CN110441706A - A kind of battery SOH estimation method and equipment - Google Patents
A kind of battery SOH estimation method and equipment Download PDFInfo
- Publication number
- CN110441706A CN110441706A CN201910784277.0A CN201910784277A CN110441706A CN 110441706 A CN110441706 A CN 110441706A CN 201910784277 A CN201910784277 A CN 201910784277A CN 110441706 A CN110441706 A CN 110441706A
- Authority
- CN
- China
- Prior art keywords
- battery
- soh
- tested
- online data
- curve
- 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
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
Abstract
The present invention relates to a kind of battery SOH estimation method and equipment, are applied to battery technology field, wherein method includes the online data for obtaining tested battery；According to online data, the mode input variable of tested battery is calculated, mode input variable is the characteristic quantity extracted according to the IC curve for being tested battery；By the mode input variable input of tested battery battery SOH discrimination model trained in advance, determine that the SOH classification that the SOH of tested battery belongs to, SOH classification are a variety of altogether.The SOH of battery can be directly obtained according to online data, eliminate the time tested in the lab, shorten the test period, and pass through the calculating to online data, mode input variable is finally obtained, is inputted in battery SOH discrimination model, the classification of SOH belonging to the SOH of tested battery can be directly determined, the real time discriminating of battery SOH is realized, it is easy to operate.
Description
Technical field
The present invention relates to battery technology fields, and in particular to a kind of battery SOH estimation method and equipment.
Background technique
With the popularization of new-energy automobile market and the raising of user cognition degree, the sales volume of electric car just increases year by year,
Its power resources is mainly battery, and SOH (State Of Health) has important shadow to using and developing for electric car
It rings, therefore, the SOH assessment of electric automobile power battery is then particularly important.
In the related technology, the test environment for being mainly based upon laboratory assesses battery SOH, generallys use capacity and declines
Subtraction, partial discharge method etc., based on these methods are mainly tested by battery cycle charge-discharge, to the variation feelings of battery behavior
Condition is analyzed, to estimate the SOH of battery, not only the test period is long for this mode, but also cannot be carried out to battery online real
When measure so that there are errors for measurement result.
Summary of the invention
In view of this, the present invention is in order to overcome the problems, such as to provide one kind present in the relevant technologies at least to some extent
Battery SOH estimation method and equipment.
In order to solve the above technical problems, the present invention adopts the following technical scheme:
In a first aspect, a kind of battery SOH estimation method, comprising:
Obtain the online data of tested battery；
According to the online data, the mode input variable of the tested battery, the mode input variable is calculated
For the characteristic quantity extracted according to the IC curve of the tested battery；
By the mode input variable input of tested battery battery SOH discrimination model trained in advance, the quilt is determined
The SOH classification that the SOH of battery belongs to is surveyed, the SOH classification is a variety of altogether.
Optionally, the online data includes: voltage, battery real time electrical quantity, described according to the online data, is calculated
To the mode input variable of the tested battery, comprising:
Voltage-capacity curve is established according to the voltage and the battery real time electrical quantity；
It is IC curve by the voltage-capacity Curve transform；
The mode input variable is extracted on the IC curve.
Optionally, the mode input variable are as follows: the height value and corresponding positional value at each peak on the IC curve.
Optionally, after the online data for obtaining tested battery, the method also includes:
Using supporting vector machine model trained in advance, the tested battery is divided by failure according to the online data
Battery and non-dead battery, so that the SOH to non-dead battery differentiates.
Optionally, further includes:
Obtain the online data of first group of training sample；
According to the online data of first group of training sample, estimate that the battery of first group of training sample is estimated to hold
Amount；
The corresponding identification information of the battery estimated capacity of first group of training sample is obtained, the identification information is
Dead battery or non-dead battery, wherein when the battery estimated capacity of first group of training sample is greater than or equal to preset value
When, the identification information is non-dead battery, when the battery estimated capacity of first group of training sample is less than preset value, institute
Stating identification information is dead battery；
According to the battery estimated capacity and corresponding identification information of first group of training sample, calculated using support vector machines
Method is trained, and obtains the supporting vector machine model.
Optionally, the online data includes: electric current, battery real time electrical quantity, described according to first group of training sample
Online data estimate the battery estimated capacity of first group of training sample, comprising:
First group of training is calculated using current integration method according to the electric current and the battery real time electrical quantity
The battery estimated capacity of sample.
Optionally, further includes:
The characteristic quantity of second group of training sample is obtained, second group of training sample is non-dead battery, the characteristic quantity
It is to be extracted according to the IC curve of the non-dead battery；
The characteristic quantity of second group of training sample is clustered, plurality of classes is clustered into, using every kind of classification as one
Kind SOH classification；
Using the characteristic quantity as the input of the battery SOH discrimination model, the SOH classification is used as the battery SOH
The output of discrimination model constructs the battery SOH discrimination model.
It is optionally, described to convert IC curve for the voltage-capacity curve, comprising:
The ratio of electricity increment and voltage increment is calculated according to the voltage-capacity curve；
Using the electricity increment and voltage increment ratio as ordinate, the voltage value establishes IC song as abscissa
Line.
Optionally, further includes:
Using generalized linear model trained in advance, user is estimated the service condition of tested battery and showed,
The service condition includes at least one in following item: using the battery SOH after preset duration；After preset times
Battery SOH；Battery SOH decays to use duration when preset value.
Second aspect, a kind of battery SOH estimation device, comprising:
Module is obtained, for obtaining the online data of tested battery；
Computing module, it is described for the mode input variable of the tested battery to be calculated according to the online data
Mode input variable is the characteristic quantity extracted according to the IC curve of the tested battery；
SOH discrimination module, the battery SOH trained in advance for the mode input variable input by the tested battery differentiate
Model determines the SOH classification that the SOH of the tested battery belongs to, and the SOH classification is a variety of altogether.
The third aspect, a kind of battery SOH estimation equipment, comprising:
Processor, and the memory being connected with the processor；
The memory is for storing computer program；
The processor is for calling and executing the computer program in the memory, to execute such as first aspect
The battery SOH estimation method.
Fourth aspect, a kind of storage medium, the storage medium are stored with computer program, and the computer program is located
When managing device execution, the battery SOH estimation method as described in any one of first aspect present invention is realized.
The invention adopts the above technical scheme, and following technical effect may be implemented: by the tested battery of acquisition in line number
According to the mode input variable of the tested battery being calculated, wherein the mode input then according to the online data
Variable is the characteristic quantity extracted according to the IC curve of the tested battery；The mode input variable input of the tested battery is pre-
First trained battery SOH discrimination model determines the SOH classification that the SOH of the tested battery belongs to, and the SOH classification is altogether more
Kind.In this way, replacing using test data in the prior art using the online data of battery, On-line sampling system is realized, so that
The result of measurement is more accurate；Also, the SOH that battery can be directly obtained according to online data, eliminates in the lab
The time tested, shorten the test period.In addition, mode input variable is finally obtained by the calculating to online data,
It is inputted in battery SOH discrimination model, the classification of SOH belonging to the SOH of tested battery can be directly determined, realize battery
The real time discriminating of SOH, it is easy to operate.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow diagram for the battery SOH estimation method that one embodiment of the invention provides；
Fig. 2 be another embodiment of the present invention provides battery SOH estimation method flow diagram；
Fig. 3 is the schematic diagram for the battery IC curve that one embodiment of the invention provides；
Fig. 4 is the structural schematic diagram for the battery SOH estimation device that one embodiment of the invention provides；
Fig. 5 is the structural schematic diagram for the battery SOH estimation equipment that one embodiment of the invention provides.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, technical solution of the present invention will be carried out below
Detailed description.Obviously, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, those of ordinary skill in the art are obtained all without making creative work
Other embodiment belongs to the range that the present invention is protected.
In order to better understand scheme provided by the present application, it is to be understood that the following contents:
The development of intelligent car networking technology has driven the upgrading of automobile industry, and the fast development of new-energy automobile is also spread out
New battery product is born, the research of this respect is actually rare on domestic market.This method can be applied to more large sample
In (such as lithium manganate battery, lead-acid battery, ferric phosphate lithium cell), the IC curve of the battery under different performance state is portrayed.
Embodiment
Fig. 1 is the flow diagram for the battery SOH estimation method that one embodiment of the invention provides.As shown in Figure 1, this implementation
Example provides a kind of battery SOH estimation method, comprising:
Step 101, the online data for obtaining tested battery.
In some embodiments, online data may include the voltage, electric current, battery real time electrical quantity (State of of battery
Charge, SOC) etc..
Step 102, according to the online data, the mode input variable of the tested battery, the model is calculated
Input variable is the characteristic quantity extracted according to the IC curve of the tested battery；
In some embodiments, there are many modes of calculating battery model input variable, for example, online data can be first passed through
In voltage and battery real time electrical quantity draw the voltage-capacity curve of tested battery, then convert voltage-capacity curve to
IC curve extracts characteristic quantity in IC curve, and then using characteristic quantity as mode input variable.
Step 103, the battery SOH discrimination model for training the mode input variable input of the tested battery in advance, really
The SOH classification that the SOH of the fixed tested battery belongs to, the SOH classification are a variety of altogether.
In the present embodiment, institute is calculated then according to the online data in the online data by obtaining tested battery
State the mode input variable of tested battery, wherein the mode input variable is to extract according to the IC curve of the tested battery
Characteristic quantity；By the mode input variable input of tested battery battery SOH discrimination model trained in advance, the quilt is determined
The SOH classification that the SOH of battery belongs to is surveyed, the SOH classification is a variety of altogether.In this way, being replaced using the online data of battery existing
Test data is used in technology, realizes On-line sampling system, so that the result of measurement is more accurate；Also, according in line number
According to the SOH that can directly obtain battery, the time tested in the lab is eliminated, the test period is shortened.In addition,
By the calculating to online data, mode input variable is finally obtained, is inputted in battery SOH discrimination model, it can be straight
It connects and determines SOH classification belonging to the SOH of tested battery, realize the real time discriminating of battery SOH, it is easy to operate.
Fig. 2 be another embodiment of the present invention provides battery SOH estimation method flow diagram.As shown in Fig. 2, this reality
It applies example and another battery SOH estimation method is provided, comprising:
Step 201, the online data for obtaining first group of training sample；
In some embodiments, online data may include the voltage, electric current, battery real time electrical quantity (State of of battery
Charge, SOC) etc..
Wherein, training sample can obtain in vehicle net system.Vehicle net system refers to by pacifying in vehicle instrument desk
Vehicle-mounted terminal equipment is filled, realize to vehicle all working situation and quiet, multidate information acquisition, storage and is sent.The fortune of vehicle
Row often relates to multinomial switching value, sensor analog quantity, CAN signal data etc., driver in operation vehicle operation,
The vehicle data of generation constantly postbacks background data base, forms mass data, realizes " the filtering to data by cloud computing platform
Cleaning ", Data Analysis Platform carry out statement form processing to data, check for administrative staff.By car networking system, can obtain
The data information that vehicle is generated in operational process, in data information, include Vehicular battery voltage, electric current, battery it is electric in real time
The online datas information such as amount.
Step 202, according to the online data of first group of training sample, estimate the battery of first group of training sample
Estimated capacity.
In some embodiments, to there are many modes of the battery estimated capacity of battery estimation, for example, can be pacified by electric current
When integration method calculate, specifically, being referred to following formula:
In formula, C indicates that battery estimated capacity, I indicate that the transient current of battery, t indicate the time, and SOC indicates the surplus of battery
Remaining electricity.
Wherein, electric current and battery real time electrical quantity can be the electric current and battery reality in the online data obtained in step 201
When electricity.
Furthermore it is also possible to pass through open circuit voltage method, current integration method, internal resistance method, neural network and Kalman filtering method etc.
Mode estimates battery estimated capacity.Wherein open circuit voltage method is due to open-circuit voltage to be expected, it is therefore desirable to quiet for a long time
Battery pack is set, internal resistance method is also difficult to realize, neural network and Kalman filtering method there is the difficulty of estimation internal resistance on hardware
Then due to the difficulty of system setting, and cost is very high when application in battery management system, provides no advantage against.Relative to open circuit
For voltage method, internal resistance method, neural network and Kalman filtering method sheet, current integration method estimation procedure is simple, effective, therefore this
Battery estimated capacity is estimated using current integration method in implementation.
Step 203, the corresponding identification information of the battery estimated capacity for obtaining first group of training sample.
In some embodiments, trained sample can be estimated by above-mentioned current integration method by the online data of training sample
This battery estimated capacity, stamps corresponding identification information according to how many pairs of training samples of battery estimated capacity.Wherein, described
Identification information is dead battery or non-dead battery, wherein when the battery estimated capacity of the training sample is greater than or equal in advance
If when value, the identification information is non-dead battery, described when the battery estimated capacity of the training sample is less than preset value
Identification information is dead battery.
In some embodiments, preset value can be, but not limited to be 80%.
Step 204, according to the battery estimated capacity and corresponding identification information of first group of training sample, using support
Vector machine algorithm is trained, and obtains the supporting vector machine model.
In some embodiments, support vector machines (Support Vector Machine, SVM) is one kind by supervised learning side
Formula carries out the generalized linear classifier of binary classification to data, and decision boundary is super flat to the maximum back gauge of learning sample solution
Face.It by online data and corresponding identification information, is trained using algorithm of support vector machine, obtains supporting vector machine model
Afterwards, it directly can directly judge that tested battery is effective battery or inert cell by online data, thus to tested electricity
Pond carries out preliminary differentiation.
Specifically, the essence of SVM training is to solve quadratic programming problem (Quadruple Programming, a feeling the pulse with the finger-tip
Scalar functions are quadratic function, and constraint condition is the optimization problem of linear restriction), what is obtained is globally optimal solution, this has it
The superiority that hardly matches of other statistical learning technologies.The text classification effect of SVM classifier is fine, is best classifier
One of.Original sample space is converted using kernel function to higher dimensional space simultaneously, is able to solve original sample linearly not
The problem of can dividing.The disadvantage is that the selection of kernel function lacks guidance, it is difficult to select optimal kernel function for particular problem；In addition
SVM training speed is greatly influenced by training set scale, and computing cost is bigger, for the training speed problem of SVM, is ground
The person of studying carefully proposes many improved methods, including Chunking method, Osuna algorithm, SMO algorithm and interaction SVM etc..Svm classifier
The advantages of device, is that versatility is preferable, and nicety of grading is high, classification speed is fast, classification speed is unrelated with training sample number, In
All slightly it is better than kNN and Nae Bayesianmethod in terms of looking into quasi- and recall ratio.
Step 205, the online data for obtaining tested battery；
In some embodiments, online data may include voltage, electric current, battery real time electrical quantity of battery etc..
Step 206, using the trained supporting vector machine model, will be described according to the online data of the tested battery
Tested battery divides into dead battery and non-dead battery, so that the SOH to non-dead battery differentiates.
It is more accurate to tested cell classification subsequently through battery SOH discrimination model in order to make in some embodiments, this reality
It applies in example, tested battery is first divided into dead battery and non-dead battery, so, it is possible to reduce in battery SOH discrimination model
Classification, keep the differentiation of battery SOH discrimination model more accurate.
Wherein it is possible to which non-dead battery is labeled as E class.
Step 207, according to the online data, the mode input variable of the tested battery, the model is calculated
Input variable is the characteristic quantity extracted according to the IC curve of the tested battery；
In some embodiments, there are many modes of calculating battery model input variable, for example, online data can be first passed through
In voltage and battery real time electrical quantity draw the voltage-capacity curve of tested battery, then convert voltage-capacity curve to
IC curve extracts characteristic quantity in IC curve, and then using characteristic quantity as mode input variable.
Specifically, the acquisition pattern of the mode input variable of tested battery can be realized by following steps:
1, voltage-capacity curve is established according to the voltage and the battery real time electrical quantity that are tested battery；
Wherein, being tested the voltage of battery and battery real time electrical quantity can be obtained by online data, the voltage-capacity of foundation
Curve can be using voltage as abscissa, and battery real time electrical quantity establishes rectangular coordinate system as ordinate.
It 2, is IC curve by the voltage-capacity Curve transform；
It wherein, can be with by the mode that the voltage-capacity Curve transform is IC curve are as follows:
The first step calculates the ratio of electricity increment and voltage increment according to the voltage-capacity curve；
Second step, using the electricity increment and voltage increment ratio as ordinate, the voltage value is built as abscissa
Vertical IC curve.
3, the mode input variable is extracted on the IC curve.
Fig. 3 is the schematic diagram for the battery IC curve that one embodiment of the invention provides.Referring in Fig. 3, peak3 and peak4 are
Two different peak values will come out both as Characteristic Extraction and be used as mode input variable.
Wherein, the mode input variable are as follows: the height value and corresponding positional value at each peak on the IC curve.It is corresponding
Positional value is the corresponding voltage value of the peak value.In battery variety difference, the variation at each peak is also different on IC curve, example
Such as, the position at the peak of ferric phosphate lithium cell can move closer to, and the summit of lithium manganate battery gradually merges.
Step 208, the characteristic quantity for obtaining second group of training sample.
In some embodiments, second group of training sample can also be by obtaining in above-mentioned vehicle net system.Wherein, described
Two groups of training samples are non-dead battery, and non-dead battery can screen to obtain by above-mentioned supporting vector machine model.In addition,
The characteristic quantity is extracted according to the IC curve of the non-dead battery.
Step 209 clusters the characteristic quantity of second group of training sample, plurality of classes is clustered into, by every type
It Zuo Wei not a kind of SOH classification.
In some embodiments, clustering to characteristic quantity can be completed by Clustering Model, and Clustering Model can be by physics
Or the set of abstract object is divided into the multiple classes being made of similar object.In the present embodiment, can be by Clustering Model, it will
Training sample is divided into multiple SOH classification.
Step 210, using the characteristic quantity as the input of the battery SOH discrimination model, described in the SOH classification is used as
The output of battery SOH discrimination model constructs the battery SOH discrimination model.
In some embodiments, battery SOH discrimination model is obtained by characteristic quantity and SOH classification based training, it is subsequent to lead to
Cross the SOH classification that battery SOH discrimination model directly judges tested battery.
Step 211, the battery SOH discrimination model for training the mode input variable input of the tested battery in advance, really
The SOH classification that the SOH of the fixed tested battery belongs to.
In some embodiments, by the battery SOH discrimination model that training obtains in advance, pass through step 206 in tested battery
Method the mode input variable of tested battery is calculated after, can directly be determined by battery SOH discrimination model tested
Classification described in the SOH of battery.
Specifically, tested battery has been divided into non-dead battery by above-mentioned supporting vector machine model there are many SOH classification
And inert cell, and pass through battery SOH discrimination model and non-dead battery has been divided into multiple SOH classification.For example, 4 can be divided into
Class, respectively corresponds A class, B class, C class, D class, and dead battery is divided into E class.
Step 212, using generalized linear model trained in advance, the service condition of tested battery is estimated and is shown
To user, the service condition includes at least one in following item: using the battery SOH after preset duration；It uses default time
Battery SOH after number；Battery SOH decays to use duration when preset value.
In some embodiments, trained generalized linear model can be by the way that in car networking system database, difference makes in advance
It is established with time but the identical a large amount of real vehicle datas of other use conditions.It can be for tested battery by generalized linear model
Particular condition in use is estimated.Wherein, service condition can be, but not limited to use the battery SOH after preset duration；It uses
Battery SOH after preset times；Battery SOH decays to use duration when preset value.Specifically, (such as using specific duration
1 year) after battery SOH；Use the battery SOH after fixed number of times (such as 100 times)；Battery SOH decays to 80% use
Duration.
In the present embodiment, online data is used by online data, determines that battery health degree is gone forward side by side by IC curvilinear characteristic
Row performance prediction；The law characteristic between IC curvilinear characteristic and performance variations is obtained by mass data, and is realized according to this feature
The judgement of battery health degree, performance prediction.
The present invention can not only identify dead battery, can also differentiate to non-dead battery SOH.First with a large amount of instructions
Practice data and battery SOH demarcation interval is obtained into multiple SOH classification, then finds matching degree most for battery SOH discrimination model
High SOH classification, differentiates that result is more acurrate.
The application uses the data obtained based on car networking technology, analyze simultaneously real time discriminating battery using online data
SOH carries out performance prediction to battery in the current use state at the same time, based on personalization (according to the currently used habit of user
Used prediction) and two kinds of use patterns of standardization (being predicted according to specification electrochemical cell usage mode) provide a user following three points and believe
Breath: 1, using the battery SOH after specific duration (such as 1 year)；2, using the battery SOH after fixed number of times (such as 100 times)；
3, battery SOH decays to 80% use duration.
Fig. 4 is the structural schematic diagram for the battery SOH estimation device that one embodiment of the invention provides.As shown in figure 4, this implementation
Example provides a kind of battery SOH estimation device, comprising:
Module 401 is obtained, for obtaining the online data of tested battery；
Computing module 402, for the mode input variable of the tested battery to be calculated according to the online data,
The mode input variable is the characteristic quantity extracted according to the IC curve of the tested battery；
SOH discrimination module 403, the battery SOH trained in advance for the mode input variable input by the tested battery
Discrimination model determines the SOH classification that the SOH of the tested battery belongs to, and the SOH classification is a variety of altogether.
In the present embodiment, the online data of tested battery is obtained by obtaining module 401, then passes through computing module 402
According to the online data, the mode input variable of the tested battery is calculated, wherein the mode input variable is root
The characteristic quantity extracted according to the IC curve of the tested battery；Finally according to SOH discrimination module 403 by the model of the tested battery
Input variable input battery SOH discrimination model trained in advance determines the SOH classification that the SOH of the tested battery belongs to, described
SOH classification is a variety of altogether.In this way, replacing using test data in the prior art using the online data of battery, realize online
Real-time measurement, so that the result of measurement is more accurate；Also, the SOH that battery can be directly obtained according to online data, is saved
Time for being tested in the lab, shorten the test period.In addition, being finally obtained by the calculating to online data
Mode input variable is inputted in battery SOH discrimination model, can directly determine SOH belonging to the SOH of tested battery points
Class realizes the real time discriminating of battery SOH, easy to operate.
The specific implementation of the present embodiment may refer to the battery SOH estimation method and method reality of previous embodiment record
The related description in example is applied, details are not described herein again.
Fig. 5 is a kind of structural schematic diagram for battery SOH estimation equipment that one embodiment of the application provides.Referring to Fig. 5, this Shen
Please embodiment provide a kind of battery SOH estimation equipment, comprising:
Processor 501, and the memory 502 being connected with processor；
Memory 502 is for storing computer program；
Processor 501 is for calling and executing the computer program in memory 502, to execute as in embodiment one or two
Battery SOH estimation method.
The specific implementation of the present embodiment may refer to the battery SOH estimation method and method reality of previous embodiment record
The related description in example is applied, details are not described herein again.
Another embodiment of the present invention provides a kind of storage medium, storage medium is stored with computer program, computer program
When being executed by processor, realize such as each step in battery SOH estimation method.
The specific implementation of the present embodiment may refer to the related description in above-mentioned battery SOH estimation method embodiment,
Details are not described herein again.
It is understood that same or similar part can mutually refer in the various embodiments described above, in some embodiments
Unspecified content may refer to the same or similar content in other embodiments.
It should be noted that in the description of the present invention, term " first ", " second " etc. are used for description purposes only, without
It can be interpreted as indication or suggestion relative importance.In addition, in the description of the present invention, unless otherwise indicated, the meaning of " multiple "
Refer at least two.
Any process described otherwise above or method description are construed as in flow chart or herein, and expression includes
It is one or more for realizing specific logical function or process the step of executable instruction code module, segment or portion
Point, and the range of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discussed suitable
Sequence, including according to related function by it is basic simultaneously in the way of or in the opposite order, Lai Zhihang function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
It should be appreciated that each section of the invention can be realized with hardware, software, firmware or their combination.Above-mentioned
In embodiment, software that multiple steps or method can be executed in memory and by suitable instruction executing device with storage
Or firmware is realized.It, and in another embodiment, can be under well known in the art for example, if realized with hardware
Any one of column technology or their combination are realized: having a logic gates for realizing logic function to data-signal
Discrete logic, with suitable combinational logic gate circuit specific integrated circuit, programmable gate array (PGA), scene
Programmable gate array (FPGA) etc..
Those skilled in the art are understood that realize all or part of step that above-described embodiment method carries
It suddenly is that relevant hardware can be instructed to complete by program, the program can store in a kind of computer-readable storage medium
In matter, which when being executed, includes the steps that one or a combination set of embodiment of the method.
It, can also be in addition, each functional unit in each embodiment of the present invention can integrate in a processing module
It is that each unit physically exists alone, can also be integrated in two or more units in a module.Above-mentioned integrated mould
Block both can take the form of hardware realization, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized and when sold or used as an independent product in the form of software function module, also can store in a computer
In read/write memory medium.
Storage medium mentioned above can be read-only memory, disk or CD etc..
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
Centainly refer to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be any
One or more embodiment or examples in can be combined in any suitable manner.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (10)
1. a kind of battery SOH estimation method characterized by comprising
Obtain the online data of tested battery；
According to the online data, the mode input variable of the tested battery is calculated, the mode input variable is root
The characteristic quantity extracted according to the IC curve of the tested battery；
By the mode input variable input of tested battery battery SOH discrimination model trained in advance, the tested electricity is determined
The SOH classification that the SOH in pond belongs to, the SOH classification are a variety of altogether.
2. the method according to claim 1, wherein the online data includes: voltage, battery real time electrical quantity,
It is described according to the online data, the mode input variable of the tested battery is calculated, comprising:
Voltage-capacity curve is established according to the voltage and the battery real time electrical quantity；
It is IC curve by the voltage-capacity Curve transform；
The mode input variable is extracted on the IC curve.
3. according to the method described in claim 2, it is characterized in that, the mode input variable are as follows: each peak on the IC curve
Height value and corresponding positional value.
4. described the method according to claim 1, wherein after the online data for obtaining tested battery
Method further include:
Using supporting vector machine model trained in advance, the tested battery is divided by mistake according to the online data of tested battery
Battery and non-dead battery are imitated, so that the SOH to non-dead battery differentiates.
5. according to the method described in claim 4, it is characterized by further comprising:
Obtain the online data of first group of training sample；
According to the online data of first group of training sample, the battery estimated capacity of first group of training sample is estimated；
The corresponding identification information of the battery estimated capacity of first group of training sample is obtained, the identification information is failure
Battery or non-dead battery, wherein when the battery estimated capacity of first group of training sample is greater than or equal to preset value, institute
Stating identification information is non-dead battery, when the battery estimated capacity of first group of training sample is less than preset value, the mark
Knowledge information is dead battery；
According to the battery estimated capacity and corresponding identification information of first group of training sample, using algorithm of support vector machine into
Row training, obtains the supporting vector machine model.
6. according to the method described in claim 5, it is characterized in that, the online data includes: electric current, battery real time electrical quantity,
The online data according to first group of training sample estimates the battery estimated capacity of first group of training sample, packet
It includes:
First group of training sample is calculated using current integration method according to the electric current and the battery real time electrical quantity
Battery estimated capacity.
7. the method according to claim 1, wherein further include:
The characteristic quantity of second group of training sample is obtained, second group of training sample is non-dead battery, and the characteristic quantity is root
It is extracted according to the IC curve of the non-dead battery；
The characteristic quantity of second group of training sample is clustered, plurality of classes is clustered into, using every kind of classification as a kind of SOH
Classification；
Using the characteristic quantity as the input of the battery SOH discrimination model, the SOH classification differentiates as the battery SOH
The output of model constructs the battery SOH discrimination model.
8. according to the method described in claim 3, it is characterized in that, described convert IC curve for the voltage-capacity curve,
Include:
The ratio of electricity increment and voltage increment is calculated according to the voltage-capacity curve；
Using the electricity increment and voltage increment ratio as ordinate, the voltage establishes IC curve as abscissa.
9. the method according to claim 1, wherein further include:
Using generalized linear model trained in advance, user is estimated the service condition of tested battery and showed, it is described
Service condition includes at least one in following item: using the battery SOH after preset duration；Use the battery after preset times
SOH；Battery SOH decays to use duration when preset value.
10. a kind of battery SOH estimates equipment characterized by comprising
Processor, and the memory being connected with the processor；
The memory is for storing computer program；
The processor is for calling and executing the computer program in the memory, to execute such as claim 1-9
Described in any item methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910784277.0A CN110441706A (en) | 2019-08-23 | 2019-08-23 | A kind of battery SOH estimation method and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910784277.0A CN110441706A (en) | 2019-08-23 | 2019-08-23 | A kind of battery SOH estimation method and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110441706A true CN110441706A (en) | 2019-11-12 |
Family
ID=68437429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910784277.0A Pending CN110441706A (en) | 2019-08-23 | 2019-08-23 | A kind of battery SOH estimation method and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110441706A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826645A (en) * | 2019-11-22 | 2020-02-21 | 四川长虹电器股份有限公司 | Adaboost algorithm-based lithium battery retirement detection method and system |
| CN112014745A (en) * | 2020-08-27 | 2020-12-01 | 洛阳理工学院 | Three-level fault diagnosis method for power battery |
| WO2021185308A1 (en) * | 2020-03-18 | 2021-09-23 | 北京理工大学 | Online determination method and system for state of health of power battery pack of electric vehicle |
| CN113835036A (en) * | 2020-06-24 | 2021-12-24 | 丰翼科技(深圳)有限公司 | Battery health state evaluation method and device, computer equipment and storage medium |
| CN114298507A (en) * | 2021-12-17 | 2022-04-08 | 欣旺达电子股份有限公司 | Lithium battery failure determination method, model generation method, device and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102749589A (en) * | 2012-07-13 | 2012-10-24 | 哈尔滨工业大学深圳研究生院 | Recession-mode predicting method of power battery of electric automobile |
| CN103675702A (en) * | 2013-12-04 | 2014-03-26 | 清华大学 | Method for evaluating state of health battery in real time |
| WO2015031437A1 (en) * | 2013-08-27 | 2015-03-05 | The Regents Of The University Of Michigan | On-board state of health monitoring of batteries using incremental capacity analysis |
| CN106980091A (en) * | 2017-03-29 | 2017-07-25 | 北京理工大学 | A kind of electrokinetic cell system health status method of estimation based on fractional model |
| CN108303649A (en) * | 2017-01-13 | 2018-07-20 | 重庆邮电大学 | A kind of cell health state recognition methods |
| CN108414938A (en) * | 2018-01-18 | 2018-08-17 | 武汉理工大学 | Batteries of electric automobile SOH online evaluation methods based on electric vehicle monitor supervision platform |
| CN108445406A (en) * | 2018-03-13 | 2018-08-24 | 桂林电子科技大学 | A kind of power battery health status method of estimation |
| CN108983106A (en) * | 2018-07-27 | 2018-12-11 | 国网重庆市电力公司电力科学研究院 | Novel power battery health state evaluation method |
| CN109932663A (en) * | 2019-03-07 | 2019-06-25 | 清华四川能源互联网研究院 | Cell health state appraisal procedure, device, storage medium and electronic device |
| CN110058178A (en) * | 2019-05-14 | 2019-07-26 | 郑州大学 | A kind of lithium battery method for detecting health status and system |
-
2019
- 2019-08-23 CN CN201910784277.0A patent/CN110441706A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102749589A (en) * | 2012-07-13 | 2012-10-24 | 哈尔滨工业大学深圳研究生院 | Recession-mode predicting method of power battery of electric automobile |
| WO2015031437A1 (en) * | 2013-08-27 | 2015-03-05 | The Regents Of The University Of Michigan | On-board state of health monitoring of batteries using incremental capacity analysis |
| CN103675702A (en) * | 2013-12-04 | 2014-03-26 | 清华大学 | Method for evaluating state of health battery in real time |
| CN108303649A (en) * | 2017-01-13 | 2018-07-20 | 重庆邮电大学 | A kind of cell health state recognition methods |
| CN106980091A (en) * | 2017-03-29 | 2017-07-25 | 北京理工大学 | A kind of electrokinetic cell system health status method of estimation based on fractional model |
| CN108414938A (en) * | 2018-01-18 | 2018-08-17 | 武汉理工大学 | Batteries of electric automobile SOH online evaluation methods based on electric vehicle monitor supervision platform |
| CN108445406A (en) * | 2018-03-13 | 2018-08-24 | 桂林电子科技大学 | A kind of power battery health status method of estimation |
| CN108983106A (en) * | 2018-07-27 | 2018-12-11 | 国网重庆市电力公司电力科学研究院 | Novel power battery health state evaluation method |
| CN109932663A (en) * | 2019-03-07 | 2019-06-25 | 清华四川能源互联网研究院 | Cell health state appraisal procedure, device, storage medium and electronic device |
| CN110058178A (en) * | 2019-05-14 | 2019-07-26 | 郑州大学 | A kind of lithium battery method for detecting health status and system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826645A (en) * | 2019-11-22 | 2020-02-21 | 四川长虹电器股份有限公司 | Adaboost algorithm-based lithium battery retirement detection method and system |
| WO2021185308A1 (en) * | 2020-03-18 | 2021-09-23 | 北京理工大学 | Online determination method and system for state of health of power battery pack of electric vehicle |
| CN113835036A (en) * | 2020-06-24 | 2021-12-24 | 丰翼科技(深圳)有限公司 | Battery health state evaluation method and device, computer equipment and storage medium |
| CN112014745A (en) * | 2020-08-27 | 2020-12-01 | 洛阳理工学院 | Three-level fault diagnosis method for power battery |
| CN114298507A (en) * | 2021-12-17 | 2022-04-08 | 欣旺达电子股份有限公司 | Lithium battery failure determination method, model generation method, device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110441706A (en) | A kind of battery SOH estimation method and equipment | |
| US10408881B2 (en) | Method and apparatus for estimating state of battery | |
| CN110058178A (en) | A kind of lithium battery method for detecting health status and system | |
| US7333962B2 (en) | Techniques to organize test results | |
| CN107632262A (en) | A kind of detection method and device of power battery pack cycle life | |
| JP2022514992A (en) | Battery diagnostic system, battery diagnostic method, and storage medium | |
| CN109991545B (en) | Battery pack electric quantity detection method and device and terminal equipment | |
| CN112818602B (en) | Battery digital twin control method and device based on big data analysis | |
| CN109613440A (en) | Stage division, device, equipment and the storage medium of battery | |
| CN114047452B (en) | Method and device for determining cycle life of battery | |
| CN109817339A (en) | Patient's group technology and device based on big data | |
| Zhang et al. | In-situ battery life prognostics amid mixed operation conditions using physics-driven machine learning | |
| CN112180258A (en) | Method, device, medium, terminal and system for measuring average coulomb efficiency of battery | |
| CN100470418C (en) | Polypropylene melting index softsensoring instrument based on blind signal analysis and method thereof | |
| CN105807231A (en) | Method and system for detecting residual capacity of storage battery | |
| CN109768340A (en) | A kind of battery discharge procedure voltage inconsistency estimation method and device | |
| CN112084459A (en) | Method and device for predicting battery charge-discharge cycle life, electronic terminal and storage medium | |
| CN114594378A (en) | Lithium battery SOH model training method, prediction method, device and equipment | |
| Khaleghi et al. | Towards machine-learning driven prognostics and health management of Li-ion batteries. A comprehensive review | |
| CN114912720A (en) | Memory network-based power load prediction method, device, terminal and storage medium | |
| CN117434429A (en) | Chip stability testing method and related device | |
| Lu et al. | A feature fusion-based convolutional neural network for battery state-of-health estimation with mining of partial voltage curve | |
| CN117725476A (en) | Training method and device of battery health degree prediction model and energy storage system | |
| CN118033432A (en) | Battery state of charge estimation method and device and computer equipment | |
| CN208350965U (en) | A kind of pure electric automobile battery state-of-health measures device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191112 |
|
| RJ01 | Rejection of invention patent application after publication |