant.
Then, the CNN-seq2seq model builds a pooling layer, reduces the number of parameters and data dimension, and reduces the prediction accuracy reduction of the maximum capacity caused by over fitting. Using a maximum value strategy, the extracted characteristic elements of the input layer are provided with different weights according to the size of the probability value, and the occurrence frequency of the characteristic elements and the set weights are positively correlated, as shown in the following formula:
Next, the CNN-seq2seq model builds the encoder. The input sequence is converted into one-dimensional intermediate vectors using a long and short term memory network as encoder nodes. The input sequence is (x) 1 、x 2 、、x t ) And (5) reducing the dimension of the data by the convolution network. The encoder converts the input into a fixed-length sequence and then inputs the data into the hidden layer. Information f of node state reserved by input t C t-1 Hidden layer state h t-1 And vector offset b c Determining W i Is a weight matrix, using sigmoid and tanh is used as an activation function of the node input gate and the hidden layer, respectively, as shown in the following equation: c (C) t ＝f t C t-1 +σ(W i [h t-1 ,x t ]+b i )tanh(W c [h t-1 ,x t )+b c )。
Finally, the CNN-seq2seq model adds a decoder and a full connection layer. The decoder is implemented using a unidirectional recurrent neural network and the encoder has the same hidden unit, the initialization of the decoder requires the acquisition of the hidden state of the last step of the encoder, the last layer of the decoder uses a fully connected layer to transform the hidden state, and the next output is predicted by maximizing the historical probability distribution and hidden layer vectors as shown in the following equation:
In addition, when the CNN-seq2seq model is trained, a 70% data set is selected, 660 groups of data are input, each group comprises four variables including voltage, current, mileage and temperature, the remaining 30% data set is selected for testing, when the root mean square error of the battery health state obtained by calculation of the model and the health state of a test collecting battery is lower than 0.15, the CNN-seq2seq model can be applied to the prediction of the battery health state, after the model is trained, the state quantity of discharge process data of the vehicle-mounted battery in the current quarter is input, the model can output the maximum discharge capacity of the vehicle-mounted battery in the future quarter according to days, and the predicted value of the vehicle-mounted battery health state in the future quarter can be obtained by comparing the maximum discharge capacity with the rated capacity of the vehicle-mounted battery, wherein the predicted vehicle-mounted battery health state is compared with the actual value, and the predicted vehicle-mounted battery health state is shown in a graph of FIG. 7.
In summary, the present application has at least the following beneficial effects:
1. the method and the device can calculate the maximum discharge capacity of the vehicle and the state of health of the battery in the given time range based on the vehicle running data, and can improve the equipment sensing capability of the battery management system.
2. According to different vehicle travel frequencies and charge and discharge habits, the method is divided into two different calculation methods of calculating coulomb efficiency multiplied by full charge energy of the battery and integrating, so that adaptability of the algorithm under different conditions is enhanced, and accuracy of calculating the maximum capacity of the battery is improved.
3. The method and the device do not need to completely charge and discharge the battery strictly according to a certain current multiplying power, but use actual running data of the vehicle to calculate the health state of the battery, are easy to operate, can be applied to a vehicle health management system, improve the battery state monitoring efficiency of enterprises, and reduce the running and maintenance cost of the vehicle.
4. The method is based on the calculated vehicle health state and data of voltage, current, driving mileage and temperature, and a network model of sequence input and sequence output is trained. Under the condition that the charge and discharge habits and the travel frequency of the vehicle are kept stable, the battery health condition of a quarter in the future can be predicted in real time, the battery state is fed back in time, the battery capacity attenuation early warning is achieved in advance, and the driving experience of the user is improved.
It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
In one embodiment, as shown in fig. 8, there is provided a prediction apparatus for a state of health of a vehicle-mounted battery, including:
a frequency acquisition module 801, configured to acquire a travel frequency of a sample vehicle;
a calculating module 802, configured to select a discharge process with a high discharge depth from among a plurality of discharge processes of the sample vehicle if the travel frequency of the sample vehicle is high, and calculate a maximum discharge capacity of the vehicle-mounted battery of the sample vehicle in a corresponding discharge process according to data of the discharge process with the high discharge depth;
The calculating module 802 is further configured to, if the travel frequency of the sample vehicle is low, select a discharge process with low discharge depth and high charge amplitude from among a plurality of discharge processes of the sample vehicle, and calculate a maximum discharge capacity of a sample vehicle-mounted battery of the sample vehicle in a corresponding discharge process according to data of charge processes adjacent to the discharge process with low discharge depth and high charge amplitude;
the training module 803 is configured to obtain an input sequence according to data of a discharge process with a high discharge depth and/or data of a discharge process with a low discharge depth and a high charge amplitude;
the training module 803 is further configured to obtain an output sequence according to a maximum discharge capacity corresponding to a discharge process with a high discharge depth and/or a maximum discharge capacity corresponding to a discharge process with a low discharge depth and a high charge amplitude;
a prediction module 804, configured to train a target model according to the input sequence and the output sequence; and predicting the health state of the vehicle-mounted battery according to the maximum discharge capacity predicted by the target model.
In one embodiment, the calculating module 802 is further configured to calculate, according to the current and the voltage in the data of the discharge process with the high depth of discharge, the energy released by the corresponding SOC consumed by the sample vehicle-mounted battery of the sample vehicle, and calculate the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle according to the proportional relationship between the corresponding SOC consumed by the sample vehicle-mounted battery of the sample vehicle and the released energy.
In one embodiment, the calculating module 802 is further configured to calculate, according to a linear regression algorithm, a current health status of the sample vehicle-mounted battery of the sample vehicle according to a fitting slope and a correlation coefficient, where the fitting slope and the correlation coefficient are fitting slopes and correlation coefficients between the sample vehicle-mounted batteries of the plurality of groups of sample vehicles consuming corresponding SOCs and released energy.
In one embodiment, the calculating module 802 is further configured to calculate, according to the data of the charging process adjacent to the discharging process with the low depth of discharge and the high charging amplitude, a maximum charging capacity and a battery coulomb efficiency coefficient of the sample vehicle battery of the sample vehicle, and multiply the maximum charging capacity and the battery coulomb efficiency coefficient to obtain a maximum discharging capacity of the sample vehicle battery of the sample vehicle.
In one embodiment, the calculating module 802 is further configured to calculate, according to data in a single window, a maximum charging capacity corresponding to the single window by using a sliding window method for data of a charging process adjacent to the discharging process with a low discharging depth and a high charging amplitude, where the window is a time window, slide the window multiple times to obtain the maximum charging capacity corresponding to each window, reject an outlier in the maximum charging capacity corresponding to each window, and average the maximum charging capacities after rejecting the outlier to obtain the maximum charging capacity of the sample vehicle-mounted battery of the sample vehicle.
In one embodiment, the calculating module 802 is further configured to calculate, when the SOC variation ranges of the sample vehicle-mounted battery of the sample vehicle are the same, a ratio of the released charge amount to the absorbed charge amount according to the discharge process data with low discharge depth and high charge amplitude and the data of the charge processes adjacent to the discharge process with low discharge depth and high charge amplitude, so as to obtain the coulomb efficiency coefficient of the battery.
In one embodiment, the discharging process data and the charging process data include a sampling time, a sample on-board battery SOC of the sample vehicle, a voltage of the sample on-board battery of the sample vehicle, a current of the sample on-board battery of the sample vehicle, a temperature of the sample on-board battery of the sample vehicle, and a driving range of the sample vehicle.
For specific limitations on the prediction apparatus of the vehicle-mounted battery state of health, reference may be made to the above limitation on the prediction method of the vehicle-mounted battery state of health, and no further description is given here. The modules in the above-mentioned prediction device for the state of health of the vehicle-mounted battery may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 9. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured 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, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing the prediction data of the vehicle-mounted battery health state. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer equipment also comprises an input/output interface, wherein the input/output interface is a connecting circuit for exchanging information between the processor and the external equipment, and the input/output interface is connected with the processor through a bus and is called as an I/O interface for short. The computer program, when executed by a processor, implements a method for predicting the state of health of a vehicle-mounted battery.
It will be appreciated by those skilled in the art that the structure shown in fig. 9 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application applies, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory storing a computer program and a processor implementing the steps of the method embodiments described above when the processor executes the computer program.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the respective method embodiments described above.
In one embodiment, a computer program product is provided, on which a computer program is stored, which computer program is executed by a processor for performing the steps of the various method embodiments described above.
It should be noted that, user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party.
Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples represent only a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (10)
1. A method for predicting the state of health of a vehicle-mounted battery, the method comprising:
acquiring the travel frequency of a sample vehicle;
if the travel frequency of the sample vehicle is high, selecting a discharge process with high discharge depth from a plurality of discharge processes of the sample vehicle, and calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the discharge process with high discharge depth;
If the travel frequency of the sample vehicle is low, selecting a discharge process with low discharge depth and high charge amplitude from a plurality of discharge processes of the sample vehicle, and calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to data of charge processes adjacent to the discharge process with low discharge depth and high charge amplitude;
according to the data of the discharging process with high discharging depth and/or the data of the discharging process with low discharging depth and high charging amplitude, an input sequence is obtained;
obtaining an output sequence according to the maximum discharge capacity corresponding to the discharge process with high discharge depth and/or the maximum discharge capacity corresponding to the discharge process with low discharge depth and high charge amplitude;
training a target model according to the input sequence and the output sequence; and predicting the health state of the vehicle-mounted battery according to the maximum discharge capacity predicted by the target model.
2. The method according to claim 1, wherein calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle during the corresponding discharge process from the data of the discharge process with a high discharge depth comprises:
according to the current and the voltage in the data of the discharging process with high discharging depth, calculating the energy released by corresponding SOC consumed by a sample vehicle-mounted battery of the sample vehicle;
And according to the proportional relation between the corresponding SOC consumed by the sample vehicle-mounted battery of the sample vehicle and the released energy, obtaining the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle.
3. The method of claim 2, wherein after calculating the energy released by the respective SOC for the sample on-board battery of the sample vehicle, the method further comprises:
and calculating the current health state of the sample vehicle-mounted battery of the sample vehicle according to a fitting slope and a correlation coefficient by a linear regression algorithm, wherein the fitting slope and the correlation coefficient are the fitting slope and the correlation coefficient between the corresponding SOC consumed by the sample vehicle-mounted batteries of a plurality of groups of sample vehicles and the released energy.
4. The method according to claim 1, wherein calculating the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle during the corresponding discharge process from data of the charge process adjacent to the discharge process having the low discharge depth and the high charge amplitude comprises:
calculating the maximum charge capacity and the battery coulomb efficiency coefficient of a sample vehicle-mounted battery of the sample vehicle according to the data of the charge process adjacent to the discharge process with low discharge depth and high charge amplitude;
And multiplying the maximum charge capacity by a battery coulomb efficiency coefficient to obtain the maximum discharge capacity of the sample vehicle-mounted battery of the sample vehicle.
5. The method of claim 4, wherein calculating the maximum charge capacity of the sample in-vehicle battery of the sample vehicle from data of charge processes adjacent to the discharge process with the low depth of discharge and the high magnitude of charge comprises:
for the data of the adjacent charging process of the discharging process with low discharging depth and high charging amplitude, a sliding window method is adopted, and the maximum charging capacity corresponding to a single window is calculated according to the data in the single window; the window is a time window;
after sliding the windows for a plurality of times, obtaining the maximum charging capacity corresponding to each window, and eliminating abnormal values in the maximum charging capacity corresponding to each window;
and taking an average value of the plurality of maximum charging capacities after the abnormal values are removed, and obtaining the maximum charging capacity of the sample vehicle-mounted battery of the sample vehicle.
6. The method of claim 4, wherein calculating a battery coulombic efficiency coefficient of a sample on-board battery of the sample vehicle from data of charging processes adjacent to the discharging process with the low depth of discharge and the high magnitude of charge comprises:
When the variation range of the SOC of the sample vehicle-mounted battery of the sample vehicle is the same, according to the discharge process data with low discharge depth and high charge amplitude and the data of the charge processes adjacent to the discharge processes with low discharge depth and high charge amplitude;
the ratio of the released charge amount to the absorbed charge amount is calculated to obtain the battery coulomb efficiency coefficient.
7. The method of claim 1, wherein the discharging process data and the charging process data comprise:
sampling time, sample vehicle-mounted battery SOC of the sample vehicle, voltage of the sample vehicle-mounted battery of the sample vehicle, current of the sample vehicle-mounted battery of the sample vehicle, temperature of the sample vehicle-mounted battery of the sample vehicle, and driving range of the sample vehicle.
8. A prediction apparatus for a state of health of a vehicle-mounted battery, the apparatus comprising:
the frequency acquisition module is used for acquiring the travel frequency of the sample vehicle;
the calculation module is used for selecting a discharge process with high discharge depth from a plurality of discharge processes of the sample vehicle if the travel frequency of the sample vehicle is high, and calculating the maximum discharge capacity of the vehicle-mounted battery of the sample vehicle in the corresponding discharge process according to the data of the discharge process with high discharge depth;
The calculation module is further used for selecting a discharging process with low discharging depth and high charging amplitude from a plurality of discharging processes of the sample vehicle if the travel frequency of the sample vehicle is low, and calculating the maximum discharging capacity of the sample vehicle-mounted battery of the sample vehicle in the corresponding discharging process according to the data of the charging processes adjacent to the discharging process with low discharging depth and high charging amplitude;
the training module is used for obtaining an input sequence according to the data of the discharge process with high discharge depth and/or the data of the discharge process with low discharge depth and high charge amplitude;
the training module is also used for obtaining an output sequence according to the maximum discharge capacity corresponding to the discharge process with high discharge depth and/or the maximum discharge capacity corresponding to the discharge process with low discharge depth and high charge amplitude;
the prediction module is used for training to obtain a target model according to the input sequence and the output sequence; and predicting the health state of the vehicle-mounted battery according to the maximum discharge capacity predicted by the target model.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the method of any one of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310012734.0A CN116203427A (en) | 2023-01-05 | 2023-01-05 | Method, device, equipment and storage medium for predicting vehicle-mounted battery health state |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310012734.0A CN116203427A (en) | 2023-01-05 | 2023-01-05 | Method, device, equipment and storage medium for predicting vehicle-mounted battery health state |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116203427A true CN116203427A (en) | 2023-06-02 |
Family
ID=86518371
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310012734.0A Pending CN116203427A (en) | 2023-01-05 | 2023-01-05 | Method, device, equipment and storage medium for predicting vehicle-mounted battery health state |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116203427A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116879759A (en) * | 2023-09-06 | 2023-10-13 | 深圳闻储创新科技有限公司 | SOH correction method, battery manager, storage medium and energy storage device |
-
2023
- 2023-01-05 CN CN202310012734.0A patent/CN116203427A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116879759A (en) * | 2023-09-06 | 2023-10-13 | 深圳闻储创新科技有限公司 | SOH correction method, battery manager, storage medium and energy storage device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220114421A1 (en) | Method for estimating the capacity of lithium battery based on convolution long-short-term memory neural network | |
| Xu et al. | Improving the state of charge estimation of reused lithium-ion batteries by abating hysteresis using machine learning technique | |
| Zhao et al. | Elman neural network using ant colony optimization algorithm for estimating of state of charge of lithium-ion battery | |
| CN112213643B (en) | Method, system and equipment for predicting initial capacity and state of health of battery | |
| CN113805064A (en) | Lithium ion battery pack health state prediction method based on deep learning | |
| CN113030764A (en) | Battery pack health state estimation method and system | |
| CN114818831B (en) | Bidirectional lithium ion battery fault detection method and system based on multi-source perception | |
| CN113391211B (en) | Method for predicting remaining life of lithium battery under small sample condition | |
| CN114290960B (en) | Method and device for acquiring battery health of power battery and vehicle | |
| CN116609676B (en) | Method and system for monitoring state of hybrid energy storage battery based on big data processing | |
| CN115201686B (en) | Lithium ion battery health state assessment method under incomplete charge and discharge data | |
| CN116203427A (en) | Method, device, equipment and storage medium for predicting vehicle-mounted battery health state | |
| CN112834927A (en) | Lithium battery residual life prediction method, system, device and medium | |
| Zheng et al. | State of health estimation for lithium battery random charging process based on CNN-GRU method | |
| CN115308608A (en) | All-vanadium redox flow battery voltage prediction method, device and medium | |
| CN113917336A (en) | Lithium ion battery health state prediction method based on segment charging time and GRU | |
| CN116125306A (en) | Power battery thermal runaway diagnosis method and system based on hybrid prediction framework | |
| CN116400241A (en) | Battery health assessment method and device based on full life cycle operation data | |
| CN115684936A (en) | Power battery health failure probability calculation and health failure map establishment method based on machine learning | |
| CN115015789A (en) | Lithium ion battery health state estimation method and device based on incremental capacitance curve | |
| Wang et al. | Fast modeling of the capacity degradation of lithium-ion batteries via a conditional temporal convolutional encoder–decoder | |
| CN112557908A (en) | SOC and SOH joint estimation method for lithium ion power battery | |
| CN113406522A (en) | Method and system for predicting and evaluating service life of lithium battery system of electric vehicle | |
| CN110658461B (en) | Random charge and discharge battery capacity prediction method based on double e index model | |
| CN114200328B (en) | Lithium ion battery SOC estimation method under non-Gaussian Levy noise |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |