CN110598300A - Battery SOH prediction method and device - Google Patents
Battery SOH prediction method and device Download PDFInfo
- Publication number
- CN110598300A CN110598300A CN201910838561.1A CN201910838561A CN110598300A CN 110598300 A CN110598300 A CN 110598300A CN 201910838561 A CN201910838561 A CN 201910838561A CN 110598300 A CN110598300 A CN 110598300A
- Authority
- CN
- China
- Prior art keywords
- battery
- soh
- parameters
- battery performance
- target
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Abstract
The application discloses a battery SOH prediction method and device, which can effectively improve the prediction accuracy of target battery SOH. The method comprises the following steps: firstly, a target battery to be predicted is obtained, battery performance parameters of the target battery in a preset time period are obtained, then parameter characteristics representing performance information of the target battery are extracted from the obtained battery performance parameters, and the parameter characteristics extracted from the battery performance parameters can be used as input data and input into a pre-constructed battery SOH prediction model to accurately predict the SOH of the target battery.
Description
Technical Field
The present application relates to the field of battery technologies, and in particular, to a method and an apparatus for predicting a battery SOH.
Background
In recent years, with the increasing serious energy crisis, new energy automobiles have become the development focus of the automobile industry in the future due to the excellent energy-saving and environment-friendly characteristics of the new energy automobiles. The power battery is used as a power source Of the new energy automobile, and the capacity Of the power battery gradually attenuates in the long-term operation process, and the capacity attenuation degree Of the power battery is generally represented by a Health State (SOH) value Of the battery. At present, research on the SOH of a power battery is an important subject in the technical field of power batteries, and the accurate prediction of the SOH of the battery is very helpful for the practical application of the power battery.
However, when predicting the SOH of the battery at present, usually, only the fitting formula of the attenuation is obtained through an experimental method, and then the SOH of the battery is calculated according to the fitting formula, this prediction method does not consider the influence that various different use conditions of the vehicle may cause on the vehicle-mounted battery in the actual operation, for example, the change of parameters such as temperature and mileage during the driving process all can affect the prediction of the SOH of the battery, at this time, if the fitting formula obtained through the experimental method is still used to predict the SOH of the battery, an erroneous prediction result is easily generated, the prediction accuracy rate is reduced, and therefore, a method for quickly and accurately predicting the SOH of the battery is lacked in the prior art.
Disclosure of Invention
In view of this, the present disclosure provides a method and an apparatus for predicting a battery SOH, so as to solve the technical problem in the prior art that the battery SOH cannot be predicted quickly and accurately.
In order to solve the above problem, the technical solution provided by the embodiment of the present application is as follows:
in a first aspect, an embodiment of the present application provides a method for predicting an SOH of a battery, including:
acquiring a target battery to be predicted;
acquiring battery performance parameters of the target battery within a preset time period;
extracting parameter characteristics representing the performance information of the target battery from the battery performance parameters;
and inputting the parameter characteristics extracted from the battery performance parameters into a battery SOH prediction model constructed in advance to predict the SOH of the target battery.
Optionally, the battery performance parameters include a temperature rise and an average temperature of the target battery in the preset time period, a historical driving mileage and a historical usage time of the target battery before the preset time period, and an initial SOC and a final SOC of the target battery at a start time point and an end time point of the preset time period.
Optionally, constructing the battery SOH prediction model includes:
acquiring a battery performance training parameter;
extracting parameter characteristics of the battery performance training parameters;
and training an initial battery SOH prediction model according to the parameter characteristics of the battery performance training parameters and the SOH labels corresponding to the battery performance training parameters to generate the battery SOH prediction model.
Optionally, the initial battery SOH prediction model includes a long short term memory network LSTM layer, a Pooling layer, and a full connectivity layer DNN.
Optionally, the method further includes:
acquiring a battery performance verification parameter;
extracting parameter characteristics of the battery performance verification parameters;
inputting the parameter characteristics of the battery performance verification parameters into the battery SOH prediction model to obtain an SOH prediction result of the battery performance verification parameters;
and when the SOH prediction result of the battery performance verification parameter is inconsistent with the SOH marking result corresponding to the battery performance verification parameter, the battery performance verification parameter is used as the battery performance training parameter again, and the battery SOH prediction model is updated.
In a second aspect, the present application provides a battery SOH prediction apparatus comprising:
a first acquisition unit configured to acquire a target battery to be predicted;
the second acquisition unit is used for acquiring the battery performance parameters of the target battery within a preset time period;
the first extraction unit is used for extracting parameter characteristics representing the performance information of the target battery from the battery performance parameters;
and the first prediction unit is used for inputting the parameter characteristics extracted from the battery performance parameters into a battery SOH prediction model constructed in advance so as to predict the SOH of the target battery.
Optionally, the battery performance parameters include a temperature rise and an average temperature of the target battery in the preset time period, a historical driving mileage and a historical usage time of the target battery before the preset time period, and an initial SOC and a final SOC of the target battery at a start time point and an end time point of the preset time period.
Optionally, the apparatus further comprises:
the third acquisition unit is used for acquiring battery performance training parameters;
the second extraction unit is used for extracting the parameter characteristics of the battery performance training parameters;
and the generating unit is used for training an initial battery SOH prediction model according to the parameter characteristics of the battery performance training parameters and the SOH labels corresponding to the battery performance training parameters to generate the battery SOH prediction model.
Optionally, the initial battery SOH prediction model includes a long short term memory network LSTM layer, a Pooling layer, and a full connectivity layer DNN.
Optionally, the apparatus further comprises:
the fourth acquisition unit is used for acquiring the battery performance verification parameters;
a third extraction unit, configured to extract a parameter feature of the battery performance verification parameter;
the second prediction unit is used for inputting the parameter characteristics of the battery performance verification parameters into the battery SOH prediction model to obtain the SOH prediction result of the battery performance verification parameters;
and the updating unit is used for updating the battery SOH prediction model by taking the battery performance verification parameters as the battery performance training parameters again when the SOH prediction result of the battery performance verification parameters is inconsistent with the SOH marking result corresponding to the battery performance verification parameters.
According to the method and the device for predicting the SOH of the battery, the parameter characteristics representing the performance information of the target battery are extracted firstly, then the parameter characteristics are used as input data and input into a battery SOH prediction model which is constructed in advance, and further the SOH of the target battery can be predicted quickly and accurately according to the model output result. Compared with the mode that the vehicle is predicted by using a fitting formula obtained through experiments at present, the method and the device fully consider the influence of various different service conditions of the vehicle on the vehicle-mounted battery in actual operation, and therefore the prediction accuracy of the target battery SOH can be effectively improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flowchart of a method for predicting SOH of a battery according to an embodiment of the present disclosure;
FIG. 2 is a schematic flow chart illustrating a process for constructing a battery SOH prediction model according to an embodiment of the present disclosure;
FIG. 3 is a schematic structural diagram of a battery SOH prediction model provided by an embodiment of the present application;
fig. 4 is a schematic flowchart of a battery SOH prediction model verification method according to an embodiment of the present disclosure;
fig. 5 is a schematic composition diagram of a device for predicting SOH of a battery according to an embodiment of the present disclosure.
Detailed Description
In some methods for predicting the SOH of a battery, a fitting formula of the attenuation is obtained through an experimental method, and then the SOH of the battery is calculated according to the fitting formula. However, the prediction method does not consider the influence of various different use conditions of the vehicle on the vehicle-mounted battery in actual operation, so that an obtained experimental conclusion (namely an obtained fitting formula) may be wrong, a wrong battery SOH is calculated, and the prediction accuracy is low.
In order to solve the above-mentioned drawbacks, the present application provides a method for predicting battery SOH, which includes obtaining battery performance parameters of a target battery within a preset time period after obtaining the target battery to be predicted, then extracting parameter features representing performance information of the target battery from the obtained battery performance parameters, and further inputting the parameter features extracted from the battery performance parameters as input data into a pre-constructed battery SOH prediction model to accurately predict SOH of the target battery, and thus, in the present application, only one battery SOH prediction model is required, and based on various battery performance parameters of the target battery within the preset time period, SOH of the target battery can be accurately predicted, and prediction basis is more comprehensive, and influences that may be caused by various different use conditions of a vehicle in actual operation on a vehicle-mounted target battery are fully considered, therefore, the prediction accuracy of the target battery SOH can be effectively improved.
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, 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 obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
First embodiment
Referring to fig. 1, a schematic flow chart of a method for predicting the SOH of a battery provided in this embodiment is shown, where the method includes the following steps:
s101: and acquiring a target battery to be predicted.
In the present embodiment, any battery that performs SOH prediction using the present embodiment is defined as a target battery. Also, the present embodiment does not limit the battery type of the target battery, for example, the target battery may be a lithium ion battery, a nickel hydrogen battery, or the like.
S102: and acquiring the battery performance parameters of the target battery in a preset time period.
In this embodiment, after the target battery to be predicted is obtained in step S101, battery performance parameters of the target battery in a preset time period may be further obtained, where the preset time period refers to a preset time duration of a time slice, and a specific value may be set according to an actual situation. In the time slice duration range, the battery performance parameters of the target battery can be obtained by using the existing or future parameter obtaining method, so as to predict the SOH of the target battery through the subsequent steps S103-S104.
The battery performance parameters refer to parameters representing various performances of the target battery, and an optional implementation manner is that the battery performance parameters may include a temperature rise and an average temperature of the target battery in a preset time period, a historical driving mileage and a historical usage time of the target battery before the preset time period, and an initial State of Charge (SOC) of the target battery at a starting time point and a cut-off SOC at an ending time point of the preset time period.
The temperature rise of the target battery in the preset time period refers to a temperature rise condition that the temperature of the target battery rises from a temperature value at a starting time point of the preset time period to a temperature value at an ending time point of the preset time period.
It should be noted that the battery performance parameters are not limited to the above, and those skilled in the art may determine other battery performance parameters of the target battery according to practical situations, such as the total charge and discharge amount of the target battery before the preset time period, and the like, which is not limited in the embodiment of the present application.
S103: and extracting parameter characteristics representing the performance information of the target battery from the battery performance parameters.
In this embodiment, after the battery performance parameters of the target battery within the preset time period are obtained in step S102, the battery performance parameters may be processed by using an existing or future feature extraction method, for example, a Principal Component Analysis (PCA) or a feature extraction method such as Word2vec, and the like, to extract features capable of representing performance information of the target battery from the battery performance parameters, where the features are defined as parameter features, and the parameter features should carry all performance information of the corresponding target battery. Further, step S104 may be executed using the extracted parameter characteristics of the target battery.
S104: and inputting the parameter characteristics extracted from the battery performance parameters into a battery SOH prediction model constructed in advance to predict the SOH of the target battery.
In this embodiment, after the parameter features representing the performance information of the target battery are extracted in step S103, the parameter features may be input into a battery SOH prediction model constructed in advance, so as to predict the SOH of the target battery according to the output value of the model.
Specifically, after the parameter characteristics of the target battery in the preset time period are extracted in step S103, the parameter characteristics may be input into a battery SOH prediction model constructed in advance, and the available battery capacity consumed by the target battery in the preset time period may be output through a dead battery network (DNN) of the model and defined as Δ C, and the unit of Δ C may be ampere-hour, symbol: and A, H, dividing the difference value between the delta C and the initial SOC of the target battery at the starting time point and the ending SOC of the target battery at the ending time point of the preset time period to obtain the total available battery capacity of the target battery at the ending time point of the preset time period, and further dividing the total available battery capacity by the rated capacity of the target battery to predict the SOH of the target battery, wherein the specific calculation formula is as follows:
wherein Δ C represents an available battery capacity consumed by the target battery within a preset time period; SOCInitialRepresenting an initial SOC of the target battery at a starting time point of the preset time period; SOCCut-offAn off-SOC indicating an end time point of the preset time period of the target battery; cGeneral assemblyRepresenting the total available battery capacity of the target battery at the end time point of the preset time period; cRated valueIndicating a rated capacity of the target battery; SO (SO)HBattery with a battery cellIndicating the predicted SOH of the target battery.
For example, the following steps are carried out: assuming that the available battery capacity Δ C of the target battery output by the battery SOH prediction model consumed during the preset time period is 5AH, and the initial SOC and the end SOC of the target battery at the start time point and the end time point of the preset time period are 90% and 80%, respectively, the total available battery capacity C of the target battery at the end time point of the preset time period can be calculated by the above equation (1)General assemblyIs 50AH, i.e., 5/(90% -80%) -50, and further, the rated capacity C of the target battery is assumedRated valueAt 60AH, the SOH of the target battery is predicted to be 83.3%, that is, 50/60 is 83.3% by the above equation (2).
It should be noted that, in order to implement step S103, a battery SOH prediction model needs to be constructed in advance, and the specific construction process can be referred to in the related description of the second embodiment.
In summary, according to the battery SOH prediction method provided in this embodiment, after the target battery to be predicted is obtained, the battery performance parameters of the target battery in the preset time period may be obtained first, then, the parameter features representing the performance information of the target battery are extracted from the obtained battery performance parameters, and then, the parameter features extracted from the battery performance parameters may be used as input data and input to a pre-constructed battery SOH prediction model, so as to accurately predict the SOH of the target battery. Therefore, in the embodiment, only one battery SOH prediction model is needed, the SOH of the target battery can be accurately predicted based on various battery performance parameters of the target battery in a preset time period, the prediction basis is more comprehensive, the influence of various different service conditions of the vehicle on the vehicle-mounted target battery in actual operation is fully considered, and the prediction accuracy of the SOH of the target battery can be effectively improved.
Second embodiment
The present embodiment will describe a specific construction process of the battery SOH prediction model mentioned in the first embodiment. By utilizing the pre-constructed battery SOH prediction model, the SOH of the target battery can be accurately and quickly predicted.
Referring to fig. 2, it shows a schematic flow chart of constructing a battery SOH prediction model provided in this embodiment, and the flow chart includes the following steps:
s201: and acquiring battery performance training parameters.
In this embodiment, in order to construct a battery SOH prediction model, a large amount of preparation work needs to be performed in advance, first, it is necessary to collect and acquire battery performance training parameters, for example, performance parameters of the battery in 100 preset time periods (the performance parameters may include temperature rise and average temperature of the battery in each preset time period, historical driving mileage and historical usage time of the battery before the preset time period, and initial SOC and ending SOC of the battery at a starting time point and an ending time point of the preset time period) may be collected in advance, and the collected performance parameters of the battery in each time period are respectively used as sample parameter data, and real SOH of the battery represented by these sample data is marked manually in advance to train the battery SOH prediction model.
S202: and extracting the parameter characteristics of the battery performance training parameters.
In this embodiment, after the battery performance training parameters are obtained in step S201, the training parameters cannot be directly used for training and generating the battery SOH prediction model, but the parameter characteristics of the battery performance training parameters need to be extracted, wherein, the extraction of the parameter characteristics of the battery performance training parameters refers to converting each parameter in the battery performance training parameters into a group of time sequence characteristic vectors with obvious physical significance, such as, a time sequence feature vector corresponding to the average temperature of the battery in each preset time period and a time sequence vector feature corresponding to the historical mileage of the battery before each preset time period can be generated, in the process of feature extraction, the method of extraction such as PCA or Word2vec can be used for feature extraction, namely the conversion of the chronological feature vector, and then training by using the extracted parameter characteristics of the battery performance training parameters to obtain a battery SOH prediction model.
S203: and training the initial battery SOH prediction model according to the parameter characteristics of the battery performance training parameters and the SOH labels corresponding to the battery performance training parameters to generate the battery SOH prediction model.
In this embodiment, after the parameter features of the battery performance training parameters are extracted in step S202, further, the initial battery SOH prediction model may be trained according to the parameter features of the battery performance training parameters and the real SOH labeling results corresponding to the battery performance training parameters, so as to generate the battery SOH prediction model.
In an alternative implementation manner, the initial battery SOH prediction model may include a Long Short-Term Memory network (LSTM) layer, a Pooling layer, and a full connection layer DNN.
In this implementation, in order to train and generate a battery SOH prediction model for accurately predicting the SOH of the battery, a model architecture based on a multi-layer network including an LSTM layer, a pooled poolling layer, and a full-connected layer DNN may be obtained, as shown in fig. 3, for constructing an initial battery SOH prediction model.
Then, a set of sample parameter data may be sequentially extracted from the model training data, and multiple rounds of model training may be performed until the training end condition is satisfied, at which time, the battery SOH prediction model is generated.
Specifically, during the current round of training, the battery performance parameters of the target battery in the first embodiment in the preset time period may be replaced by the sample parameter data extracted in the current round, and the SOH of the battery represented by the sample parameter data may be predicted through the current initial battery SOH prediction model according to the execution process in the first embodiment. Specifically, according to the steps S101 to S104 in the first embodiment, after feature extraction is performed on sample parameter data, the available battery capacity Δ C consumed by the battery in the preset time period corresponding to the sample parameter data is output through the full connection layer, then, the SOH of the battery is calculated through the formulas (1) and (2), the predicted SOH of the battery is compared with the corresponding artificially labeled real SOH labeling result, the model parameter is updated according to the difference between the predicted SOH and the artificially labeled real SOH labeling result, until a preset condition is met, for example, the difference change range is small, the update of the model parameter is stopped, the training of the battery SOH prediction model is completed, and a trained battery SOH prediction model is generated.
Through the embodiment, the battery SOH prediction model can be generated by training the battery performance training parameters, and further, the generated battery SOH prediction model can be verified by using the battery performance verification parameters.
The battery SOH prediction model verification method provided by the embodiment of the present application is described below with reference to the accompanying drawings.
Referring to fig. 4, which shows a flowchart of a battery SOH prediction model verification method provided in an embodiment of the present application, as shown in fig. 4, the method includes:
s401: and acquiring battery performance verification parameters.
In this embodiment, in order to implement the verification of the battery SOH prediction model, first, battery performance verification parameters need to be obtained, where the battery performance verification parameters refer to various performance parameters of the battery that can be used for performing the battery SOH prediction model verification, and after the battery performance verification parameters are obtained, step S402 may be continuously executed.
S402: and extracting the parameter characteristics of the battery performance verification parameters.
In this embodiment, after the battery performance verification parameters are obtained in step S401, the battery performance verification parameters cannot be directly used for verifying the battery SOH prediction model, but the parameter characteristics of the battery performance verification parameters need to be extracted, wherein, the extraction of the parameter feature of the battery performance verification parameter refers to converting each parameter in the battery performance verification parameter into a group of time sequence feature vectors with obvious physical significance, such as, a time sequence feature vector corresponding to the average temperature of the battery in each preset time period and a time sequence vector feature corresponding to the historical mileage of the battery before each preset time period can be generated, in the process of feature extraction, the method of extraction such as PCA or Word2vec can be used for feature extraction, namely the conversion of the chronological feature vector, and further, the obtained battery SOH prediction model can be verified by using the extracted parameter characteristics of the battery performance verification parameters.
S403: and inputting the parameter characteristics of the battery performance verification parameters into a battery SOH prediction model to obtain an SOH prediction result of the battery performance verification parameters.
In this embodiment, after the parameter features of the battery performance verification parameters are extracted in step S402, the parameter features of the battery performance verification parameters may be further input into a battery SOH prediction model, and a prediction result of the battery SOH corresponding to the battery performance verification parameters is obtained by calculation according to a model output result, so that step S404 may be continuously performed.
S404: and when the SOH prediction result of the battery performance verification parameter is inconsistent with the SOH marking result corresponding to the battery performance verification parameter, the battery performance verification parameter is used as the battery performance training parameter again, and the battery SOH prediction model is updated.
In this embodiment, after the battery SOH is predicted in step S403, when the prediction result is inconsistent with the manual labeling result corresponding to the battery performance verification parameter, the battery performance verification parameter may be used as the battery performance training parameter again to update the battery SOH prediction model.
Through the embodiment, the battery SOH prediction model can be effectively verified by using the battery performance verification parameters, and when the battery SOH prediction result corresponding to the battery performance verification parameters is inconsistent with the manual marking result corresponding to the battery performance verification parameters, the battery SOH prediction model can be timely adjusted and updated, so that the prediction precision and accuracy of the prediction model can be improved.
In summary, the battery SOH prediction model trained by the embodiment can predict the SOH of the target battery by using the parameter characteristics representing the performance information of the target battery, and fully considers the influence of various different use conditions of the vehicle on the vehicle-mounted battery in the actual operation in the prediction process, so that the prediction accuracy of the SOH of the target battery can be effectively improved.
Third embodiment
In this embodiment, a device for predicting SOH of a battery will be described, and please refer to the above method embodiments for related contents.
Referring to fig. 5, a schematic composition diagram of a battery SOH prediction apparatus provided in this embodiment is shown, where the apparatus includes:
a first obtaining unit 501, configured to obtain a target battery to be predicted;
a second obtaining unit 502, configured to obtain a battery performance parameter of the target battery within a preset time period;
a first extracting unit 503, configured to extract a parameter feature representing performance information of the target battery from the battery performance parameters;
a first prediction unit 504, configured to input the parameter features extracted from the battery performance parameters into a battery SOH prediction model constructed in advance, so as to predict the SOH of the target battery.
In one implementation manner of this embodiment, the battery performance parameters include a temperature rise and an average temperature of the target battery in the preset time period, a historical driving mileage and a historical usage time of the target battery before the preset time period, and an initial SOC and a cutoff SOC of the target battery at a start time point and an end time point of the preset time period.
In an implementation manner of this embodiment, the apparatus further includes:
the third acquisition unit is used for acquiring battery performance training parameters;
the second extraction unit is used for extracting the parameter characteristics of the battery performance training parameters;
and the generating unit is used for training an initial battery SOH prediction model according to the parameter characteristics of the battery performance training parameters and the SOH labels corresponding to the battery performance training parameters to generate the battery SOH prediction model.
In one implementation of this embodiment, the initial battery SOH prediction model includes a long short term memory network LSTM layer, a pooled Pooling layer, and a full connectivity layer DNN.
In an implementation manner of this embodiment, the apparatus further includes:
the fourth acquisition unit is used for acquiring the battery performance verification parameters;
a third extraction unit, configured to extract a parameter feature of the battery performance verification parameter;
the second prediction unit is used for inputting the parameter characteristics of the battery performance verification parameters into the battery SOH prediction model to obtain the SOH prediction result of the battery performance verification parameters;
and the updating unit is used for updating the battery SOH prediction model by taking the battery performance verification parameters as the battery performance training parameters again when the SOH prediction result of the battery performance verification parameters is inconsistent with the SOH marking result corresponding to the battery performance verification parameters.
In summary, according to the battery SOH prediction apparatus provided in this embodiment, after the target battery to be predicted is obtained, the battery performance parameters of the target battery in the preset time period may be obtained first, then, the parameter features representing the performance information of the target battery are extracted from the obtained battery performance parameters, and then, the parameter features extracted from the battery performance parameters may be used as input data and input into the battery SOH prediction model constructed in advance, so as to accurately predict the SOH of the target battery. Therefore, in the embodiment, only one battery SOH prediction model is needed, the SOH of the target battery can be accurately predicted based on various battery performance parameters of the target battery in a preset time period, the prediction basis is more comprehensive, the influence of various different service conditions of the vehicle on the vehicle-mounted target battery in actual operation is fully considered, and the prediction accuracy of the SOH of the target battery can be effectively improved.
As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that all or part of the steps in the above embodiment methods can be implemented by software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present application may be essentially or partially implemented in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network communication device such as a media gateway, etc.) to execute the method according to the embodiments or some parts of the embodiments of the present application.
It should be noted that, in the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for predicting SOH of a battery, comprising:
acquiring a target battery to be predicted;
acquiring battery performance parameters of the target battery within a preset time period;
extracting parameter characteristics representing the performance information of the target battery from the battery performance parameters;
and inputting the parameter characteristics extracted from the battery performance parameters into a battery SOH prediction model constructed in advance to predict the SOH of the target battery.
2. The method according to claim 1, wherein the battery performance parameters include a temperature rise and an average temperature of the target battery over the preset time period, a historical driving mileage and a historical usage time of the target battery before the preset time period, and an initial SOC and a cutoff SOC of the target battery at a start time point and an end time point of the preset time period.
3. The method of claim 1, wherein constructing the battery SOH prediction model comprises:
acquiring a battery performance training parameter;
extracting parameter characteristics of the battery performance training parameters;
and training an initial battery SOH prediction model according to the parameter characteristics of the battery performance training parameters and the SOH labels corresponding to the battery performance training parameters to generate the battery SOH prediction model.
4. The method of claim 3, wherein the initial battery SOH prediction model comprises a long short term memory network (LSTM) layer, a pooled Pooling layer, and a fully-connected layer (DNN).
5. The method according to any one of claims 3 to 4, further comprising:
acquiring a battery performance verification parameter;
extracting parameter characteristics of the battery performance verification parameters;
inputting the parameter characteristics of the battery performance verification parameters into the battery SOH prediction model to obtain an SOH prediction result of the battery performance verification parameters;
and when the SOH prediction result of the battery performance verification parameter is inconsistent with the SOH marking result corresponding to the battery performance verification parameter, the battery performance verification parameter is used as the battery performance training parameter again, and the battery SOH prediction model is updated.
6. A battery SOH prediction apparatus, comprising:
a first acquisition unit configured to acquire a target battery to be predicted;
the second acquisition unit is used for acquiring the battery performance parameters of the target battery within a preset time period;
the first extraction unit is used for extracting parameter characteristics representing the performance information of the target battery from the battery performance parameters;
and the first prediction unit is used for inputting the parameter characteristics extracted from the battery performance parameters into a battery SOH prediction model constructed in advance so as to predict the SOH of the target battery.
7. The apparatus of claim 6, wherein the battery performance parameters comprise a temperature rise and an average temperature of the target battery over the preset time period, a historical driving mileage and a historical usage time of the target battery before the preset time period, and an initial SOC and a cutoff SOC of the target battery at a starting time point and an ending time point of the preset time period.
8. The apparatus of claim 6, further comprising:
the third acquisition unit is used for acquiring battery performance training parameters;
the second extraction unit is used for extracting the parameter characteristics of the battery performance training parameters;
and the generating unit is used for training an initial battery SOH prediction model according to the parameter characteristics of the battery performance training parameters and the SOH labels corresponding to the battery performance training parameters to generate the battery SOH prediction model.
9. The apparatus of claim 8, wherein the initial battery SOH prediction model comprises a long short term memory network LSTM layer, a pooled Pooling layer, and a fully-connected layer DNN.
10. The apparatus of any one of claims 8 to 9, further comprising:
the fourth acquisition unit is used for acquiring the battery performance verification parameters;
a third extraction unit, configured to extract a parameter feature of the battery performance verification parameter;
the second prediction unit is used for inputting the parameter characteristics of the battery performance verification parameters into the battery SOH prediction model to obtain the SOH prediction result of the battery performance verification parameters;
and the updating unit is used for updating the battery SOH prediction model by taking the battery performance verification parameters as the battery performance training parameters again when the SOH prediction result of the battery performance verification parameters is inconsistent with the SOH marking result corresponding to the battery performance verification parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910838561.1A CN110598300A (en) | 2019-09-05 | 2019-09-05 | Battery SOH prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910838561.1A CN110598300A (en) | 2019-09-05 | 2019-09-05 | Battery SOH prediction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110598300A true CN110598300A (en) | 2019-12-20 |
Family
ID=68857752
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910838561.1A Pending CN110598300A (en) | 2019-09-05 | 2019-09-05 | Battery SOH prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110598300A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111308375A (en) * | 2020-02-04 | 2020-06-19 | 浙江大学 | LSTM-FFNN-based electric forklift lithium ion battery health state prediction method |
| CN111722139A (en) * | 2020-06-29 | 2020-09-29 | 重庆邮电大学 | Lithium battery health monitoring model self-learning method based on micro-continuous mapping |
| CN111983457A (en) * | 2020-07-15 | 2020-11-24 | 江苏大学 | Battery pack SOH estimation method based on LSTM neural network |
| CN112542621A (en) * | 2020-11-25 | 2021-03-23 | 东软睿驰汽车技术（沈阳）有限公司 | Battery temperature vector generation method and related equipment thereof |
| CN113655397A (en) * | 2020-05-12 | 2021-11-16 | 北京京东乾石科技有限公司 | Method, device and storage medium for determining full charge and discharge frequency of battery |
| CN113743661A (en) * | 2021-08-30 | 2021-12-03 | 西安交通大学 | Method, system, equipment and storage medium for predicting online capacity of lithium ion battery |
| CN114167301A (en) * | 2021-11-30 | 2022-03-11 | 同济大学 | Power battery evaluation method based on real vehicle data of electric vehicle |
| CN114791993A (en) * | 2022-05-16 | 2022-07-26 | 江苏大学 | SOH prediction method and system for power battery pack |
| CN116400227A (en) * | 2023-06-08 | 2023-07-07 | 长安大学 | SOH prediction method, system, equipment and medium for power battery of electric automobile |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109001631A (en) * | 2018-04-27 | 2018-12-14 | 张家港莫特普数据科技有限公司 | Battery performance online evaluation system and method |
| US20190019080A1 (en) * | 2015-12-31 | 2019-01-17 | Vito Nv | Methods, controllers and systems for the control of distribution systems using a neural network architecture |
| CN109799463A (en) * | 2019-01-18 | 2019-05-24 | 上海卡鲁自动化科技有限公司 | The estimation and prediction technique of power battery SOC/SOH/SOP under actual operating mode based on big data method |
| CN109934294A (en) * | 2019-03-18 | 2019-06-25 | 常伟 | A method of batteries of electric automobile SOH prediction is carried out based on big data machine learning |
| CN109991542A (en) * | 2019-03-27 | 2019-07-09 | 东北大学 | Lithium ion battery residual life prediction technique based on WDE optimization LSTM network |
-
2019
- 2019-09-05 CN CN201910838561.1A patent/CN110598300A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190019080A1 (en) * | 2015-12-31 | 2019-01-17 | Vito Nv | Methods, controllers and systems for the control of distribution systems using a neural network architecture |
| CN109001631A (en) * | 2018-04-27 | 2018-12-14 | 张家港莫特普数据科技有限公司 | Battery performance online evaluation system and method |
| CN109799463A (en) * | 2019-01-18 | 2019-05-24 | 上海卡鲁自动化科技有限公司 | The estimation and prediction technique of power battery SOC/SOH/SOP under actual operating mode based on big data method |
| CN109934294A (en) * | 2019-03-18 | 2019-06-25 | 常伟 | A method of batteries of electric automobile SOH prediction is carried out based on big data machine learning |
| CN109991542A (en) * | 2019-03-27 | 2019-07-09 | 东北大学 | Lithium ion battery residual life prediction technique based on WDE optimization LSTM network |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111308375A (en) * | 2020-02-04 | 2020-06-19 | 浙江大学 | LSTM-FFNN-based electric forklift lithium ion battery health state prediction method |
| CN111308375B (en) * | 2020-02-04 | 2021-12-07 | 浙江大学 | LSTM-FFNN-based electric forklift lithium ion battery health state prediction method |
| CN113655397A (en) * | 2020-05-12 | 2021-11-16 | 北京京东乾石科技有限公司 | Method, device and storage medium for determining full charge and discharge frequency of battery |
| CN111722139A (en) * | 2020-06-29 | 2020-09-29 | 重庆邮电大学 | Lithium battery health monitoring model self-learning method based on micro-continuous mapping |
| CN111722139B (en) * | 2020-06-29 | 2022-08-09 | 重庆邮电大学 | Lithium battery health monitoring model self-learning method based on micro-continuous mapping |
| CN111983457A (en) * | 2020-07-15 | 2020-11-24 | 江苏大学 | Battery pack SOH estimation method based on LSTM neural network |
| CN112542621A (en) * | 2020-11-25 | 2021-03-23 | 东软睿驰汽车技术（沈阳）有限公司 | Battery temperature vector generation method and related equipment thereof |
| CN113743661A (en) * | 2021-08-30 | 2021-12-03 | 西安交通大学 | Method, system, equipment and storage medium for predicting online capacity of lithium ion battery |
| CN114167301A (en) * | 2021-11-30 | 2022-03-11 | 同济大学 | Power battery evaluation method based on real vehicle data of electric vehicle |
| CN114791993A (en) * | 2022-05-16 | 2022-07-26 | 江苏大学 | SOH prediction method and system for power battery pack |
| CN116400227A (en) * | 2023-06-08 | 2023-07-07 | 长安大学 | SOH prediction method, system, equipment and medium for power battery of electric automobile |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110598300A (en) | Battery SOH prediction method and device | |
| Tang et al. | A novel framework for Lithium-ion battery modeling considering uncertainties of temperature and aging | |
| CN111090047B (en) | Lithium battery health state estimation method based on multi-model fusion | |
| CN110658460B (en) | Battery life prediction method and device for battery pack | |
| CN107957562B (en) | Online prediction method for residual life of lithium ion battery | |
| Zhang et al. | Battery modelling methods for electric vehicles-A review | |
| CN110873841B (en) | Battery life prediction method based on combination of data driving and battery characteristics | |
| CN110806541B (en) | AD-BAS-based lithium battery model parameter identification method | |
| CN112834945A (en) | Evaluation model establishing method, battery health state evaluation method and related product | |
| CN103197251A (en) | Identification method of second order resistance and capacitance (RC) equivalent model of power lithium battery | |
| CN102937704A (en) | Method for identifying RC (resistor-capacitor) equivalent model of power battery | |
| CN110568373A (en) | Lithium battery health state evaluation method, system, terminal and storage medium | |
| CN105891715A (en) | Lithium ion battery health state estimation method | |
| CN110806540B (en) | Battery cell test data processing method, device and system and storage medium | |
| CN114114051B (en) | Battery aging value determining method and device and electronic equipment | |
| CN113125967A (en) | Lithium battery SOE calculation method based on temperature rise prediction | |
| CN112765726A (en) | Service life prediction method and device | |
| CN113255199A (en) | Battery remaining life prediction method based on particle filtering | |
| CN112731160A (en) | Battery hysteresis model training method, and method and device for estimating battery SOC | |
| CN111319510B (en) | Method and device for predicting driving range of electric vehicle | |
| CN116679213A (en) | SOH estimation method for electric vehicle power battery based on integrated deep learning | |
| CN116609669A (en) | Method and device for extracting charging condition of power battery based on big data | |
| Xia et al. | Short-Term Capacity Estimation and Long-Term Remaining Useful Life Prediction of Lithium-Ion Batteries Based on a Data-Driven Method | |
| CN115327389A (en) | Lithium battery SOC estimation method based on genetic algorithm improved double-Kalman filtering | |
| CN113255215B (en) | Lithium battery health state estimation method based on voltage segments |
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 |