CN112798962A - Battery hysteresis model training method, and method and device for estimating battery SOC - Google Patents
Battery hysteresis model training method, and method and device for estimating battery SOC Download PDFInfo
- Publication number
- CN112798962A CN112798962A CN202110277793.1A CN202110277793A CN112798962A CN 112798962 A CN112798962 A CN 112798962A CN 202110277793 A CN202110277793 A CN 202110277793A CN 112798962 A CN112798962 A CN 112798962A
- Authority
- CN
- China
- Prior art keywords
- battery
- current
- hysteresis model
- training
- circuit voltage
- 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.)
- Granted
Links
- 238000012549 training Methods 0.000 title claims abstract description 90
- 238000000034 method Methods 0.000 title claims abstract description 80
- 238000012360 testing method Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 8
- 230000010354 integration Effects 0.000 claims description 6
- 238000002372 labelling Methods 0.000 claims description 4
- 238000004146 energy storage Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 7
- 238000007599 discharging Methods 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 238000002790 cross-validation Methods 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 1
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 description 1
- 239000002253 acid Substances 0.000 description 1
- OJIJEKBXJYRIBZ-UHFFFAOYSA-N cadmium nickel Chemical compound [Ni].[Cd] OJIJEKBXJYRIBZ-UHFFFAOYSA-N 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011038 discontinuous diafiltration by volume reduction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 229910052744 lithium Inorganic materials 0.000 description 1
- 229910001416 lithium ion Inorganic materials 0.000 description 1
- GELKBWJHTRAYNV-UHFFFAOYSA-K lithium iron phosphate Chemical compound [Li+].[Fe+2].[O-]P([O-])([O-])=O GELKBWJHTRAYNV-UHFFFAOYSA-K 0.000 description 1
- 229910052987 metal hydride Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 229920000642 polymer Polymers 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- 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]
- G01R31/385—Arrangements for measuring battery or accumulator variables
- G01R31/387—Determining ampere-hour charge capacity or SoC
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
Abstract
The application provides a battery hysteresis model training method, which comprises the following steps: collecting the battery from t on line0Time to the tkOperating current I of timek(ii) a Calculating the current integral quantity Q (t) according to the working currentk) Wherein, in the step (A),
if the battery is determined to be at t according to the working currentkConstantly satisfying the condition of collecting open circuit voltage, collecting the battery at tkOpen circuit voltage OCV (t) at timek) (ii) a Collecting said battery at tkTemperature at time Tbat(tk) (ii) a According to the current integral quantity Q (t)k) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Constructing a sample set; and training the battery hysteresis model according to the sample set. The application also provides a method and a device for estimating the SOC of the battery. The method and the device can obtain the open-circuit voltage of the battery on line according to the current and the temperature acquired in real time.
Description
Technical Field
The application relates to the technical field of batteries, in particular to a battery hysteresis model training method, and a method and a device for estimating battery SOC.
Background
State of Charge (SOC) estimation of a battery is one of core functions of a battery management system. Accurate SOC estimation can ensure safe and reliable operation of the battery system, optimize the battery system, and provide basis for energy management and safety management of the power-consuming devices (e.g., electric vehicles). In many conventional SOC estimation methods, SOC and Open Circuit Voltage (OCV) are obtained by using a corresponding relationship. However, the open-circuit voltage and the OCV in the battery do not completely correspond to each other one to one, and a hysteresis relationship exists. Therefore, when the SOC of the battery is estimated using the open circuit voltage, the influence of the hysteresis characteristic on the OC of the battery needs to be considered. The existing modeling method for the hysteresis characteristic of the open-circuit voltage of the battery introduces more simplification, so that the modeling precision of a hysteresis model is low, and the SOC estimation is influenced. Therefore, meeting the requirement of precision and obtaining the battery SOC in real time are always problems to be solved urgently in the industry.
Disclosure of Invention
In view of the above, there is a need to provide a method and an apparatus for training a hysteresis model of a battery, and a method and an apparatus for estimating SOC of a battery, which can obtain an open-circuit voltage of the battery on-line according to a current and a temperature collected in real time.
An embodiment of the application provides a battery hysteresis model training method, which collects the battery hysteresis time t on line0Time to the tkOperating current I of timekWherein k is greater than 0; calculating from said t from said operating current0Time to the tkIntegral quantity of current Q (t) at timek) Wherein, in the step (A),
if the battery is determined to be at t according to the working currentkConstantly satisfying the condition of collecting open circuit voltage, collecting the battery at tkOpen circuit voltage OCV (t) at timek) (ii) a Collecting said battery at tkTemperature at time Tbat(tk) (ii) a According to the current integral quantity Q (t)k) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Constructing a sample set; and training the battery hysteresis model according to the sample set.
According to the present applicationIn some embodiments, the sample set comprises a positive sample set and a negative sample set, and the integral quantity Q (t) is based on the currentk) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Constructing a sample set includes: obtaining the current integral quantity Q (t) of the positive samples in the positive sample setk) Temperature Tbat(tk) And open circuit voltage OCV (t)k) And the current integral quantity Q (t) of the negative samples in the negative sample setk) Temperature Tbat(tk) And open circuit voltage OCV (t)k) (ii) a Integrating the current of the positive sample by an amount Q (t)k) Temperature Tbat(tk) And open circuit voltage OCV (t)k) Labeling the class data to integrate the current of the positive sample by Q (t)k) Temperature Tbat(tk) And open circuit voltage OCV (t)k) Carrying a category label.
According to some embodiments of the present application, the training of the battery hysteresis model according to the sample set comprises: generating a sample training set and a sample testing set according to the sample set; training the battery hysteresis model according to the sample training set, and verifying the accuracy of the trained battery hysteresis model according to the sample testing set; and if the accuracy is greater than or equal to the preset accuracy, ending the training process of the battery hysteresis model.
According to some embodiments of the present application, the training the battery hysteresis model according to the sample set comprises further comprising: if the accuracy is less than the preset accuracy, increasing the number of the sample training sets to retrain the battery hysteresis model until the accuracy is greater than or equal to the preset accuracy.
According to some embodiments of the present application, the generating of the sample training set and the sample testing set from the sample set comprises: randomly selecting a first preset number of sample training sets from the generated sample training sets for training; randomly selecting a second preset number of sample test sets in the generated sample test sets for verification.
According to some embodiments of the present application, if the operating current is at a first predetermined levelKeeping smaller than a first preset value in a time period, and determining to be at tkThe battery meets the open-circuit voltage acquisition condition at the moment, wherein the starting time point of the first preset time period is tk-iThe ending time point is tk，0＜i＜k。
According to some embodiments of the application, the method further comprises: if the working current is greater than or equal to a second preset value within a second preset time period, ending the collection of the open-circuit voltage of the battery; and continuously collecting the working current of the battery, and updating historical data according to the collected working current.
According to some embodiments of the present application, the first preset value is related to at least one of a battery capacity or a battery temperature.
According to some embodiments of the present application, the battery hysteresis model includes an input layer, a hidden layer, and an output layer.
An embodiment of the present application provides a method for estimating a battery SOC by using a hysteresis model trained by the battery hysteresis model training method, the method including: collecting the battery from t on line0Time to the tkOperating current I of timekWherein k is greater than 0; calculating from said t from said operating current0Time to the tkIntegral quantity of current Q (t) at timek) (ii) a Obtaining the battery at tkTemperature at time Tbat(tk) (ii) a Inputting the current integral quantity Q (t)k) And temperature Tbat(tk) To the battery hysteresis model to obtain the battery at tkOpen circuit voltage at time; and inquiring the SOC-OCV corresponding relation according to the open-circuit voltage to obtain the state of charge of the battery.
According to some embodiments of the present application, the SOC-OCV correspondence is obtained by combining an ampere-hour integration method and an open-circuit voltage method.
An embodiment of the present application provides an electrical device comprising a memory and a processor for implementing a battery hysteresis model training method as described above or a method of estimating a battery SOC as described above when executing a computer program stored in the memory.
According to some embodiments of the application, the power consuming device comprises an energy storage device, or a two or more wheeled electric vehicle, or a drone, or a power tool.
According to the embodiment of the application, the working current data of the battery are collected on line, and the working current data are abstracted into the current integral quantity. And training by taking the current integral quantity and the battery temperature and the open-circuit voltage acquired at the current moment as sample data to obtain the battery hysteresis model. In the using process, the current integral quantity can be calculated according to the current collected in real time, the current integral quantity and the collected battery temperature are jointly used as input quantities, the corresponding open-circuit voltage is obtained through the trained battery hysteresis model, and the nuclear power state of the battery is obtained according to the open-circuit voltage. The method does not carry out off-line experiments in the modeling process, and the input quantity of the battery hysteresis model is simple, so that the requirements on the computing capacity and the storage capacity of the electric device are low. The method has the advantages of low experimental amount, low calculation amount and low storage amount while the accuracy is acceptable.
Drawings
Fig. 1 is a schematic structural diagram of an electric device according to an embodiment of the present application.
Fig. 2 is a flow chart of a battery hysteresis model training method according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a battery hysteresis model according to an embodiment of the present application.
FIG. 4 is a flow chart of a method of estimating battery SOC according to an embodiment of the present application.
Description of the main elements
Power utilization device 1
Memory 11
Processor 12
Battery 13
Collection device 14
Time-meter 15
The following detailed description will explain the present application in further detail in conjunction with the above-described figures.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application.
Referring to fig. 1, fig. 1 is a schematic view of an electric device according to an embodiment of the present disclosure. Referring to fig. 1, the power utilization apparatus 1 includes, but is not limited to, a memory 11, at least one processor 12, a battery 13, a collection device 14, and a timer 15, and the above elements may be connected via a bus or directly.
In one embodiment, the battery 13 is a rechargeable battery for providing electrical energy to the electrical device 1. For example, the battery 13 may be a lead-acid battery, a nickel-cadmium battery, a nickel-metal hydride battery, a lithium ion battery, a lithium polymer battery, a lithium iron phosphate battery, or the like. The Battery 13 is logically connected to the processor 12 through a Battery Management System (BMS), so that functions such as charging and discharging are realized through the Battery Management System. The battery management System CAN be in communication connection with a Power Conversion System (PCS) through CAN or RS 485. The battery 13 includes cells (not shown) that can be repeatedly recharged in a rechargeable fashion.
In the present embodiment, the collecting device 14 includes an analog-to-digital converter for collecting the voltage of the battery 13 and the current of the battery 13, and a thermometer for collecting the temperature of the battery 13. It is understood that the collection device 14 may also be other voltage collection devices and current collection devices. The timer 15 is used for recording the working time of the battery 13. It is understood that the powered device 1 may also include other devices, such as pressure sensors, light sensors, gyroscopes, hygrometers, infrared sensors, etc.
Fig. 1 is a diagram illustrating the electric power consuming apparatus 1. In other embodiments, the powered device 1 may include more or fewer elements, or have a different configuration of elements. The electric device 1 may be an energy storage product, or an electric tool, or a cleaning tool, a cargo unicycle or 2-wheel or more electric vehicle, an unmanned aerial vehicle, or any other suitable rechargeable device.
Although not shown, the electric device 1 may further include a Wireless Fidelity (WiFi) unit, a bluetooth unit, a speaker, and other components, which are not described in detail herein.
Referring to fig. 2, fig. 2 is a flowchart illustrating a battery hysteresis model training method according to an embodiment of the present disclosure. The battery hysteresis model training method is used for establishing a battery hysteresis model, and the state of charge of the battery 13 can be estimated on line through the battery hysteresis model, and the battery hysteresis model training method can comprise the following steps:
step S21: on-line collecting battery 13 from t0Time to the tkOperating current I of timekWherein k is greater than 0.
In the present embodiment, the operating current of the battery 13 may be collected by the collecting device 14. For example, the collecting device 14 is a hall current sensor.
For example, t is acquired0Operating current at time I0，t1Operating current at time I1，t2Operating current at time I2，tkOperating current at time Ik. A current information sequence can be generated according to the collected working current and time information, and the current information sequence is { I }0，I1，I2...Ik}。
Step S22: calculating from said t from said operating current0Time to the tkIntegral quantity of current Q (t) at timek)。
The method aims to solve the problems that in the existing method, the experiment time consumption is high due to the fact that experiment data need to be acquired offline, or the calculation capacity requirement is high due to the fact that a large amount of recent historical current data train the battery hysteresis model on line. According to the battery hysteresis model training method, the working current data acquired in real time can be abstracted into a parameter (such as current integral quantity), and the parameter and the battery temperature acquired at the current moment are jointly used as data for constructing a training sample, so that the battery hysteresis model is trained. The technical effects of on-line calculation and data volume reduction are achieved.
In the present embodiment, the current integration quantity Q (t)k) The calculation formula of (2) is as follows:
i.e. by calculating the battery from the t0Time to the tkThe electric quantity at the moment is taken as sample data. It should be noted that the operating current I is takenkIs positive.
Step S23: determining that the battery 13 is at tkWhether the open-circuit voltage acquisition condition is met at any moment. If it is determined that the battery is at tkThe open-circuit voltage acquisition condition is satisfied at any time, and the flow proceeds to step S24; if it is determined that the battery is at tkThe time does not satisfy the open circuit voltage collection condition, and the flow returns to step S21.
In the present embodiment, it is determined that the battery 13 is at t by determining whether the operating current of the battery remains less than a first preset value for a first preset time period Δ t1kWhether the open-circuit voltage acquisition condition is met at any moment. Wherein the starting time point of the delta t1 is tiThe ending time point is tkAnd i is more than 0 and less than k. If the working current of the battery 13 is kept smaller than a first preset value within a first preset time period deltat 1, determining that the battery is at tkThe open-circuit voltage acquisition condition is satisfied at any time, and the flow proceeds to step S24; if the working current of the battery 13 is not kept smaller than the first preset value within a first preset time period, determining that the battery is in tkThe time does not satisfy the open circuit voltage collection condition, and the flow returns to step S21.
In the present embodiment, it is determined whether the battery 13 is in a static state by determining whether the working current of the battery 13 remains smaller than a first preset value within a first preset time period, so as to satisfy the collection condition of the open-circuit voltage. In general, after the charge and discharge of the battery 13 are completed, the battery 13 is left at rest, and the open circuit voltage of the battery 13 may be collected while the battery 13 is left at rest. While during the rest of the battery 13, the electrical operating flow of the battery 13 is kept at a small value (for example zero). In this embodiment, when the working current of the battery 13 is set to be smaller than the first preset value within the first preset time period, the battery 13 enters a static state, and the collection condition of the open-circuit voltage of the battery 13 is satisfied. In this embodiment, the first preset value is related to at least one of a battery capacity or a battery temperature.
Step S24: collecting the battery at tkOpen circuit voltage OCV (t) at timek)。
In this embodiment, if the working current of the battery 13 is kept smaller than a first preset value within a first preset time period, it is determined that the battery 13 is in a static state, and the working current of the battery 13 at the time t may be collectedkThe open circuit voltage at the moment.
Step S25: obtaining the battery at tkTemperature at time Tbat(tk)。
In the present embodiment, the battery is collected at t by the collecting device 14kTemperature at time Tbat(tk). Because the temperature of the battery has a large influence on the performance of the battery, the temperature of the battery is used as sample data for training the hysteresis model of the battery, and a more accurate open-circuit voltage value can be obtained.
Step S26: according to the current integral quantity Q (t)k) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) A sample set is constructed, wherein the sample set includes a positive sample set and a negative sample set.
In the present embodiment, in order to train the battery hysteresis model, it is necessary to construct training data required for the battery hysteresis model, and then construct a sample set from the training data. The training digit comprises the integrated quantity of current Q (t) of the battery 13k) The temperature Tbat(tk) And said open circuit voltage OCV (t)k)。
In particular, said basisThe integrated current Q (t) of the battery 13k) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Constructing a sample set includes: obtaining the current integral quantity Q (t) of the positive samples in the positive sample setk) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) And the current integral quantity Q (t) of the negative samples in the negative sample setk) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) (ii) a Integrating the current of the positive sample by an amount Q (t)k) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Labeling the class data to integrate the current of the positive sample by the amount Q (t)k) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Carrying a category label. For example, 500 open-circuit voltage data corresponding to the current integral quantity and the temperature at different times are respectively selected, and category data are labeled to each open-circuit voltage data. Can use "1" as t0An open circuit voltage data tag corresponding to the current integral quantity and the temperature at the moment takes '2' as t1An open circuit voltage data tag corresponding to the current integral quantity and the temperature at the moment takes '3' as t2And the current integral quantity at the moment and the open-circuit voltage data label corresponding to the temperature.
As an alternative embodiment, in order to make the battery hysteresis model more intelligent, the negative samples input into the battery hysteresis model can be identified, and the integral quantity Q (t) of the current of the positive samples in the positive sample set can be obtainedk) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) And the current integral quantity Q (t) of the negative samples in the negative sample setk) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Then; integrating the current of the negative sample by the quantity Q (t)k) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Labeling the class data to make the current integral quantity Q (t) of the negative samplek) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Carrying a category label. It will be appreciated that for the samplePositive samples of a set carry class labels from which they can be identified by the battery hysteresis model; class labels are also carried on negative examples of the sample set, and the negative examples can be identified according to the class labels through the battery hysteresis model. However, in general, only the class label corresponding to the positive sample input into the battery hysteresis model needs to be identified, so as to obtain an accurate output result according to the battery hysteresis model.
Step S27: and training the battery hysteresis model according to the sample set.
In this embodiment, after training data is acquired and the sample set is constructed from the training data, the battery hysteresis model is trained from the sample set.
Specifically, the training of the battery hysteresis model according to the sample set comprises:
(1) and generating a sample training set and a sample testing set according to the sample set.
In the present embodiment, the battery hysteresis model may be trained by dividing a structured sample set into a sample training set and a sample testing set in an appropriate ratio by using a cross validation (CrossValidation) concept. For example, a suitable division ratio is 6: 4.
Further, if the total number of the divided sample training sets is still large, that is, all the sample training sets are used for training the battery hysteresis model, the cost of searching for the parameter corresponding to the battery hysteresis model is large. Thus, the generating the sample training set may further comprise: randomly selecting a first preset number of sample training sets from the generated sample training sets for training.
In the preferred embodiment, in order to increase the randomness of the training set of the training samples, a random number generation algorithm may be used for random selection.
In the preferred embodiment, the first predetermined number may be a predetermined fixed value, for example, 500, that is, 500 samples are randomly selected from the generated sample training set for training the battery hysteresis model. The first predetermined number may also be a preset scaling value, e.g., 1/10, i.e., 1/10 scaled samples are randomly selected from the generated training set of samples for training of the battery hysteresis model.
(2) And training the battery hysteresis model according to the sample training set, and verifying the accuracy of the trained battery hysteresis model according to the sample testing set.
In this embodiment, the sample training set is used to train the battery hysteresis model, and the sample testing set is used to test the performance of the battery hysteresis model.
In this embodiment, the training samples in the sample training set are distributed to different folders. For example, let t0Distributing training samples of current integral quantity and temperature at the moment to a first folder t1The training samples of the current integral and temperature at the moment are distributed to a second folder t2A training sample of the integrated amount of current at time and temperature is distributed into a third folder. Then, training samples with a first preset proportion (for example, 70%) are respectively extracted from different folders to serve as total training samples to perform training of the battery hysteresis model, and training samples with a second remaining preset proportion (for example, 30%) are respectively extracted from different folders to serve as total test samples to perform accuracy verification on the trained battery hysteresis model.
In the present embodiment, as shown in fig. 3, the battery hysteresis model includes an input layer, a hidden layer, and an output layer. During the subsequent charge and discharge cycles of the battery 13, the integrated current amount and the temperature at different times are input into the battery hysteresis model, and the corresponding open-circuit voltage can be output.
(3) And determining whether the accuracy is greater than or equal to a preset accuracy.
In the embodiment, if the accuracy of the test is higher, the battery hysteresis model is better in performance; if the accuracy of the test is low, the battery hysteresis model trained by the method is poor in performance. And determining whether to train a battery hysteresis model with good performance or not by determining whether the accuracy is greater than or equal to a preset accuracy. If the accuracy is greater than or equal to the preset accuracy, the trained battery hysteresis model is confirmed to have good performance, and the process enters the step (4); and (5) if the accuracy is less than the preset accuracy, confirming that the performance of the trained battery hysteresis model is not good, and carrying out the process.
(4) And finishing the training process of the battery hysteresis model.
In this embodiment, if the accuracy is greater than or equal to the preset accuracy, it is determined that the trained battery hysteresis model has good performance and meets the requirements, and the training process of the battery hysteresis model is ended.
(5) Increasing the number of sample training sets to retrain the battery hysteresis model until the accuracy is greater than or equal to the preset accuracy.
In this embodiment, if the accuracy is smaller than the preset accuracy, it is determined that the trained battery hysteresis model has poor performance and does not meet the requirement, the number of the sample training sets needs to be increased to retrain the battery hysteresis model until the accuracy is greater than or equal to the preset accuracy, and a battery hysteresis model meeting the requirement is obtained.
In this embodiment, the battery hysteresis model training method further includes:
determining whether the working current of the battery 13 is greater than or equal to a second preset value within a second preset time period Δ t2, and if the working current of the battery 13 is greater than or equal to the second preset value within the second preset time period Δ t2, ending the collection of the open-circuit voltage of the battery 13; and continuously collecting the working current of the battery 13, and updating historical data according to the collected working current. If the working current of the battery 13 is continuously kept smaller than the second preset value within the second preset time period Δ t2, continuously determining whether the current of the battery 13 is greater than or equal to the second preset value within a third preset time period Δ t 3.
It should be noted that the second preset time period is the time after the first preset time period, and the third preset time period is the time after the first preset time periodThe time after the second preset period of time, and so on. The starting time point of Δ t2 is tjThe ending time point is tnJ is more than k and less than n; the starting time point of Δ t3 is tlThe ending time point is tm，n＜l＜m。
In the present embodiment, it may be determined whether the open-circuit voltage collection exit condition of the battery 13 is satisfied by determining whether the operating current of the battery 13 is greater than or equal to a second preset value within a second preset time period. If the working current of the battery 13 is greater than or equal to a second preset value within a second preset time period, determining that the battery 13 enters a charging and discharging process, and ending the collection of the open-circuit voltage of the battery 13; and continuously collecting the working current of the battery 13, and calculating and updating historical data according to the collected working current.
It should be noted that the second preset value is a larger current value, and can be used to determine that the battery 13 enters a charging and discharging process.
Referring to fig. 4, fig. 4 is a flowchart illustrating a method for estimating a state of charge of the battery 13 according to an embodiment of the present disclosure. The method for estimating the state of charge of the battery 13 specifically includes the following steps, and the order of the steps in the flowchart may be changed and some steps may be omitted according to different requirements.
Step S31: on-line collecting battery 13 from t0Time to the tkOperating current I of timekWherein k is greater than 0.
In the present embodiment, during the cyclic charge and discharge of the battery 13, the operating current of the battery 13 is collected in real time by the collecting device 14.
Step S32: calculating from said t from said operating current0Time to the tkIntegral quantity of current Q (t) at timek). In the present embodiment, the current integration quantity Q (t)k) The calculation formula of (2) is as follows:
i.e. by calculating the battery from the t0Time to the tkThe electric quantity at the moment is taken as sample data.It should be noted that the operating current I is takenkIs positive.
Step S33: obtaining the battery at tkTemperature at time Tbat(tk)。
Step S34: inputting the current integral quantity Q (t)k) And temperature Tbat(tk) To the battery hysteresis model to obtain the battery at tkThe open circuit voltage at the moment.
In the present embodiment, the current integration quantity Q (t) is inputk) And temperature Tbat(tk) To the trained battery hysteresis model, the battery hysteresis model outputs t according to the currentkOpen circuit voltage OCV (t) at timek)。
Step S35: and inquiring the SOC-OCV corresponding relation according to the open-circuit voltage to obtain the state of charge of the battery 13.
In the present embodiment, the power consumption device 1 stores an SOC-OCV correspondence relationship in advance. It should be noted that, when the battery system is determined, the SOC-OCV correspondence of the battery is usually fixed and unchanged. Even if the battery is subjected to cyclic charge and discharge for several times, the SOC-OCV corresponding relation of the battery is not changed.
Specifically, the state of charge (SOC) -Open Circuit Voltage (OCV) correspondence of the battery may be obtained by:
1) taking a battery, charging the battery to a full charge state, and then discharging the battery to an emptying state by using a first preset current; in this embodiment, the first predetermined current is a small-rate current, such as 0.01C, but may be other currents.
2) And recording the voltage and capacity changes of the battery in the charging and discharging processes.
3) And acquiring the charge state of the battery in the discharging process. For example, the maximum discharge capacity of the battery is taken as the full-load capacity of the battery, and the value of the capacity of the battery changing with time during discharge is divided by the full-load capacity to obtain the state of charge of the battery during discharge.
4) And respectively establishing corresponding relations of the battery voltages of the battery in different charge states in the discharging process to obtain the SOC-OCV corresponding relation of the battery.
In another embodiment, the SOC-OCV correspondence relationship may also be obtained by combining an ampere-hour integration method and an open-circuit voltage method.
The application provides a battery hysteresis model training method with low experimental amount and low calculation amount on the premise of acceptable precision. According to the method, the working current data of the battery are acquired on line, and the working current data are abstracted into a current integral quantity. And training by taking the current integral quantity and the battery temperature and the open-circuit voltage acquired at the current moment as sample data to obtain the battery hysteresis model. In the using process, the current integral quantity can be calculated according to the current collected in real time, the current integral quantity and the collected battery temperature are jointly used as input quantities, the corresponding open-circuit voltage is obtained through the trained battery hysteresis model, and the nuclear power state of the battery is obtained according to the open-circuit voltage. The method does not carry out off-line experiments in the modeling process, and the input quantity of the battery hysteresis model is simple, so that the requirements on the computing capacity and the storage capacity of the electric device are low. The method has the advantages of low experimental amount, low calculation amount and low storage amount while the accuracy is acceptable.
Referring to fig. 1, in the present embodiment, the memory 11 may be an internal memory of an electric device, that is, a memory built in the electric device. In other embodiments, the memory 11 may also be an external memory of the electric device, i.e. a memory externally connected to the electric device.
In some embodiments, the memory 11 is used for storing program codes and various data, and realizes high-speed and automatic access to programs or data during the operation of the electric device.
The memory 11 may include random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
In one embodiment, the Processor 12 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete gate or transistor logic device, discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any other conventional processor or the like.
The program code and various data in the memory 11 may be stored in a computer-readable storage medium if they are implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the present application may also implement all or part of the processes in the methods of the embodiments, for example, implement the steps in the method for detecting a short circuit in a battery, and may also be implemented by using a computer program to instruct related hardware, where the computer program may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the embodiments of the methods may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), or the like.
It is understood that the above described module division is a logical function division, and there may be other division ways in actual implementation. In addition, functional modules in the embodiments of the present application may be integrated into the same processing unit, or each module may exist alone physically, or two or more modules are integrated into the same unit. The integrated module can be realized in a hardware form, and can also be realized in a form of hardware and a software functional module.
Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present application and not for limiting, and although the present application is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions can be made on the technical solutions of the present application without departing from the spirit and scope of the technical solutions of the present application.
Claims (13)
1. A method for training a hysteresis model of a battery, the method comprising:
collecting the battery from t on line0Time to the tkOperating current I of timekWherein k is>0；
Calculating from said t from said operating current0Time to the tkIntegral quantity of current Q (t) at timek) Wherein, in the step (A),
if the battery is determined to be at t according to the working currentkConstantly satisfying the condition of collecting open circuit voltage, collecting the battery at tkOpen circuit voltage OCV (t) at timek)；
Obtaining the battery at tkTemperature at time Tbat(tk)；
According to the current integral quantity Q (t)k) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Constructing a sample set; and
and training the battery hysteresis model according to the sample set.
2. The method of claim 1, wherein the sample set comprises a positive sample set and a negative sample set, and wherein the integral quantity Q (t) is calculated according to the currentk) The temperature Tbat(tk) And said open circuit voltage OCV (t)k) Constructing a sample set comprising：
Obtaining the current integral quantity Q (t) of the positive samples in the positive sample setk) Temperature Tbat(tk) And open circuit voltage OCV (t)k) And the current integral quantity Q (t) of the negative samples in the negative sample setk) Temperature Tbat(tk) And open circuit voltage OCV (t)k)；
Integrating the current of the positive sample by an amount Q (t)k) Temperature Tbat(tk) And open circuit voltage OCV (t)k) Labeling the class data to integrate the current of the positive sample by Q (t)k) Temperature Tbat(tk) And open circuit voltage OCV (t)k) Carrying a category label.
3. The battery hysteresis model training method of claim 2, wherein said training the battery hysteresis model according to the sample set comprises:
generating a sample training set and a sample testing set according to the sample set;
training the battery hysteresis model according to the sample training set, and verifying the accuracy of the trained battery hysteresis model according to the sample testing set; and
and if the accuracy is greater than or equal to the preset accuracy, ending the training process of the battery hysteresis model.
4. The battery hysteresis model training method of claim 3, wherein said training the battery hysteresis model according to the sample set further comprises:
if the accuracy is less than the preset accuracy, increasing the number of the sample training sets to retrain the battery hysteresis model until the accuracy is greater than or equal to the preset accuracy.
5. The method of claim 3, wherein the generating a sample training set and a sample test set from the sample set comprises:
randomly selecting a first preset number of sample training sets from the generated sample training sets for training;
randomly selecting a second preset number of sample test sets in the generated sample test sets for verification.
6. The battery hysteresis model training method of claim 1, wherein if the operating current remains less than a first predetermined value for a first predetermined period of time, determining at tkThe battery meets the open-circuit voltage acquisition condition at the moment, wherein the starting time point of the first preset time period is tk-iThe ending time point is tk，0<i<k。
7. The battery hysteresis model training method of claim 6, the method further comprising:
if the working current is greater than or equal to a second preset value within a second preset time period, ending the collection of the open-circuit voltage of the battery;
and continuously collecting the working current of the battery, and updating historical data according to the collected working current.
8. The battery hysteresis model training method of claim 6, wherein the first predetermined value is related to at least one of battery capacity or battery temperature.
9. The battery hysteresis model training method of claim 1, wherein the battery hysteresis model comprises an input layer, a hidden layer, and an output layer.
10. A method for estimating SOC of a battery using a hysteresis model trained according to the method of any one of claims 1 to 9, the method comprising:
collecting the battery from t on line0Time to the tkOperating current I of timekWherein k is>0；
Calculating from the operating currentSaid t is0Time to the tkIntegral quantity of current Q (t) at timek)；
Obtaining the battery at tkTemperature at time Tbat(tk)；
Inputting the current integral quantity Q (t)k) And temperature Tbat(tk) To the battery hysteresis model to obtain the battery at tkOpen circuit voltage at time; and
and inquiring the SOC-OCV corresponding relation according to the open-circuit voltage to obtain the state of charge of the battery.
11. The method of estimating SOC of a battery according to claim 10, wherein the SOC-OCV correspondence is obtained by combining an ampere-hour integration method and an open-circuit voltage method.
12. An electrical device, comprising:
a memory; and
a processor for implementing the battery hysteresis model training method of any one of claims 1 to 9 or the method of estimating the SOC of a battery of any one of claims 10 to 11 when executing the computer program stored in the memory.
13. The power consumption device of claim 12, wherein the power consumption device comprises an energy storage device, or a two or more wheeled electric vehicle, or a drone, or a power tool.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110277793.1A CN112798962B (en) | 2021-03-15 | 2021-03-15 | Battery hysteresis model training method, method and device for estimating battery SOC |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110277793.1A CN112798962B (en) | 2021-03-15 | 2021-03-15 | Battery hysteresis model training method, method and device for estimating battery SOC |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112798962A true CN112798962A (en) | 2021-05-14 |
| CN112798962B CN112798962B (en) | 2024-04-30 |
Family
ID=75817063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110277793.1A Active CN112798962B (en) | 2021-03-15 | 2021-03-15 | Battery hysteresis model training method, method and device for estimating battery SOC |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112798962B (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201293826Y (en) * | 2008-11-24 | 2009-08-19 | 南开大学 | Energy detector for wireless sensor network node |
| CN203037829U (en) * | 2013-01-23 | 2013-07-03 | 天津雅迪实业有限公司 | Electric car continuous mileage obtaining apparatus |
| CN103616647A (en) * | 2013-12-09 | 2014-03-05 | 天津大学 | Battery remaining capacity estimation method for electric car battery management system |
| CN103675701A (en) * | 2013-11-29 | 2014-03-26 | 宇龙计算机通信科技(深圳)有限公司 | Battery state-of-charge correction method and device |
| CN105203963A (en) * | 2015-09-11 | 2015-12-30 | 同济大学 | Charge state estimation method based on open-circuit voltage hysteretic characteristics |
| CN105699907A (en) * | 2016-01-28 | 2016-06-22 | 广州市香港科大霍英东研究院 | A battery SOC estimation method and system based on dynamic impedance correction |
| CN106646265A (en) * | 2017-01-22 | 2017-05-10 | 华南理工大学 | Method for estimating SOC of lithium battery |
| CN106918787A (en) * | 2017-03-20 | 2017-07-04 | 国网重庆市电力公司电力科学研究院 | A kind of electric automobile lithium battery residual charge evaluation method and device |
| CN107861073A (en) * | 2017-11-06 | 2018-03-30 | 合肥工业大学 | A kind of vehicle-mounted electric quantity of lead-acid storage battery On-line Estimation method based on CHVT models |
| CN109884528A (en) * | 2019-02-25 | 2019-06-14 | 江苏大学 | A kind of lithium ion battery open-circuit voltage evaluation method and system with temperature input |
| CN110045286A (en) * | 2019-03-21 | 2019-07-23 | 江苏大学 | A kind of battery open circuit voltage evaluation method and device based on air-or liquid-operated mechanical model |
| CN110333450A (en) * | 2019-04-30 | 2019-10-15 | 蜂巢能源科技有限公司 | Battery open circuit voltage predictor method and system |
| CN110609232A (en) * | 2019-03-29 | 2019-12-24 | 深圳猛犸电动科技有限公司 | Lithium ion battery SOC estimation method and device and terminal equipment |
| CN110850315A (en) * | 2019-11-29 | 2020-02-28 | 北京邮电大学 | Method and device for estimating state of charge of battery |
| CN111458646A (en) * | 2020-05-25 | 2020-07-28 | 安徽理工大学 | Lithium battery SOC estimation method based on PSO-RBF neural network |
| CN111537887A (en) * | 2020-04-27 | 2020-08-14 | 南京航空航天大学 | Hybrid power system battery open-circuit voltage model optimization method considering hysteresis characteristic |
-
2021
- 2021-03-15 CN CN202110277793.1A patent/CN112798962B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201293826Y (en) * | 2008-11-24 | 2009-08-19 | 南开大学 | Energy detector for wireless sensor network node |
| CN203037829U (en) * | 2013-01-23 | 2013-07-03 | 天津雅迪实业有限公司 | Electric car continuous mileage obtaining apparatus |
| CN103675701A (en) * | 2013-11-29 | 2014-03-26 | 宇龙计算机通信科技(深圳)有限公司 | Battery state-of-charge correction method and device |
| CN103616647A (en) * | 2013-12-09 | 2014-03-05 | 天津大学 | Battery remaining capacity estimation method for electric car battery management system |
| CN105203963A (en) * | 2015-09-11 | 2015-12-30 | 同济大学 | Charge state estimation method based on open-circuit voltage hysteretic characteristics |
| CN105699907A (en) * | 2016-01-28 | 2016-06-22 | 广州市香港科大霍英东研究院 | A battery SOC estimation method and system based on dynamic impedance correction |
| CN106646265A (en) * | 2017-01-22 | 2017-05-10 | 华南理工大学 | Method for estimating SOC of lithium battery |
| CN106918787A (en) * | 2017-03-20 | 2017-07-04 | 国网重庆市电力公司电力科学研究院 | A kind of electric automobile lithium battery residual charge evaluation method and device |
| CN107861073A (en) * | 2017-11-06 | 2018-03-30 | 合肥工业大学 | A kind of vehicle-mounted electric quantity of lead-acid storage battery On-line Estimation method based on CHVT models |
| CN109884528A (en) * | 2019-02-25 | 2019-06-14 | 江苏大学 | A kind of lithium ion battery open-circuit voltage evaluation method and system with temperature input |
| CN110045286A (en) * | 2019-03-21 | 2019-07-23 | 江苏大学 | A kind of battery open circuit voltage evaluation method and device based on air-or liquid-operated mechanical model |
| CN110609232A (en) * | 2019-03-29 | 2019-12-24 | 深圳猛犸电动科技有限公司 | Lithium ion battery SOC estimation method and device and terminal equipment |
| CN110333450A (en) * | 2019-04-30 | 2019-10-15 | 蜂巢能源科技有限公司 | Battery open circuit voltage predictor method and system |
| CN110850315A (en) * | 2019-11-29 | 2020-02-28 | 北京邮电大学 | Method and device for estimating state of charge of battery |
| CN111537887A (en) * | 2020-04-27 | 2020-08-14 | 南京航空航天大学 | Hybrid power system battery open-circuit voltage model optimization method considering hysteresis characteristic |
| CN111458646A (en) * | 2020-05-25 | 2020-07-28 | 安徽理工大学 | Lithium battery SOC estimation method based on PSO-RBF neural network |
Non-Patent Citations (1)
| Title |
|---|
| 邓晔;胡越黎;滕华强;: "锂电池开路电压的预估及SOC估算", 仪表技术, no. 02, 15 February 2015 (2015-02-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112798962B (en) | 2024-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lipu et al. | A review of state of health and remaining useful life estimation methods for lithium-ion battery in electric vehicles: Challenges and recommendations | |
| Qian et al. | Convolutional neural network based capacity estimation using random segments of the charging curves for lithium-ion batteries | |
| Hannan et al. | Neural network approach for estimating state of charge of lithium-ion battery using backtracking search algorithm | |
| Yang et al. | A novel method on estimating the degradation and state of charge of lithium-ion batteries used for electrical vehicles | |
| CN107271905B (en) | Battery capacity active estimation method for pure electric vehicle | |
| CN106383316B (en) | A kind of echelon utilizes lithium battery method of evaluating performance | |
| CN109991554B (en) | Battery electric quantity detection method and device and terminal equipment | |
| Mamo et al. | Long short-term memory with attention mechanism for state of charge estimation of lithium-ion batteries | |
| CN109100655B (en) | Data processing method and device for power battery | |
| CN110850298B (en) | Lithium battery SOH estimation method and system based on data driving | |
| CN109991545B (en) | Battery pack electric quantity detection method and device and terminal equipment | |
| CN110290972A (en) | Estimate the method for the charged state of battery cell | |
| Zhang et al. | A novel battery state-of-health estimation method for hybrid electric vehicles | |
| Liu et al. | Comparative study of incremental capacity curve determination methods for lithium-ion batteries considering the real-world situation | |
| US11360148B2 (en) | Method for updating capacity of battery, device for updating capacity of battery, electronic device, and storage unit | |
| Vennam et al. | A dynamic soh-coupled lithium-ion cell model for state and parameter estimation | |
| Zhao et al. | The li-ion battery state of charge prediction of electric vehicle using deep neural network | |
| CN112731160A (en) | Battery hysteresis model training method, and method and device for estimating battery SOC | |
| CN113075562A (en) | Battery differential pressure updating method, electric quantity estimating method, electronic device and storage medium | |
| Vennam et al. | A novel coupled electro-thermal-aging model for simultaneous SOC, SOH, and parameter estimation of lithium-ion batteries | |
| CN108845267B (en) | Data processing method and device for power battery | |
| Devi et al. | Lithium-ion battery management system: a review | |
| WO2021226797A1 (en) | Battery capacity estimation method, electronic device and storage medium | |
| CN112798962B (en) | Battery hysteresis model training method, method and device for estimating battery SOC | |
| CN116047339A (en) | Lithium ion battery pack SOC estimation method and device based on thermoelectric coupling model |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |