CN113820608A - Method for predicting remaining capacity of battery in echelon and electronic equipment - Google Patents

Method for predicting remaining capacity of battery in echelon and electronic equipment Download PDF

Info

Publication number
CN113820608A
CN113820608A CN202110964990.0A CN202110964990A CN113820608A CN 113820608 A CN113820608 A CN 113820608A CN 202110964990 A CN202110964990 A CN 202110964990A CN 113820608 A CN113820608 A CN 113820608A
Authority
CN
China
Prior art keywords
training
voltage
obtaining
groups
echelon
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110964990.0A
Other languages
Chinese (zh)
Other versions
CN113820608B (en
Inventor
徐鹏
郝一
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Posts and Telecommunications
Original Assignee
Beijing University of Posts and Telecommunications
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing University of Posts and Telecommunications filed Critical Beijing University of Posts and Telecommunications
Priority to CN202110964990.0A priority Critical patent/CN113820608B/en
Publication of CN113820608A publication Critical patent/CN113820608A/en
Application granted granted Critical
Publication of CN113820608B publication Critical patent/CN113820608B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
    • 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
    • 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
    • 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
    • G06N3/048—Activation functions
    • 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
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Abstract

The disclosure provides a prediction method of echelon battery residual capacity and an electronic device. The method comprises the following steps: obtaining online use data of the echelon battery, and generating a feature vector of the online use data according to the online use data of the echelon battery; and obtaining the residual capacity of the echelon battery at the stable voltage starting point according to the feature vector of the on-line use data and a pre-trained residual capacity prediction network model. In the embodiment of the description, the residual capacity of the echelon battery is finally obtained through the online use data of the echelon battery and the pre-trained residual capacity prediction network model.

Description

Method for predicting remaining capacity of battery in echelon and electronic equipment
Technical Field
The disclosure relates to the field of echelon battery management, in particular to a method for predicting remaining capacity of a echelon battery.
Background
In modern society, batteries are increasingly used in a variety of devices. In the process of repeated charging and discharging, the capacity of the storage battery is attenuated continuously, and the ex-service treatment is required to be carried out to a certain degree. In order to utilize the storage battery to the maximum extent, the battery needs to be utilized in a gradient manner, so that the estimation of the residual capacity of the battery is an important link in terms of measuring whether the battery is retired or not and evaluating the gradient utilization of the retired battery.
Most of the existing methods for estimating the residual capacity of the battery have the defect of high dependency on the type of the battery, a model constructed through a large number of experiments is only suitable for a certain type, and the estimation methods of the residual capacity of the battery are influenced by the aging of the battery, so that the accuracy is reduced.
Disclosure of Invention
In view of the above, an object of the present disclosure is to provide a method for predicting a remaining capacity of a battery in a stepped manner and an electronic device.
Based on the above purpose, the present disclosure provides a method for predicting remaining capacity of a battery in a echelon, including:
obtaining online use data of the echelon battery, and generating a feature vector of the online use data according to the online use data of the echelon battery;
and obtaining the residual capacity of the echelon battery at the stable voltage starting point according to the feature vector of the on-line use data and a pre-trained residual capacity prediction network model.
Optionally, the process of obtaining the online usage data of the echelon battery and generating the feature vector of the online usage data according to the online usage data of the echelon battery specifically includes:
acquiring a current value and a voltage value of the online use data of the echelon battery;
respectively obtaining a current curve I '(t) and a voltage curve V' (t) according to the current value and the voltage value;
obtaining a discharged electricity quantity curve Q ' according to the current curve I ' (t) 'dis(t)；
According to the voltage curve V '(t) and the discharged electricity quantity curve Q'dis(t) obtaining a stable voltage starting point V'0Time t'0And discharged electric quantity Q'dis0；
According to the discharged electric quantity Q'dis0And a unit discharged electric quantity delta Q 'to obtain a discharge quantity Q'dis0Several electric quantity values Q 'inside'diss；
According to a plurality of said electric quantitiesValue Q'dissAnd the discharged electricity quantity curve Q'dis(t) obtaining each of said electric quantity values Q'dissCorresponding time t's；
According to the time t'sAnd the voltage curve V' (t) obtains a voltage value V corresponding to each times′；
Corresponding voltage value V to each times' as a feature vector of said on-net usage data.
Optionally, the training process of the remaining capacity prediction network model includes:
establishing a Back Propagation Neural Network (BPNN);
obtaining a plurality of groups of ex-factory discharge test data of training echelon batteries;
taking a part of the factory discharge test data of the plurality of groups of training echelon batteries as a training set; the training set includes: a training current value I and a training voltage value V;
obtaining a plurality of groups of stable voltage initial points V for training according to the voltage value V for training0And a plurality of groups of discharge inflection points n for training;
according to the plurality of groups of stable voltage initial points V for training0Obtaining a plurality of groups of training voltage values V by the training current value I and the unit discharge quantity delta Qs(ii) the sequence of (a);
according to the plurality of groups of stable voltage initial points V for training0And the discharge inflection points n for the training of the groups are used for obtaining a plurality of groups of discharged electric quantities Q for the trainingn；
According to the plurality of groups of voltage values V for trainingsComposed sequence and sets of said discharged quantity of electricity Q for trainingnAnd training the back propagation neural network BPNN to obtain the residual capacity prediction network.
Optionally, the current curve I '(t) is obtained by interpolation, fitting and calculation of the current value, and the voltage curve V' (t) is obtained by interpolation, fitting and calculation of the voltage value.
Optionally, the residual capacity prediction network model is composed of a 4-layer details full-connection network layer and a Relu activation layer.
Optionally, the stable voltage starting point V0Is the first point of the voltage value V, wherein V is less than or equal to 3.3 volts.
Optionally, the discharge inflection point n is calculated in the following manner:
according to the voltage value V, making difference delta VtV (t +1) -V (t), where t is time; when at a certain time t, if Δ V is satisfied at the same timetLess than-0.03V and delta Vt+1Less than-0.03V and delta Vt+2The time t at this time is obtained as the discharge inflection point n under the condition of < -0.03V.
Optionally, the training voltage value VsThe way the composed sequence is calculated is:
obtaining a voltage acquisition time interval m according to the current value I and the unit discharge quantity delta Q; the time interval m of the collected voltage is delta Q/I;
according to the acquisition voltage time interval m and the stable voltage starting point V for training0Obtaining the training voltage value VsThe sequence of the composition.
Optionally, the electric quantity QnFor the starting point V of the stabilized voltage0And the electric quantity of the discharge inflection point n.
Based on the same inventive concept, the present disclosure also provides a device for predicting the remaining capacity of a battery in a echelon, comprising:
the acquisition module is used for acquiring online use data of the echelon battery and generating a feature vector of the online use data according to the online use data of the echelon battery;
and the result obtaining module is used for predicting a network model according to the feature vector of the on-line use data and the pre-trained residual capacity to obtain the residual capacity of the echelon battery at the stable voltage starting point.
Based on the same inventive concept, the present disclosure also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for predicting the remaining capacity of a battery in a echelon manner as described in any one of the above.
From the above, according to the method for predicting the residual capacity of the echelon battery provided by the disclosure, the residual capacity of the echelon battery at the stable voltage starting point is obtained by acquiring the on-line use data of the echelon battery, generating the feature vector of the on-line use data, and giving the pre-trained residual capacity prediction network model. For the existing echelon battery, the residual capacity of the echelon battery is predicted by establishing a residual capacity prediction network model, so that the retired echelon battery can be more effectively utilized; meanwhile, the residual capacity prediction network model has stronger applicability, can predict the capacity of various echelon battery types after effective training, is not like the traditional method which can only predict one type of battery, and has higher prediction accuracy rate on the residual capacity through the effectively trained residual capacity prediction network model.
Drawings
In order to more clearly illustrate the technical solutions in the present disclosure or related technologies, the drawings needed to be used in the description of the embodiments or related technologies are briefly introduced below, and it is obvious that the drawings in the following description are only embodiments of the present disclosure, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a flowchart of a method for predicting remaining capacity of a battery in a echelon according to an embodiment of the present disclosure;
FIG. 2 is a flowchart illustrating obtaining of feature vectors of online usage data of a echelon battery according to an embodiment of the disclosure;
fig. 3 is a schematic diagram of a training process of a echelon battery remaining capacity prediction network model according to an embodiment of the present disclosure;
fig. 4 is a structural diagram of a device for predicting the remaining capacity of a battery in a echelon according to an embodiment of the present disclosure;
fig. 5 is an electronic device structure according to an embodiment of the disclosure.
Detailed Description
For the purpose of promoting a better understanding of the objects, aspects and advantages of the present disclosure, reference is made to the following detailed description taken in conjunction with the accompanying drawings.
It is to be noted that technical terms or scientific terms used in the embodiments of the present disclosure should have a general meaning as understood by those having ordinary skill in the art to which the present disclosure belongs, unless otherwise defined. The use of "first," "second," and similar terms in the embodiments of the disclosure is not intended to indicate any order, quantity, or importance, but rather to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", and the like are used merely to indicate relative positional relationships, and when the absolute position of the object being described is changed, the relative positional relationships may also be changed accordingly.
As described in the background section, the battery remaining capacity estimation methods in the related art are generally applicable to only a certain model, however, the usage amount of the storage battery is gradually increasing nowadays, and the prior art can only predict the remaining capacity of a certain model of battery, and virtually increase a lot of workload, which brings great inconvenience to the work of predicting the remaining capacity of the battery, and meanwhile, the precision of many prediction methods decreases along with the aging of the battery, resulting in inaccurate capacity prediction results.
In view of the above, the embodiment of the present disclosure provides a method for predicting remaining capacity of a battery in a echelon mode, which predicts the remaining capacity of the battery in the echelon mode based on a pre-trained remaining capacity prediction network model by using battery factory discharge test data such as a training current value I and a training voltage value V and based on the pre-trained remaining capacity prediction network model, so that the prediction of the remaining capacity of the battery is more accurate, and the utilization rate of the battery is further improved.
Hereinafter, the technical means of the embodiments of the present disclosure will be described in detail by specific examples.
Referring to fig. 1, a method for predicting remaining capacity of a battery in a echelon according to an embodiment of the present disclosure includes the following steps:
step S101, obtaining the online use data of the echelon battery, and generating a feature vector of the online use data according to the online use data of the echelon battery.
In the step, firstly, on-line use data of the battery in the echelon is obtained, a corresponding feature vector is further generated, and the residual capacity of the battery in the subsequent echelon is predicted based on the feature vector of the on-line use data, wherein the on-line use data comprises a current value and a voltage value.
And step S103, obtaining the residual capacity of the echelon battery at the stable voltage starting point according to the feature vector of the online use data and a pre-trained residual capacity prediction network model.
In the step, the feature vector of the on-line use data obtained in the step is input into a pre-trained residual capacity prediction network model to obtain the residual capacity of the echelon battery at the stable voltage starting point.
In some embodiments, the step of obtaining the feature vector of the grid use data of the echelon battery is described in detail in this embodiment. As shown in fig. 2, obtaining the feature vector of the online usage data of the echelon battery includes the following steps:
step S201, obtaining the current value and the voltage value of the echelon battery on-line use data.
The current value and the voltage value of the grid use data of the battery in the echelon mode are obtained firstly. In this embodiment, the data of the grid-connected use of the echelon battery is preprocessed, where the preprocessing includes validity check of the data, data cleaning, and data smoothing, and then the current value and the voltage value of the processed data of the grid-connected use of the echelon battery are extracted.
In step S203, a current curve I '(t) and a voltage curve V' (t) are obtained according to the current value and the voltage value, respectively. In this embodiment, the voltage curve V '(t) is obtained by performing interpolation fitting processing on the voltage values of the grid use data of the battery in steps, and the current value of the collected grid use data of the battery in steps is not a constant value, so that the current curve I' (t) needs to be obtained by performing interpolation fitting processing on the current value.
In step S205, a discharged electricity quantity curve Q ' is obtained from the current curve I ' (t) 'dis(t) of (d). In this embodiment, the current curve I '(t) is integrated to obtain the discharged electricity amount curve Q'dis(t)。
Step S207, according to the voltage curve V '(t) and the discharged electricity quantity curve Q'dis(t) obtaining a stable voltage starting point V'0Time t'0And discharged electric quantity Q'dis0. In the present embodiment, the stable voltage starting point V0' As a first point satisfying V.ltoreq.3.3 volts in the voltage curve V ' (t), after V.ltoreq.3.3 volts, the voltage curve V ' (t) starts to become smoothly stable, and therefore the stable voltage starting point V0'Take the first Point, time t', that satisfies V.ltoreq.3.3 volts in Voltage Curve V '(t)'0For stabilizing the voltage starting point V0' at a corresponding point in time in the voltage curve V ' (t), in the quantity curve Q 'dis(t) middle time t'0The corresponding electric quantity value is discharged electric quantity Q'dis0。
Step S209, according to the discharged electricity quantity Q'dis0And a unit discharged electric quantity delta Q 'to obtain a discharge quantity Q'dis0Several electric quantity values Q 'inside'diss. In this example, the unit discharged electric energy Δ Q ═ 1.67AH, electric energy value Q'dissIs derived from discharged electricity quantity Q'dis0An electric quantity value Q 'is obtained at every unit discharged electric quantity delta Q'dissIn this embodiment, 20 are taken out in total, and in actual operation, other numbers, at least 10, can be taken out. In electric quantity value Q'dis1For example, the calculation formula is Q'dis1＝Q′dis0+ΔQ。
Step S211, according to a plurality of electric quantity values Q'dissAnd the discharged electricity quantity curve Q'dis(t) obtaining each of said electric quantity values Q'dissCorresponding time t's. In this example, there were 20 in total of Q'dissAt the time of placingElectricity output quantity curve Q'dis(t) finding corresponding 20 times t's。
Step S213, according to the time t'sAnd the voltage curve V '(t) is obtained for each time t'sCorresponding voltage value Vs'. In this example, there are 20 times t 'in total'sRespectively finding 20 voltage values V 'on the voltage curve V' (t)s', the 20 voltage values Vs' as a feature vector of said on-net usage data.
In some embodiments, the residual capacity prediction network model is obtained by pre-training. In this embodiment, 1498 sets of ex-factory discharge test data of the battery in echelon for training are taken as an example, and a training process of the residual capacity prediction network model is described in detail, because the data volume of the in-network use data is too small, the ex-factory discharge test data of the battery in echelon is used for training when the residual capacity prediction network model is trained, and the two data are obtained only in different manners. As shown in fig. 3, the training process of the model includes the following steps:
step S301, a back propagation neural network BPNN is established. In this embodiment, the back propagation neural network BPNN is composed of 4 layers of sense fully-connected network layers and one layer of Relu activation layer, where the parameters of the 4 layers of sense fully-connected network layers are 32, 64, 128, and 32 respectively to extract the features of the input data, and the parameters of the sense fully-connected network layers are all variables. The activation layer selects a Relu layer to add nonlinear elements into the network.
Step S303, obtaining a plurality of groups of ex-factory discharge test data of the training echelon batteries. In this embodiment, the factory discharge test data of the echelon battery is first preprocessed, where the preprocessing includes validity check, data cleaning, and data smoothing, so as to obtain 1498 groups of factory discharge test data of the echelon battery for training, where the factory discharge test data includes a current value, a voltage value, and an electric quantity.
Step S305, taking a part of factory discharge test data of the plurality of groups of training echelon batteries as a training set; the training set includes: a training current value I and a training voltage value V. In this embodiment, 1498 sets of factory discharge test data of the battery in the training echelon are obtained in total, including current value, voltage value and time, where 80% of the divided voltage value data is the training voltage value V, 20% is the testing voltage value, 80% of the divided current value data is the training current value I, and 20% is the testing current value. The training set comprises 1198 groups of voltage values V for training and 1198 groups of current values I for training corresponding to the voltage values V, and the current values I for training of the 1198 groups are both 100A.
Step S307, obtaining a plurality of groups of stable voltage starting points V for training according to the voltage value V for training0And several sets of discharge corners n for training. In this embodiment, the starting point V of the stable voltage for training0The first point of the training voltage value V is less than or equal to 3.3V to obtain 1498 stable voltage starting points V for training0. After V ≧ 3.3 volts, the voltage curve V '(t) begins to become smoothly stable, so the stable voltage starting point V'0The first point in the voltage curve V' (t) is taken that satisfies V.ltoreq.3.3 volts. The discharge inflection point n is positioned at the rear half section of discharge, and the difference is made on the training voltage value V: Δ Vt(t +1) -v (t), where t is time; when at a certain time t, if Δ V is satisfied at the same timetLess than-0.03V and delta Vt+1Less than-0.03V and delta Vt+2If the voltage is lower than-0.03V, the time t is obtained as a discharge inflection point n, and 1198 groups of discharge inflection points n are obtained.
Step S309, according to the plurality of groups of stable voltage initial points V for training0Obtaining a plurality of groups of training voltage values V by the training current value I and the unit discharge quantity delta Qs. In this example, the training current value I is 100A, the unit discharge amount Δ Q is 1.67AH, and the acquisition voltage time interval m is 1 minute, Δ Q/I. Voltage value V for trainingsThe calculation method is as follows: starting point V from stable voltage0Initially, a voltage value V is taken every acquisition voltage time interval msIn this embodiment, 20 are counted, and in actual operation, other numbers can be counted, but at least 10 are counted.
Step S311, according to the plurality of groups of stable voltage initial points V for training0And the plurality of groups of discharge inflection points n for training are obtainedAnd a plurality of groups of discharged electric quantity Q for training. In the present embodiment, 1198 groups of stable voltage starting points V for training are counted0And 1198 groups of discharge inflection points n for training, wherein the calculation mode of the discharged electricity quantity Q for training is as follows: at stable set point starting point V0There is a corresponding time t0Time t0With a corresponding electric quantity Q0At the discharge inflection point n, there is a corresponding time tnTime tnWith a corresponding electric quantity QnThe discharged power Q for training is Qn-Q0。
Step S313, according to the plurality of groups of voltage values V for trainingsAnd training the back propagation neural network BPNN by a plurality of groups of the training electricity discharging quantity Q to obtain the residual capacity prediction network. In the present embodiment, there are 1198 training voltage values V in totalsThe composed sequence and 1198 groups of discharge quantity Q for training train the back propagation neural network BPNN, and the input is a voltage value V for trainingsAnd the output of the formed sequence is the discharge capacity Q.
It should be noted that the method of the embodiments of the present disclosure may be executed by a single device, such as a computer or a server. The method of the embodiment can also be applied to a distributed scene and completed by the mutual cooperation of a plurality of devices. In such a distributed scenario, one of the devices may only perform one or more steps of the method of the embodiments of the present disclosure, and the devices may interact with each other to complete the method.
It should be noted that the above describes some embodiments of the disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments described above and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
Based on the same inventive concept, corresponding to the method of any embodiment, the disclosure also provides a device for predicting the remaining capacity of the battery in echelon.
Referring to fig. 4, the device for predicting the remaining capacity of the battery in the echelon includes:
the obtaining module 11 is configured to obtain online usage data of the echelon battery, and generate a feature vector of the online usage data according to the online usage data of the echelon battery.
And the result obtaining module 13 is configured to obtain the remaining capacity of the echelon battery at the stable voltage starting point according to the feature vector of the online use data and a pre-trained remaining capacity prediction network model.
As an optional embodiment, the acquiring process of the acquiring module 11 specifically includes:
acquiring a current value and a voltage value of the online use data of the echelon battery;
respectively obtaining a current curve I '(t) and a voltage curve V' (t) according to the current value and the voltage value;
obtaining a discharged electricity quantity curve Q ' according to the current curve I ' (t) 'dis(t)；
Obtaining a stable point voltage initial point V according to the voltage curve V' (t)0', time t'0And discharged electric quantity Q'dis0；
According to the discharged electric quantity Q'dis0And a unit discharged electric quantity delta Q 'to obtain a discharge quantity Q'dis0Several electric quantity values Q 'inside'diss；
According to a plurality of the electric quantity values Q'dissAnd the discharged electricity quantity curve Q'dis(t) obtaining each of said electric quantity values Q'dissCorresponding time t's；
According to the time t'sAnd the voltage curve V' (t) obtains a voltage value V corresponding to each times′；
Corresponding voltage value V to each times' as a feature vector of said on-net usage data.
As an alternative embodiment, the training process of the result obtaining module 13 includes:
establishing a Back Propagation Neural Network (BPNN);
obtaining a plurality of groups of ex-factory discharge test data of training echelon batteries;
taking a part of the factory discharge test data of the plurality of groups of training echelon batteries as a training set; the training set includes: a training current value I, a training voltage value V and time t;
obtaining a plurality of groups of stable voltage initial points V for training according to the voltage value V for training0And a plurality of groups of discharge inflection points n for training;
according to the plurality of groups of stable voltage initial points V for training0Obtaining a plurality of groups of training voltage values V by the training current value I and the unit discharge quantity delta Qs(ii) the sequence of (a);
according to the plurality of groups of stable voltage initial points V for training0And the plurality of groups of discharge inflection points n for training are obtained to obtain a plurality of groups of discharged electric quantities Qn for training;
according to the plurality of groups of voltage values V for trainingsAnd training a Back Propagation Neural Network (BPNN) by the formed sequence and a plurality of groups of discharged electric quantity Qn for training to obtain the residual capacity prediction network.
As an alternative embodiment, the current curve I '(t) is obtained by interpolation fitting calculation of the current values, and the voltage curve V' (t) is obtained by interpolation fitting calculation of the voltage values.
As an alternative embodiment, the residual capacity prediction network model is composed of a 4-layer density fully-connected network layer and a Relu activation layer.
As an alternative embodiment, the stable voltage starting point V0Is the first point of the voltage value V, wherein V is less than or equal to 3.3 volts.
As an optional embodiment, the discharge inflection point n is calculated by:
according to the voltage value V, making difference delta VtY (t +1) -v (t), where t is time; when at a certain time t, if at the same timeSatisfies Δ Vt＜-0.03、ΔVt+1＜-0.03、ΔVt+2The time t at this time was obtained as the discharge inflection point n under the condition of < -0.03.
As an alternative embodiment, the training voltage value VsThe way the composed sequence is calculated is:
obtaining a voltage acquisition time interval m according to the current value I and the unit discharge quantity delta Q; the time interval m of the collected voltage is delta Q/I;
according to the acquisition voltage time interval m and the stable voltage starting point V for training0Obtaining the training voltage value VsThe sequence of the composition.
As an alternative embodiment, the quantity of electricity QnFor the starting point V of the stabilized voltage0And the electric quantity of the discharge inflection point n.
For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. Of course, the functionality of the various modules may be implemented in the same one or more software and/or hardware implementations of the present disclosure.
The device of the above embodiment is used to implement the method for predicting the remaining capacity of the battery in any of the foregoing embodiments in a corresponding manner, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Based on the same inventive concept, corresponding to the method of any embodiment described above, the present disclosure further provides an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the method for predicting the remaining capacity of the battery in the echelon mode according to any embodiment described above is implemented.
Fig. 5 is a schematic diagram illustrating a more specific hardware structure of an electronic device according to this embodiment, where the electronic device may include: a processor 1010, a memory 1020, an input/output interface 1030, a communication interface 1040, and a bus 1050. Wherein the processor 1010, memory 1020, input/output interface 1030, and communication interface 1040 are communicatively coupled to each other within the device via bus 1050.
The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits, and is configured to execute related programs to implement the technical solutions provided in the embodiments of the present specification.
The Memory 1020 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, or the like. The memory 1020 may store an operating system and other application programs, and when the technical solution provided by the embodiments of the present specification is implemented by software or firmware, the relevant program codes are stored in the memory 1020 and called to be executed by the processor 1010.
The input/output interface 1030 is used for connecting an input/output module to input and output information. The i/o module may be configured as a component in a device (not shown) or may be external to the device to provide a corresponding function. The input devices may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices may include a display, a speaker, a vibrator, an indicator light, etc.
The communication interface 1040 is used for connecting a communication module (not shown in the drawings) to implement communication interaction between the present apparatus and other apparatuses. The communication module can realize communication in a wired mode (such as USB, network cable and the like) and also can realize communication in a wireless mode (such as mobile network, WIFI, Bluetooth and the like).
Bus 1050 includes a path that transfers information between various components of the device, such as processor 1010, memory 1020, input/output interface 1030, and communication interface 1040.
It should be noted that although the above-mentioned device only shows the processor 1010, the memory 1020, the input/output interface 1030, the communication interface 1040 and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, those skilled in the art will appreciate that the above-described apparatus may also include only those components necessary to implement the embodiments of the present description, and not necessarily all of the components shown in the figures.
The electronic device of the above embodiment is used to implement the method for predicting the remaining capacity of the battery in any of the foregoing embodiments in a corresponding manner, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; within the idea of the present disclosure, also technical features in the above embodiments or in different embodiments may be combined, steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present disclosure as described above, which are not provided in detail for the sake of brevity.
In addition, well-known power/ground connections to Integrated Circuit (IC) chips and other components may or may not be shown in the provided figures for simplicity of illustration and discussion, and so as not to obscure the embodiments of the disclosure. Furthermore, devices may be shown in block diagram form in order to avoid obscuring embodiments of the present disclosure, and this also takes into account the fact that specifics with respect to implementation of such block diagram devices are highly dependent upon the platform within which the embodiments of the present disclosure are to be implemented (i.e., specifics should be well within purview of one skilled in the art). Where specific details (e.g., circuits) are set forth in order to describe example embodiments of the disclosure, it should be apparent to one skilled in the art that the embodiments of the disclosure can be practiced without, or with variation of, these specific details. Accordingly, the description is to be regarded as illustrative instead of restrictive.
While the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory architectures (e.g., dynamic ram (dram)) may use the discussed embodiments.
The disclosed embodiments are intended to embrace all such alternatives, modifications and variances which fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalents, improvements, and the like that may be made within the spirit and principles of the embodiments of the disclosure are intended to be included within the scope of the disclosure.

Claims (10)

1. A prediction method for residual capacity of a battery in echelon comprises the following steps:
obtaining online use data of the echelon battery, and generating a feature vector of the online use data according to the online use data of the echelon battery;
and obtaining the residual capacity of the echelon battery at the stable voltage starting point according to the feature vector of the on-line use data and a pre-trained residual capacity prediction network model.
2. The method according to claim 1, wherein the process of obtaining the online usage data of the echelon battery and generating the feature vector of the online usage data according to the online usage data of the echelon battery specifically comprises:
acquiring a current value and a voltage value of the online use data of the echelon battery;
respectively obtaining a current curve I '(t) and a voltage curve V' (t) according to the current value and the voltage value;
obtaining a discharged electricity quantity curve Q ' according to the current curve I ' (t) 'dis(t)；
According to the voltage curve V '(t) and the discharged electricity quantity curve Q'dis(t) obtaining a stable voltage starting point V'0Time t'0And discharged electric quantity Q'dis0；
According to the discharged electric quantity Q'dis0And a unit discharged electric quantity delta Q 'to obtain a discharge quantity Q'dis0Several electric quantity values Q 'inside'diss；
According to a plurality of the electric quantity values Q'dissAnd the amount of discharged electricityCurve Q'dis(t) obtaining each of said electric quantity values Q'dissCorresponding time t's；
According to the time t'sAnd the voltage curve V' (t) obtains a voltage value V corresponding to each times′；
Corresponding voltage value V to each times' as a feature vector of said on-net usage data.
3. The method of claim 1, wherein the training process of the residual capacity prediction network model comprises:
establishing a Back Propagation Neural Network (BPNN);
obtaining a plurality of groups of ex-factory discharge test data of training echelon batteries;
taking a part of the factory discharge test data of the plurality of groups of training echelon batteries as a training set; the training set includes: a training current value I, a training voltage value V and time t;
obtaining a plurality of groups of stable voltage initial points V for training according to the voltage value V for training0And a plurality of groups of discharge inflection points n for training;
according to the plurality of groups of stable voltage initial points V for training0Obtaining a plurality of groups of training voltage values V by the training current value I and the unit discharge quantity delta Qs(ii) the sequence of (a);
according to the plurality of groups of stable voltage initial points V for training0And the plurality of groups of discharge inflection points n for training are obtained to obtain a plurality of groups of discharged electric quantities Q for training;
according to the plurality of groups of voltage values V for trainingsAnd training a Back Propagation Neural Network (BPNN) by using the formed sequence and a plurality of groups of the training discharged electric quantity Q to obtain the residual capacity prediction network.
4. The method of claim 2, wherein the current curve I '(t) is calculated by interpolation fitting of the current values and the voltage curve V' (t) is calculated by interpolation fitting of the voltage values.
5. The method of claim 3, wherein the residual capacity prediction network model consists of a 4-layer Dense fully-connected network layer and a Relu activation layer.
6. The method of claim 3, wherein the stable voltage starting point V0Is the first point of the voltage value V, wherein V is less than or equal to 3.3 volts.
7. The method of claim 3, wherein the discharge inflection point n is calculated by:
according to the voltage value V, making difference delta VtV (t +1) -V (t), where t is time; when at a certain time t, if Δ V is satisfied at the same timetLess than-0.03V and delta Vt+1Less than-0.03V and delta Vt+2The time t at this time is obtained as the discharge inflection point n under the condition of < -0.03V.
8. The method of claim 3, wherein the training voltage value VsThe way the composed sequence is calculated is:
obtaining a voltage acquisition time interval m according to the current value I and the unit discharge quantity delta Q; the time interval m of the collected voltage is delta Q/I;
according to the acquisition voltage time interval m and the stable voltage starting point V for training0Obtaining the training voltage value VsThe sequence of the composition.
9. The method of claim 3, wherein the quantity of electricity Q is the regulated voltage starting point V0And the electric quantity of the discharge inflection point n.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1 to 9 when executing the program.
CN202110964990.0A 2021-08-20 2021-08-20 Method for predicting remaining capacity of battery in echelon and electronic equipment Active CN113820608B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110964990.0A CN113820608B (en) 2021-08-20 2021-08-20 Method for predicting remaining capacity of battery in echelon and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110964990.0A CN113820608B (en) 2021-08-20 2021-08-20 Method for predicting remaining capacity of battery in echelon and electronic equipment

Publications (2)

Publication Number Publication Date
CN113820608A true CN113820608A (en) 2021-12-21
CN113820608B CN113820608B (en) 2022-08-05

Family

ID=78913389

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110964990.0A Active CN113820608B (en) 2021-08-20 2021-08-20 Method for predicting remaining capacity of battery in echelon and electronic equipment

Country Status (1)

Country Link
CN (1) CN113820608B (en)

Citations (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103389471A (en) * 2013-07-25 2013-11-13 哈尔滨工业大学 Cycle life indirect prediction method for lithium ion battery provided with uncertain intervals on basis of GPR (general purpose register)
US8855954B1 (en) * 2007-11-30 2014-10-07 Intellectual Assets Llc System and method for prognosticating capacity life and cycle life of a battery asset
FR3015123A1 (en) * 2013-12-13 2015-06-19 Renault Sa METHOD OF ESTIMATING THE RESIDUAL CAPACITIES OF A PLURALITY OF BATTERIES
JP2016053564A (en) * 2014-09-01 2016-04-14 横河電機株式会社 Secondary battery capacity measurement system and secondary battery capacity measurement method
CN107037373A (en) * 2017-05-03 2017-08-11 广西大学 Battery dump energy Forecasting Methodology based on neutral net
CN107085187A (en) * 2017-04-13 2017-08-22 华北电力科学研究院有限责任公司 Echelon utilizes the determination method and device of battery energy storage system consistency maintenance index
CN108680861A (en) * 2018-03-05 2018-10-19 北京航空航天大学 The construction method and device of lithium battery cycles left Life Prediction Model
CN109543317A (en) * 2018-04-28 2019-03-29 北京航空航天大学 A kind of method and device of PEMFC remaining life prediction
US20190257886A1 (en) * 2018-02-21 2019-08-22 Nec Laboratories America, Inc. Deep learning approach for battery aging model
CN110161425A (en) * 2019-05-20 2019-08-23 华中科技大学 A kind of prediction technique of the remaining life divided based on lithium battery catagen phase
CN110187287A (en) * 2019-06-24 2019-08-30 安徽师范大学 A kind of retired lithium battery complementary energy rapid detection method
CN110224192A (en) * 2019-05-30 2019-09-10 安徽巡鹰新能源科技有限公司 A kind of echelon utilizes power battery life-span prediction method
US20200011932A1 (en) * 2018-07-05 2020-01-09 Nec Laboratories America, Inc. Battery capacity fading model using deep learning
CN110824375A (en) * 2019-10-18 2020-02-21 广东电网有限责任公司 Method for calculating actual capacity of retired storage battery
CN110941929A (en) * 2019-12-06 2020-03-31 长沙理工大学 Battery health state assessment method based on ARMA and Elman neural network combined modeling
US20200153264A1 (en) * 2017-07-25 2020-05-14 Semiconductor Energy Laboratory Co., Ltd. Power storage system, electronic device, vehicle, and estimation method
CN111537888A (en) * 2020-05-09 2020-08-14 国网福建省电力有限公司莆田供电公司 Data-driven echelon battery SOC prediction method
US20200301408A1 (en) * 2017-05-25 2020-09-24 Johnson Controls Technology Company Model predictive maintenance system with degradation impact model
CN111866932A (en) * 2020-08-07 2020-10-30 北京邮电大学 Network measurement method and device and electronic equipment
CN112098848A (en) * 2020-09-14 2020-12-18 北京大学深圳研究生院 Battery discharge capacity prediction method and system and readable storage medium
CN112241608A (en) * 2020-10-13 2021-01-19 国网湖北省电力有限公司电力科学研究院 Lithium battery life prediction method based on LSTM network and transfer learning
CN112269136A (en) * 2020-10-13 2021-01-26 深圳市智达百川科技有限公司 Internet of things battery electric quantity measuring method, device, system, equipment and storage medium
US20210050631A1 (en) * 2019-08-13 2021-02-18 International Business Machines Corporation Capacity degradation analysis for batteries
CN112578296A (en) * 2019-09-27 2021-03-30 比亚迪股份有限公司 Battery capacity estimation method and apparatus, and computer storage medium
CN113065283A (en) * 2021-03-26 2021-07-02 深圳技术大学 Battery life prediction method, system, electronic device and storage medium
CN113093020A (en) * 2021-04-02 2021-07-09 中国矿业大学 Method for predicting remaining service life of lithium ion battery based on LSTM neural network
CN113240194A (en) * 2021-06-04 2021-08-10 河北工业大学 Energy storage battery capacity prediction method, server and computer readable storage medium

Patent Citations (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8855954B1 (en) * 2007-11-30 2014-10-07 Intellectual Assets Llc System and method for prognosticating capacity life and cycle life of a battery asset
CN103389471A (en) * 2013-07-25 2013-11-13 哈尔滨工业大学 Cycle life indirect prediction method for lithium ion battery provided with uncertain intervals on basis of GPR (general purpose register)
FR3015123A1 (en) * 2013-12-13 2015-06-19 Renault Sa METHOD OF ESTIMATING THE RESIDUAL CAPACITIES OF A PLURALITY OF BATTERIES
JP2016053564A (en) * 2014-09-01 2016-04-14 横河電機株式会社 Secondary battery capacity measurement system and secondary battery capacity measurement method
CN107085187A (en) * 2017-04-13 2017-08-22 华北电力科学研究院有限责任公司 Echelon utilizes the determination method and device of battery energy storage system consistency maintenance index
CN107037373A (en) * 2017-05-03 2017-08-11 广西大学 Battery dump energy Forecasting Methodology based on neutral net
US20200301408A1 (en) * 2017-05-25 2020-09-24 Johnson Controls Technology Company Model predictive maintenance system with degradation impact model
US20200153264A1 (en) * 2017-07-25 2020-05-14 Semiconductor Energy Laboratory Co., Ltd. Power storage system, electronic device, vehicle, and estimation method
US20190257886A1 (en) * 2018-02-21 2019-08-22 Nec Laboratories America, Inc. Deep learning approach for battery aging model
CN108680861A (en) * 2018-03-05 2018-10-19 北京航空航天大学 The construction method and device of lithium battery cycles left Life Prediction Model
CN109543317A (en) * 2018-04-28 2019-03-29 北京航空航天大学 A kind of method and device of PEMFC remaining life prediction
US20200011932A1 (en) * 2018-07-05 2020-01-09 Nec Laboratories America, Inc. Battery capacity fading model using deep learning
CN110161425A (en) * 2019-05-20 2019-08-23 华中科技大学 A kind of prediction technique of the remaining life divided based on lithium battery catagen phase
CN110224192A (en) * 2019-05-30 2019-09-10 安徽巡鹰新能源科技有限公司 A kind of echelon utilizes power battery life-span prediction method
CN110187287A (en) * 2019-06-24 2019-08-30 安徽师范大学 A kind of retired lithium battery complementary energy rapid detection method
US20210050631A1 (en) * 2019-08-13 2021-02-18 International Business Machines Corporation Capacity degradation analysis for batteries
CN112578296A (en) * 2019-09-27 2021-03-30 比亚迪股份有限公司 Battery capacity estimation method and apparatus, and computer storage medium
CN110824375A (en) * 2019-10-18 2020-02-21 广东电网有限责任公司 Method for calculating actual capacity of retired storage battery
CN110941929A (en) * 2019-12-06 2020-03-31 长沙理工大学 Battery health state assessment method based on ARMA and Elman neural network combined modeling
CN111537888A (en) * 2020-05-09 2020-08-14 国网福建省电力有限公司莆田供电公司 Data-driven echelon battery SOC prediction method
CN111866932A (en) * 2020-08-07 2020-10-30 北京邮电大学 Network measurement method and device and electronic equipment
CN112098848A (en) * 2020-09-14 2020-12-18 北京大学深圳研究生院 Battery discharge capacity prediction method and system and readable storage medium
CN112269136A (en) * 2020-10-13 2021-01-26 深圳市智达百川科技有限公司 Internet of things battery electric quantity measuring method, device, system, equipment and storage medium
CN112241608A (en) * 2020-10-13 2021-01-19 国网湖北省电力有限公司电力科学研究院 Lithium battery life prediction method based on LSTM network and transfer learning
CN113065283A (en) * 2021-03-26 2021-07-02 深圳技术大学 Battery life prediction method, system, electronic device and storage medium
CN113093020A (en) * 2021-04-02 2021-07-09 中国矿业大学 Method for predicting remaining service life of lithium ion battery based on LSTM neural network
CN113240194A (en) * 2021-06-04 2021-08-10 河北工业大学 Energy storage battery capacity prediction method, server and computer readable storage medium

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
NING YAN等: "Sorting Method for Retired Power Battery Module Considering Consistent Static and Dynamic Characteristics", 《 2020 IEEE INTERNATIONAL CONFERENCE ON APPLIED SUPERCONDUCTIVITY AND ELECTROMAGNETIC DEVICES (ASEMD)》 *
李长东等: "新能源动力电池放电剩余容量性能试验研究", 《能源研究与管理》 *
杨若岑等: "电池剩余容量估算方法综述", 《电气技术》 *
解坤等: "基于退役电池梯次利用的源-储容量优化配置", 《储能科学与技术》 *

Also Published As

Publication number Publication date
CN113820608B (en) 2022-08-05

Similar Documents

Publication Publication Date Title
Park et al. LSTM-based battery remaining useful life prediction with multi-channel charging profiles
EP3542173B1 (en) Determining a state of health of a battery and providing an alert
US20160239592A1 (en) Data-driven battery aging model using statistical analysis and artificial intelligence
JP7199568B2 (en) BATTERY DIAGNOSTIC SYSTEM, BATTERY DIAGNOSTIC METHOD, AND STORAGE MEDIUM
US20200150185A1 (en) Method for approximating algorithms for fast charging li-ion batteries based on electrochemical battery models
CN101809456B (en) System and method for estimating long term characteristics of battery
US10830821B2 (en) Methods and apparatus for battery power and energy availability prediction
CN101803104B (en) System and method for estimating long term characteristics of battery
Chao et al. State-of-health estimator based-on extension theory with a learning mechanism for lead-acid batteries
Cui et al. State of health diagnosis and remaining useful life prediction for lithium-ion battery based on data model fusion method
Mian Qaisar A proficient Li-ion battery state of charge estimation based on event-driven processing
CN109462260A (en) A kind of charging method, charging equipment and electronic equipment
CN116757378A (en) Demand response evaluation model training method, application evaluation method and related equipment
CN109075397A (en) The method and apparatus of electrochemical energy accumulator is optimally used for the service life
CN113820608B (en) Method for predicting remaining capacity of battery in echelon and electronic equipment
CN107729144B (en) Application control method and device, storage medium and electronic equipment
CN106546922A (en) Method and apparatus for determining battery parameter
CN113740755A (en) Display processing method and device for battery pack charge state and readable medium
CN115792665A (en) Method and device for determining residual capacity of battery, electronic equipment and storage medium
CN115902625A (en) Performance prediction method, device, equipment and storage medium of battery system
CN115656834A (en) Battery capacity prediction method and device and electronic equipment
CN114740389A (en) Battery health assessment method and device, electronic equipment and readable storage medium
KR20190080102A (en) A method of estimating state of charge of battery and an apparatus for managing of battery
CN107506275A (en) A kind of software protecting energy consumption evaluation method based on ARM instruction virtualization
CN109254201B (en) Electrical parameter metering method and device, charging equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant