CN115407224A - Power battery full life cycle health monitoring method and system and electronic equipment - Google Patents
Power battery full life cycle health monitoring method and system and electronic equipment Download PDFInfo
- Publication number
- CN115407224A CN115407224A CN202211174113.4A CN202211174113A CN115407224A CN 115407224 A CN115407224 A CN 115407224A CN 202211174113 A CN202211174113 A CN 202211174113A CN 115407224 A CN115407224 A CN 115407224A
- Authority
- CN
- China
- Prior art keywords
- battery
- capacity
- interval
- diving
- potential
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- 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
-
- 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]
- G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/4285—Testing apparatus
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- General Chemical & Material Sciences (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention provides a method, a system and electronic equipment for monitoring the health of a power battery in a full life cycle, wherein the full life cycle of the battery is divided into three stages, namely a degradation stage, a potential diving interval and a key stage, and different monitoring methods are adopted according to the stage of the battery: in the degradation stage, battery capacity advanced multi-step forecasting is carried out based on a battery historical capacity sequence; in the potential diving interval, carrying out battery capacity advanced prediction based on a capacity sequence in the potential diving interval, wherein the number of advanced prediction steps is larger than that of the degradation stage; and in the key stage, predicting the residual life in the whole process. According to the method, the full life cycle of the battery is divided into a plurality of stages based on the battery capacity degradation trend, then different tasks are allocated to each stage, each stage of the battery can obtain a high-reliability health state representation, each stage of the battery is managed more carefully and pertinently, and the defect that the residual life of the battery is inaccurate in early and middle periods is overcome.
Description
Technical Field
The invention belongs to the technical field of new energy automobile power batteries, and particularly relates to a health management technology of a power battery.
Background
With the gradual improvement of the permeability of new energy vehicles, new energy vehicle accidents caused by battery faults are more and more caused in the market, and the health management of the power battery of the new energy vehicle in the whole life cycle is also more and more widely concerned. An important aspect of battery health management is Remaining Life (RUL) prediction. IEEE standard 1188-1996 states that a battery should be replaced when its capacity drops to 80% of the factory capacity. Therefore, the remaining life of the battery generally refers to the time from the present time when the battery can be used before the capacity is reduced to 80% of the factory capacity. This is the full life cycle of the battery at the vehicle end, which is the power battery full life cycle referred to below in the present invention.
Most of the current literature concerns about the health management of batteries are also battery remaining life predictions based on capacity, such as: patent documents CN202110206262.3 (a method for predicting the remaining life of a lithium battery based on the multi-core GPR algorithm of MFF), CN201910238129.9 (a method for predicting the remaining life of a lithium battery based on a gray wolf group optimized LSTM network), and the like. The residual life prediction plays a crucial role in health management in the late stage of battery degradation, but the full life cycle health management of the battery is difficult to support by only depending on the residual life prediction. This is because the degradation trend of the battery system over the life cycle is generally time-varying, and it is difficult to describe the time-varying degradation trend by a single model or a single means. During the early use of the power battery, the degradation is also slow; however, when the battery capacity is degraded to a certain extent, the battery is degraded at an accelerated speed, a phenomenon of battery capacity 'water jump' occurs, and the battery capacity is degraded to the battery failure in a short time, as shown in fig. 1. In practical application, due to the fact that the water jump inflection point is difficult to predict, the residual life prediction results of the power battery in the early stage and the middle stage often have large deviation with the real residual life value, and the prediction results are difficult to support issuing of a vehicle maintenance strategy. Therefore, how to set a strategy to enable the health state representation with high reliability to be obtained at each stage of the battery has important significance on the full life cycle health management of the battery.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a battery full-life-cycle health monitoring method, electronic equipment and an automobile, wherein the battery is subjected to full-life-cycle health monitoring on a time-varying system, the full life cycle of the battery is divided into a plurality of stages based on the battery capacity degradation trend, then different tasks are allocated to each stage, each stage of the battery can obtain a high-reliability health state representation, more detailed and more targeted management on each stage of the battery is realized, and the defect that the residual life of the battery is inaccurate in early and middle periods is overcome.
The technical scheme of the invention is as follows:
the invention provides a method for monitoring the health of a power battery in a full life cycle, which comprises the steps of dividing the full life cycle of the battery into three stages based on the degradation trend of the battery capacity and assigning different tasks to each stage aiming at a time-varying system of the battery, ensuring that each stage of the battery can obtain a high-reliability health state representation, and realizing the full life cycle health monitoring of the battery. On the other hand, aiming at the problem that the battery capacity jumping point is difficult to predict in advance, a clustering model is established based on the user characteristics of the battery, the jumping points of the cloud similar batteries are counted and used as the prior estimation of the current battery jumping point, and the potential jumping interval of the current battery capacity is obtained.
The method specifically comprises the steps that the full life cycle of the battery is divided into a degradation stage, a potential diving interval and a key stage in sequence, and different management methods are adopted according to the stage of the battery:
in the degradation stage, battery capacity advanced multi-step forecasting is carried out based on a battery historical capacity sequence;
and in the potential diving interval, performing battery capacity advanced prediction based on a capacity sequence in the potential diving interval, wherein the number of advanced prediction steps is gradually increased, the prediction is gradually transited to the residual life prediction, and the advanced prediction step is connected with the third stage.
And in the key stage, predicting the residual life in the whole process.
Further, the potential diving interval is obtained by counting diving points of the cloud-end like battery as prior estimation of the current battery diving point.
Specifically, in an embodiment of the present invention, the obtaining of the potential diving interval is: firstly, extracting the characteristics of the driving speed, driving style, electric appliance information, a common SOC driving interval and a common SOC charging and discharging interval of a user from the driving behavior and charging habit of the user to which the battery belongs in a cloud-end similar battery, and establishing a clustering model for the extracted characteristics; extracting the features from the current battery, and predicting the category of the user through the clustering model; and then counting the distribution of the number of charge-discharge cycles when the capacity of the battery in the category jumps, and acquiring a confidence interval of the distribution as a potential diving interval of the current battery.
Further, in the degradation stage, outliers are removed from the battery historical capacity sequence, then a state transition equation and an observation equation are established based on the battery historical capacity sequence, and finally battery capacity advanced multi-step forecasting is carried out.
Further, in the potential diving interval, the outlier removal is performed on the battery potential diving interval capacity sequence, then a state transfer equation and an observation equation are established based on the potential diving interval battery capacity sequence, and finally the battery capacity advanced multi-step prediction is performed.
Further, in the key stage, firstly, the potential diving interval and the key stage capacity sequence of the battery are subjected to outlier removal, then a state transition equation and an observation equation are established based on the potential diving interval and the key stage battery capacity sequence, and finally, the whole-process residual life is predicted according to the following formula:
rul k ＝inf(l:f(l+k)≥γ)
in the formula, inf (l) representsLower limit of number of charge-discharge cycles l; γ represents the failure threshold of the discharge capacity, here 80% of the initial capacity of the battery; rul k Indicating the remaining life of the battery at the kth charge-discharge cycle.
The invention provides a power battery full-life cycle health monitoring system in a second aspect, which comprises three management modules of a battery full-life cycle, namely a degradation stage management module, a potential diving interval management module and a key stage management module, wherein the three modules adopt different management methods according to the stage of the battery.
And the degradation stage management module is used for performing battery capacity advanced multi-step forecasting based on the battery historical capacity sequence in the degradation stage.
And the potential diving interval management module is used for carrying out battery capacity advanced prediction in the potential diving interval based on the capacity sequence in the potential diving interval, and the advanced prediction step number is greater than the step number of the degradation stage.
And the key stage management module is used for predicting the residual life in the whole key stage.
The present invention provides, in a third aspect, an electronic device comprising:
one or more processors;
a storage device for storing one or more programs which, when executed by the one or more processors, cause the electronic device to implement the power battery full-life cycle health monitoring method of the first aspect.
By adopting the technical scheme, the invention at least has the following advantages:
1. the method divides the whole life cycle of the battery into a plurality of stages, assigns different tasks for each stage, performs battery capacity advanced multi-step forecasting based on a battery historical capacity sequence in the degradation stage, performs battery capacity advanced forecasting which is gradually improved based on the capacity sequence in a potential diving interval in the potential diving interval, and performs residual life forecasting in the whole course in the key stage. Compared with the existing schemes such as pure residual life prediction and the like, the scheme provided by the invention has more detailed and targeted health monitoring method for each stage of the battery, is easier to realize the full life cycle health monitoring of the battery, and makes up the defect of inaccurate prediction of the residual life of the battery in early and middle stages.
2. In the invention, on the acquisition of the battery capacity jump point, the potential jump interval of the current battery is obtained by counting the cloud battery jump points of the same category as the prior estimation of the current battery jump point, and compared with the scheme of obtaining the jump point single-point prediction result by mechanism analysis in the prior art, the method gets rid of complicated battery mechanism analysis.
3. The invention can fully utilize national standard and enterprise standard data uploaded to the cloud by vehicles, and is easy to deploy and implement at the cloud.
Drawings
FIG. 1 is a schematic diagram of a capacity diving inflection point of a power cell;
FIG. 2 is a schematic diagram of the present invention dividing the full life cycle of a battery into three phases;
fig. 3 is a flow chart of the power battery full life cycle health monitoring method of the present invention.
It should be noted that the above drawings are for better understanding of the present solution and do not constitute a limitation of the present application.
Detailed Description
The embodiments of the present application will be described with reference to specific examples, and other advantages and effects of the present application will be apparent to those skilled in the art from the disclosure of the present application. The application is capable of other and different embodiments and its several details are capable of modifications and various changes in detail without departing from the spirit of the application. It should be noted that, in the following embodiments, features, steps, and the like in the embodiments may be combined with each other without conflict.
As shown in fig. 2, in the present invention, the full-life cycle health monitoring of the battery divides the full-life cycle of the battery into three stages, namely, a degradation stage, a potential diving interval and a key stage.
Example 1:
the present embodiment is a method for monitoring the health of a battery in a full life cycle based on the above three phases, and the flow chart is shown in fig. 3.
Firstly, after the current charge and discharge cycle of the power battery is completed, whether the current charge and discharge cycle enters a potential diving interval is judged.
Since the power battery is in the degradation process, the capacity water jump point is difficult to predict in advance. The method provided by the invention creatively proposes to divide the potential diving interval, and obtains the potential diving interval by counting the diving points of the similar batteries at the cloud as the prior estimation of the current battery diving points.
In an embodiment of the present invention, the potential diving interval may be obtained by adopting the following calculation procedure:
1. firstly, in a cloud-like battery, the characteristics of the driving speed, the driving style, the electric appliance information, the common SOC driving interval, the common SOC charging and discharging interval and the like of a user who the battery belongs to are extracted from the driving behavior and the charging habit of the user, and the extracted characteristics are clustered based on a Gaussian mixture model to obtain a clustering model.
2. And then extracting the characteristics from the current battery, and predicting the category to which the user belongs through the established clustering model.
3. And finally, counting the distribution of the charging and discharging cycle number of the batteries of the same category at the cloud during capacity diving, and acquiring a 95% confidence interval of the distribution, namely the distribution can be used as a potential diving interval of the current battery.
The potential diving interval is obtained by utilizing data of the cloud-end similar battery and obtaining a priori estimation interval of the current battery diving point in a data driving mode instead of a single-point prediction result, and the reliability and the fault tolerance of the diving point generating point can be improved through the expression of uncertainty. Moreover, national standard and enterprise standard data uploaded to the cloud by vehicles can be fully utilized, and the cloud deployment and implementation are easy.
And when the current charge-discharge cycle is judged not to be in the potential diving interval and is smaller than the lower limit of the potential diving interval, the current battery is in a degradation stage, management is carried out according to a degradation stage method, and battery capacity advanced multi-step prediction is carried out based on the battery historical capacity sequence.
And when the current charge-discharge cycle is judged to be larger than the lower limit of the potential diving interval, whether the current charge-discharge cycle enters a key stage or not is further judged, namely whether the current charge-discharge cycle exceeds the upper limit of the potential diving interval or not is judged, if the current charge-discharge cycle does not enter the key stage, management is carried out according to a method of the potential diving interval, advanced prediction of the battery capacity is carried out based on a capacity sequence in the potential diving interval, and the number of advanced prediction steps is larger than that of the degradation stage.
If the upper limit is exceeded, the critical stage is entered, and the residual life prediction is carried out in the whole process.
1. A degradation stage: for a brand new battery, when the battery starts to run, the system enters a stable working state, and the time before the battery is subjected to capacity diving is a degradation stage. Since the accelerated degradation trend of the water jump point and later of the capacity cannot be predicted at this stage, and the prediction error is gradually accumulated as the number of advance prediction steps increases, and the deviation between the predicted value and the actual capacity value is gradually increased, it is difficult to accurately predict the remaining life at this stage. But according to the characteristic that the battery capacity recession does not generate mutation, medium and short term capacity prediction based on the existing degradation trend is relatively reliable; in addition, the early battery degradation is far from the end of life, and the medium-short term forecast of capacity can provide enough health information to ensure that the degradation condition of the battery is controllable. In conclusion, the task of the degradation stage setting is mainly to perform advanced multi-step prediction of the battery capacity.
In an embodiment of the present invention, the battery capacity advanced prediction process in the degradation stage is as follows:
1. and preprocessing the battery historical capacity sequence, mainly removing outliers.
2. Establishing a state transition equation and an observation equation according to an empirical model of the battery degradation process, as follows:
wherein, w k Representing process noise, v k Representing the observed noise, k =1,2, … represents the cycle period of battery charging and discharging.
3. And setting the current charge-discharge cycle k as a prediction starting point, and performing state tracking on the historical capacity sequence by using a particle filter algorithm so as to determine unknown parameters in a state transition equation.
4. Initializing a particle filtering algorithm, wherein parameters comprise: number of particles N, process noise w k Covariance R of (1), observation noise v k The covariance Q of (a).
5. And carrying out advanced prediction of the capacity based on a particle filter algorithm.
6. Setting the number of advanced forecasting steps as m, and sequentially and repeatedly executing the steps to finally obtain a capacity forecasting value C of m advanced steps k+1 、C k+2 、…、C k+m 。
2. Potential diving interval:
the task set by the potential diving interval is also advanced multi-step forecasting of the battery capacity. Since the capacity degradation trend of the battery after entering the potential diving interval is accelerated and the degradation trend of the capacity sequence in the degradation stage has no reference significance, the capacity advanced prediction is carried out only on the basis of the capacity sequence in the potential diving interval. In addition, the battery enters the potential water jump interval, which means that the battery is closer to the end point of the service life of the battery, so that the forecast steps are gradually increased in advance, and the prediction is gradually transited to the residual service life prediction.
In an embodiment of the present invention, the battery capacity advanced prediction process in the potential diving interval is as follows:
1. and (4) preprocessing the capacity sequence of the potential water-jumping interval of the battery, and mainly removing outliers.
2. Establishing a state transition equation and an observation equation according to an empirical model of the battery degradation process, as follows:
wherein, w k Representing process noise, v k Representing the observed noise, k =1,2, … represents the cycle period of battery charging and discharging.
3. And setting the current charge-discharge cycle k as a prediction starting point, and performing state tracking on the historical capacity sequence by using a particle filter algorithm so as to determine unknown parameters in a state transition equation.
4. Initializing a particle filtering algorithm, wherein parameters comprise: number of particles N, process noise w k Covariance R of (1), observation noise v k The covariance Q of (a).
5. And carrying out advanced prediction of the capacity based on a particle filter algorithm.
6. Setting the initial advanced forecasting step number as m 0 The steps are repeatedly executed in sequence to obtain
7. After the current charge-discharge cycle enters k +1 from k, the advanced prediction step number is changed into m 0 And +1, sequentially and repeatedly executing the steps, then increasing the current charge-discharge cycle, and gradually forecasting the step number by +1. Here, the initial advance forecasting step number m0 is not necessarily set to be a positive integer, for example, may be set to be-10, and then the forecasting step number is gradually increased, which means that forecasting is started after the potential diving interval goes through more than 10 discharge cycles, because the sequence is too short to see the trend, and the forecasting is inaccurate.
3. The key stage is as follows:
after the battery enters the stage, namely the water jump point of the battery capacity is crossed, the capacity degradation trend is greatly accelerated and reaches the failure threshold value in a short time. Since the degradation trend is substantially fixed after entering this phase, the remaining life prediction should be made throughout this phase.
The residual life prediction rule provided by the invention is as follows:
rul k ＝inf(l:f(l+k)≥γ)
wherein inf (l) represents the lower limit of the charge-discharge cycle number l; γ represents the failure threshold of the discharge capacity, here 80% of the initial capacity of the battery; rul k The remaining life (unit: number of charge/discharge cycles) of the battery at the kth charge/discharge cycle is shown.
In an embodiment of the present invention, the process of predicting the remaining battery life at the critical stage is as follows:
1. when the current charge and discharge cycle of the battery enters a key stage, first Predicting Time (FPT) is determined.
2. And (4) preprocessing the capacity sequence of the potential diving interval and the key stage of the battery, and mainly removing outliers.
3. Establishing a state transition equation and an observation equation according to an empirical model of the battery degradation process, as follows:
wherein, w k Representing process noise, v k Representing the observed noise, k =1,2, … represents the cycle period of battery charging and discharging.
4. And setting the current charge-discharge cycle k as a prediction starting point, and performing state tracking on the battery potential diving interval and the historical capacity sequence of the key stage by using a particle filter algorithm, thereby determining unknown parameters in a state transition equation.
5. Initializing a particle filtering algorithm, wherein parameters comprise: number of particles N, process noise w k Covariance R of (1), observation noise v k The covariance Q of (a).
6. And carrying out advanced prediction of the capacity based on a particle filter algorithm.
7. Determining a capacity advance prediction value C k+1 Whether the battery's failure threshold, i.e., 80% of the initial capacity, has been reached. If the threshold value is not reached, the above steps are repeated in sequence, and when the predicted value C is reached k+n And if the residual service life of the battery exceeds the battery failure threshold value, the residual service life of the battery under the current charge-discharge cycle is n-1 charge-discharge cycles.
In an embodiment of the present invention, the particle-filter-based algorithm adopted in each of the above stages for capacity advanced prediction may adopt the following method, and the specific steps include:
a) Updating the particles, N particles based on discharge cycle k
Obtaining the prior values of N particles of the discharge cycle k +1 through updating of a state transition equation
b) Calculating weight set of N particle prior values of discharge cycle k +1 by combining measurement equation and likelihood function
Normalizing the particle weights results in a set of weights
c) Utilizing weight sets
Resampling the particles to obtain a resampled particle set
d) Calculating a state estimate of battery capacity
And recorded into the sequence.
Example 2:
the embodiment is a system for monitoring the full life cycle health of a power battery, and is used for implementing the method described in the above embodiment. The system comprises three management modules of the full life cycle of the battery, namely a degradation stage management module, a potential diving interval management module and a key stage management module, wherein the three modules adopt different management methods according to the stage of the battery.
And the degradation stage management module is used for performing battery capacity advanced multi-step forecasting based on the battery historical capacity sequence in the degradation stage. The method specifically comprises the steps of firstly removing outliers from a battery historical capacity sequence, then establishing a state transfer equation and an observation equation based on the battery historical capacity sequence, and finally performing battery capacity advanced multi-step forecasting.
And the potential diving interval management module is used for carrying out battery capacity advanced prediction in the potential diving interval based on the capacity sequence in the potential diving interval, and the advanced prediction step number is greater than the step number of the degradation stage. Specifically, outlier removal is performed on a battery potential diving interval capacity sequence, then a state transfer equation and an observation equation are established based on the battery capacity sequence of the potential diving interval, and finally battery capacity advanced multi-step prediction is performed.
And the key stage management module is used for predicting the residual life in the whole key stage. Specifically, outlier removal is performed on a potential diving interval and a key stage capacity sequence of the battery, then a state transfer equation and an observation equation are established based on the potential diving interval and the key stage capacity sequence of the battery, and finally the whole-course residual life is predicted according to the following formula:
rul k ＝inf(l:f(l+k)≥γ)
wherein inf (l) represents the lower limit of the charge-discharge cycle number l; γ represents the failure threshold of the discharge capacity, here 80% of the initial capacity of the battery; rul k Indicating the remaining life of the battery at the kth charge-discharge cycle.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should be construed as the protection scope of the present invention.
Claims (16)
1. The method for monitoring the health of the full life cycle of the power battery is characterized in that the full life cycle of the battery is divided into three stages, namely a degradation stage, a potential diving interval and a key stage, and different processing methods are adopted according to the stage of the battery:
in the degradation stage, battery capacity advanced multi-step forecasting is carried out based on the historical capacity sequence of the current battery;
in the potential diving interval, carrying out battery capacity advanced prediction based on a capacity sequence in the potential diving interval, wherein the advanced prediction step number is gradually increased and gradually transits to the residual life prediction;
and in the key stage, predicting the residual life in the whole process.
2. The power battery full-life cycle health monitoring method according to claim 1, wherein the potential diving interval is obtained by counting diving point data of a cloud-like battery as a prior estimate of a current battery diving point.
3. The power battery full-life cycle health monitoring method of claim 2, wherein the potential diving interval is obtained by: firstly, extracting the characteristics of the driving speed, driving style, electric appliance information, a common SOC driving interval and a common SOC charging and discharging interval of a user from the driving behavior and charging habit of the user to which the battery belongs in a cloud-end similar battery, and establishing a clustering model for the extracted characteristics; extracting the features from the current battery, and predicting the category of the user through the clustering model; and then counting the distribution of the number of charge-discharge cycles when the capacity of the battery in the category jumps, and acquiring a confidence interval of the distribution as a potential diving interval of the current battery.
4. The method for monitoring the full-life-cycle health of the power battery as claimed in claim 1,2 or 3, wherein in the degradation stage, the battery historical capacity sequence is firstly subjected to outlier removal, then a state transition equation and an observation equation are established based on the battery historical capacity sequence, and finally battery capacity advanced multi-step forecasting is carried out.
5. The power battery full-life cycle health monitoring method according to claim 4, wherein performing battery capacity advanced prediction in the degradation stage comprises:
(1) Preprocessing a battery historical capacity sequence to remove outliers;
(2) Establishing a state transition equation and an observation equation according to an empirical model of the battery degradation process, wherein the state transition equation and the observation equation are as follows:
wherein, w k Representing process noise, v k Representing observed noise, k =1,2, … represents the cycle period of battery charging and discharging;
(3) Setting the current charge-discharge cycle k as a prediction starting point, and performing state tracking on the historical capacity sequence by using a particle filter algorithm to determine unknown parameters in a state transition equation;
(4) Initializing a particle filtering algorithm, wherein parameters comprise: number of particles N, process noise w k Covariance R of (1), observation noise v k The covariance Q of (a);
(5) The method carries out the advanced prediction of the capacity based on the particle filter algorithm, and calculates the state estimation value of the battery capacity, namely the capacity prediction value C k+1 ；
(6) Setting the number of advanced forecasting steps as m, and sequentially and repeatedly executing the steps to finally obtain a capacity forecasting value C of m advanced steps k+1 、C k+2 、…、C k+m 。
6. The method for monitoring the health of the full life cycle of the power battery as claimed in claim 1,2 or 3, wherein in the potential diving interval, the outlier removal is performed on the capacity sequence of the potential diving interval of the battery, then a state transfer equation and an observation equation are established based on the capacity sequence of the battery in the potential diving interval, and finally the battery capacity advanced multi-step forecast is performed.
7. The power battery full-life cycle health monitoring method according to claim 6, wherein battery capacity advanced multi-step forecasting is performed in the potential diving interval, and comprises the following steps:
(1) Preprocessing a capacity sequence of a potential diving interval of the battery to remove outliers;
(2) Establishing a state transition equation and an observation equation according to an empirical model of the battery degradation process, wherein the state transition equation and the observation equation are as follows:
wherein, w k Representing process noise, v k Representing observed noise, k =1,2, … represents the cycle period of battery charging and discharging;
(3) Setting the current charge-discharge cycle k as a prediction starting point, and performing state tracking on the historical capacity sequence by using a particle filter algorithm to determine unknown parameters in a state transition equation;
(4) Initializing a particle filtering algorithm, wherein parameters comprise: number of particles N, process noise w k Covariance R of (1), observation noise v k The covariance Q of (a);
(5) The method carries out the advanced prediction of the capacity based on the particle filter algorithm, and calculates the state estimation value of the battery capacity, namely the capacity prediction value C k+1 ；
(6) Setting the initial advanced prediction step number as m 0 Sequentially and repeatedly executing the steps to obtain a capacity forecast value
(7) After the current charge-discharge cycle enters k +1 from k, the advanced prediction step number is changed into m 0 And +1, sequentially and repeatedly executing the steps, then increasing the current charge-discharge cycle, and gradually forecasting the step number by +1.
8. The method for monitoring the health of the full life cycle of the power battery as claimed in claim 1,2 or 3, wherein in the key stage, the potential diving interval and the key stage capacity sequence of the battery are removed from the outlier, then a state transfer equation and an observation equation are established based on the potential diving interval and the key stage capacity sequence of the battery, and finally the whole-course residual life prediction is performed according to the following formula:
rul k ＝inf(l:f(l+k)≥γ)
wherein inf (l) represents the lower limit of the charge-discharge cycle number l; γ represents a failure threshold of the discharge capacity, here 80% of the initial capacity of the battery; rul k Indicating the remaining life of the battery at the kth charge-discharge cycle.
9. The power battery full-life cycle health monitoring method of claim 8, wherein predicting remaining battery life at the critical phase comprises:
(1) Determining a first prediction time FPT;
(2) Preprocessing a capacity sequence of a potential diving interval and a key stage of the battery to remove outliers;
(3) Establishing a state transition equation and an observation equation according to an empirical model of the battery degradation process, wherein the state transition equation and the observation equation are as follows:
wherein, w k Representing process noise, v k Representing observed noise, k =1,2, … represents the cycle period of battery charging and discharging;
(4) Setting a current charge-discharge cycle k as a prediction starting point, and performing state tracking on a battery potential diving interval and a historical capacity sequence of a key stage by using a particle filter algorithm to determine unknown parameters in a state transition equation;
(5) Initializing a particle filtering algorithm, wherein parameters comprise: number of particles N, process noise w k Covariance R of (1), observation noise v k The covariance Q of (a);
(6) Carrying out capacity advanced prediction based on particle filter algorithm, and calculating state estimation value C of battery capacity k+1 ；
(7) Determining a capacity advance prediction value C k+1 Whether the failure threshold of the battery, that is, 80% of the initial capacity, is reached, and if not, the above steps are sequentially and repeatedly performedWhen the predicted value C is k+n And if the residual service life of the battery exceeds the battery failure threshold value, the residual service life of the battery under the current charge-discharge cycle is n-1 charge-discharge cycles.
10. The method for monitoring the full-life-cycle health of the power battery according to claim 5, 7 or 9, wherein the performing of the advanced capacity prediction based on the particle filter algorithm specifically comprises: initializing a set of particles and generating particles; carrying out importance sampling; calculating the weight and normalizing the particle weight; resampling is carried out, and a grain set and weight after resampling are obtained; finally, calculating the state estimation value C of the battery capacity k+1 And recorded into the sequence.
11. The system is characterized by comprising three management modules of the full life cycle of the battery, namely a degradation stage management module, a potential diving interval management module and a key stage management module, wherein the three modules adopt different management methods according to the stage of the battery:
the degradation stage management module is used for carrying out battery capacity advanced multi-step forecasting based on a battery historical capacity sequence in a degradation stage;
the potential diving interval management module is used for carrying out battery capacity advanced prediction in the potential diving interval based on a capacity sequence in the potential diving interval, and the number of advanced prediction steps is larger than that of degradation stages;
and the key stage management module is used for predicting the residual life in the whole process of the key stage.
12. The power battery full-life-cycle health monitoring system of claim 1, wherein the potential diving interval is obtained by counting diving point data of a cloud-end like battery as a priori estimation of a current battery diving point, and specifically, the method comprises the steps of firstly extracting the characteristics of a user's customary driving speed, driving style, electric appliance information, a customary SOC driving interval and a customary SOC charging and discharging interval from the driving behavior and charging habit of a user to whom the battery belongs in the cloud-end like battery, and establishing a clustering model for the extracted characteristics; extracting the features from the current battery, and predicting the category of the user through the clustering model; and then counting the distribution of the number of charge-discharge cycles when the capacity of the battery in the category jumps, and acquiring a confidence interval of the distribution as a potential diving interval of the current battery.
13. The system for monitoring the full-life-cycle health of the power battery according to claim 11 or 12, wherein the degradation stage management module removes outliers from the battery historical capacity sequence, establishes a state transition equation and an observation equation based on the battery historical capacity sequence, and performs a battery capacity advanced multi-step forecast.
14. The power battery full-life-cycle health monitoring system according to claim 11 or 12, wherein the potential diving interval management module is configured to perform outlier removal on a battery potential diving interval capacity sequence, then establish a state transfer equation and an observation equation based on the potential diving interval battery capacity sequence, and finally perform battery capacity advanced multi-step forecasting.
15. The system for monitoring the health of the power battery in the full life cycle according to claim 11 or 12, wherein the key stage management module removes outliers from the potential diving interval and the key stage capacity sequence of the battery, establishes a state transfer equation and an observation equation based on the potential diving interval and the key stage capacity sequence of the battery, and predicts the remaining life of the battery in the full life cycle according to the following formula:
rul k ＝inf(l:f(l+k)≥γ)
wherein inf (l) represents the lower limit of the charge-discharge cycle number l; γ represents the failure threshold of the discharge capacity, here 80% of the initial capacity of the battery; rul k Indicating the remaining life of the battery at the kth charge-discharge cycle.
16. An electronic device, comprising:
one or more processors;
storage means for storing one or more programs that, when executed by the one or more processors, cause the electronic device to implement the steps of the power cell full life cycle health monitoring method of any of claims 1-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211174113.4A CN115407224A (en) | 2022-09-26 | 2022-09-26 | Power battery full life cycle health monitoring method and system and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211174113.4A CN115407224A (en) | 2022-09-26 | 2022-09-26 | Power battery full life cycle health monitoring method and system and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115407224A true CN115407224A (en) | 2022-11-29 |
Family
ID=84165662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211174113.4A Pending CN115407224A (en) | 2022-09-26 | 2022-09-26 | Power battery full life cycle health monitoring method and system and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115407224A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117806220A (en) * | 2024-03-01 | 2024-04-02 | 连云港市新达电子技术有限公司 | Intelligent control system for power supply of petroleum logging instrument |
-
2022
- 2022-09-26 CN CN202211174113.4A patent/CN115407224A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117806220A (en) * | 2024-03-01 | 2024-04-02 | 连云港市新达电子技术有限公司 | Intelligent control system for power supply of petroleum logging instrument |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107957562B (en) | Online prediction method for residual life of lithium ion battery | |
| CN107918103B (en) | Lithium ion battery residual life prediction method based on gray particle filtering | |
| CN106443478B (en) | The evaluation method of ferric phosphate lithium cell remaining capacity based on closed loop hybrid algorithm | |
| EP4019321A1 (en) | Battery managing method and device, and vehicle | |
| CN109100655B (en) | Data processing method and device for power battery | |
| CN107132490B (en) | Method for estimating state of charge of lithium battery pack | |
| CN105116344B (en) | Based on binary-coded battery open circuit voltage evaluation method | |
| CN108205114B (en) | Method and system for predicting service life of battery | |
| CN113777510A (en) | Lithium battery state of charge estimation method and device | |
| CN111967194B (en) | Battery classification method based on cloud historical data | |
| CN110568373A (en) | Lithium battery health state evaluation method, system, terminal and storage medium | |
| CN112611976A (en) | Power battery state of health estimation method based on double differential curves | |
| CN115407211B (en) | Online prediction method and system for health state of lithium battery of electric vehicle | |
| CN115407224A (en) | Power battery full life cycle health monitoring method and system and electronic equipment | |
| CN114114038A (en) | Lithium battery SOC and available capacity joint estimation method under full-life and full-temperature conditions | |
| CN114167301A (en) | Power battery evaluation method based on real vehicle data of electric vehicle | |
| CN114594382A (en) | Battery energy state prediction method and device, electronic equipment and storage medium | |
| CN115512777B (en) | Electrochemical model parameter identification method and system based on capacity change rate | |
| CN114509679B (en) | Battery SOH prediction model construction method based on deep learning | |
| CN111999659A (en) | Characteristic value method-based SOH estimation method for lithium iron phosphate battery and storage medium | |
| CN114114051A (en) | Method and device for determining battery aging value and electronic equipment | |
| CN116500475B (en) | Energy storage acquisition method and system with real-time SOC correction compensation | |
| CN111931415A (en) | Global optimal particle filter-based life prediction method for lithium ion battery | |
| CN112630658A (en) | Method and system for evaluating service life loss of storage battery | |
| CN115407217B (en) | Online estimation method and system for state of charge of lithium battery of electric vehicle |
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 | ||
| CB02 | Change of applicant information |
Address after: 401133 room 208, 2 house, 39 Yonghe Road, Yu Zui Town, Jiangbei District, Chongqing Applicant after: Deep Blue Automotive Technology Co.,Ltd. Address before: 401133 room 208, 2 house, 39 Yonghe Road, Yu Zui Town, Jiangbei District, Chongqing Applicant before: CHONGQING CHANGAN NEW ENERGY AUTOMOBILE TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information |