CN114563713A - Battery health prediction method based on partial energy characteristics of incomplete charging process - Google Patents
Battery health prediction method based on partial energy characteristics of incomplete charging process Download PDFInfo
- Publication number
- CN114563713A CN114563713A CN202210240487.5A CN202210240487A CN114563713A CN 114563713 A CN114563713 A CN 114563713A CN 202210240487 A CN202210240487 A CN 202210240487A CN 114563713 A CN114563713 A CN 114563713A
- Authority
- CN
- China
- Prior art keywords
- charging
- charging data
- constant
- voltage
- data
- 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]
- G01R31/367—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]
- G01R31/378—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC] specially adapted for the type of battery or accumulator
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
- Tests Of Electric Status Of Batteries (AREA)
Abstract
The invention relates to the technical field of battery management, and discloses a battery health prediction method based on partial energy characteristics of an incomplete charging process, which comprises the following steps: collecting charging data of an electric vehicle battery, wherein the charging data comprises constant voltage charging data and constant current charging data, and respectively extracting aging characteristics of the constant voltage charging data and the constant current charging data; if the charging data comprises constant voltage charging data and meets the condition A, taking the aging characteristic of the constant voltage charging data as input, taking the health degree of the battery as output, and establishing a first model by utilizing a machine learning algorithm; if the charging data only contain the constant-current charging data and meet the condition B, or the charging data contain the constant-voltage charging data and do not meet the condition A, taking the aging characteristic of the constant-current charging data as input, taking the battery health degree as output, and establishing a second model by utilizing a machine learning algorithm; and inputting the charging data of the electric vehicle into the first model or the second model to obtain a health state predicted value.
Description
Technical Field
The invention relates to the technical field of battery management, in particular to a battery health prediction method based on partial energy characteristics in an incomplete charging process.
Background
The lithium ion battery has the characteristics of long service life, low self-discharge rate and high energy density, can become good energy storage equipment, and can be widely applied to electrical equipment, such as pure electric vehicles, plug-in hybrid electric vehicles, mobile energy storage equipment, power grids and the like.
The current methods for estimating the state of health of a power battery of an electric vehicle can be roughly divided into two types, namely a model-based method and a data-driven method. The model-based method needs certain priori knowledge to construct appropriate models which can reflect the electrochemical characteristics of the battery to a certain extent, such as an electrochemical model and an equivalent circuit model, and then estimates the health state by using recursion algorithms such as Kalman filtering and particle filtering and based on real-time voltage and current. The data-driven method analyzes a large amount of experimental data of battery aging, extracts features related to the battery aging, and then describes the correlation between the features and the health state by using a machine learning tool.
At present, there is a method of performing a state of health estimation by extracting a characteristic using a capacity increment curve of a battery charging stage and establishing a mapping relationship between a peak characteristic of the curve and a state of health of the battery. However, these methods basically do not consider that the capacity increase curve in the actual charging condition may be incomplete, and also basically do not consider the influence of the temperature on the curve peak, and the charging and discharging capacity of the battery is easily influenced by the temperature, so that the increase curve of the charging capacity may also be influenced by the temperature. In addition, there are some methods for performing aging characteristic extraction and health state estimation by using charge and discharge segment data, but in order to apply these methods in an actual charge and discharge process full of uncertainty, the charge and discharge segments of the training data are often very large.
Because the charging and discharging behaviors of the electric automobile in the actual running process are full of uncertainty, complete discharging or complete full charging processes are rare, and many aging characteristics related to the complete discharging or full charging processes are invalid. Therefore, how to extract features by using incomplete charging and discharging data and search aging features with anti-interference capability is a common problem of the current data-driven method.
Disclosure of Invention
In order to solve the technical problem, the invention provides a battery health prediction method based on the partial energy characteristics of the incomplete charging process.
In order to solve the technical problems, the invention adopts the following technical scheme:
a battery health prediction method based on partial energy characteristics of an incomplete charging process comprises the following steps:
collecting charging data of an electric vehicle battery, wherein the charging data comprises constant voltage charging data and constant current charging data, and respectively extracting aging characteristics of the constant voltage charging data and the constant current charging data; according to the difference of charging voltage curves in different aging states, the charging voltage in the constant current charging stage can be divided into characteristic voltage intervals;
if the charging data comprises constant voltage charging data and meets the condition A, taking the aging characteristics of the constant voltage charging data as input, taking the battery health degree as output, and establishing a health state prediction model I by utilizing a machine learning algorithm;
if the charging data only contain the constant-current charging data and meet the condition B, or the charging data contain the constant-voltage charging data and do not meet the condition A, the aging characteristics of the constant-current charging data are used as input, the battery health degree is used as output, and a health state prediction model II is established by utilizing a machine learning algorithm;
wherein the condition A is: the charging cut-off current in the constant-voltage charging process is less than or equal to x times of the charging current in the constant-current charging process, and x is between 0.05 and 1; the condition B is as follows: the constant current charging data at least comprises a complete characteristic voltage interval;
and inputting the charging data of the electric vehicle into the first health state prediction model or the second health state prediction model to obtain a health state prediction value.
The closer x is to 0.05, the more complete the battery process is, and for enhancing the practicability, the greater x should be taken under the premise of meeting the requirement of extracting the aging characteristic data quantity, wherein x is 0.5.
Specifically, the aging characteristics of the constant voltage charging data include an energy increment, a charging temperature, and a charging rate in the constant voltage charging stage.
Specifically, according to the difference of charging voltage curves in different aging states, the charging voltage in the constant-current charging stage is divided into a plurality of characteristic voltage intervals, and in each characteristic voltage interval, the charging voltage curves corresponding to the different aging states are different from the areas enveloped by coordinate axes; the aging characteristics of the constant-current charging data comprise energy increment, charging temperature and charging multiplying power in each characteristic voltage interval in the constant-current charging stage as aging characteristics.
Specifically, if the charging data contains neither constant voltage charging data nor constant current charging data, the last health state prediction value is taken as the health state prediction value of this time.
Compared with the prior art, the invention has the beneficial technical effects that:
in order to reduce the complexity of the training of the health state prediction model and ensure the practicability and precision of the algorithm, the invention takes the statistical characteristics of the actual charging behavior of the electric vehicle into consideration, divides all charging processes which may occur, extracts the aging characteristics of the corresponding processes, sets the priority of the health prediction model, and realizes the adjustment of the aging characteristics and the estimation method aiming at different charging working conditions. In addition, the health state prediction method can predict the health state by using data under any charging starting and stopping point conditions, which is an effect that most of the existing health prediction methods do not have, and is also a main characteristic of the health state prediction method.
Drawings
Fig. 1 is a schematic flow chart of a battery health prediction method according to the present invention.
Detailed Description
A preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings.
The invention provides an extraction idea of battery aging characteristics. According to the research of a steam group on the charging and discharging behaviors of the under-flag electric vehicle, about 60% of users are charged until the charge state reaches 95% -100%, the charge state distribution of users in various regions during charging and starting is similar, the charge state distribution is mainly 20% -80%, and the charge state is 35% -45% in the maximum ratio. As can be seen, users generally tend to be fully charged at a time, reducing range anxiety, while the state of charge of the battery before charging is wide in floating range, and the discharging behavior of the battery is fully random and difficult to extract features.
Considering that the lithium battery usually needs complete constant current charging and partial constant voltage charging to reach more than 95% of the state of charge, the invention carries out stage division and priority division on the charging process of the battery: when the charging process comprises charging data in a constant voltage stage, selecting aging characteristics of the constant voltage charging data as input of a prediction model; and when only the charging data in the constant current stage exists, selecting the aging characteristic of the constant current charging data as the input of the prediction model, wherein the specific scheme is as follows.
As shown in fig. 1, a method for predicting battery health based on partial energy characteristics of an incomplete charging process includes the following steps:
(1) the method comprises the steps of collecting charging data of an electric vehicle battery, wherein the charging data comprise constant voltage charging data and constant current charging data, and respectively extracting aging characteristics of the constant voltage charging data and the constant current charging data.
(2) And if the charging data comprises constant voltage charging data and meets the condition A, taking the aging characteristics of the constant voltage charging data as input, taking the battery health degree as output, and establishing a first health state prediction model by utilizing a machine learning algorithm.
(3) And if the charging data only contain the constant-current charging data and satisfy the condition B, or the charging data contain the constant-voltage charging data and do not satisfy the condition A, establishing a second health state prediction model by using a machine learning algorithm and taking the aging characteristic of the constant-current charging data as input and the battery health degree as output.
Wherein the condition A is: the charging cut-off current in the constant-voltage charging process is less than or equal to x times of the charging current in the constant-current charging process, and x is between 0.05 and 1; the condition B is as follows: the constant current charging data at least comprises a complete characteristic voltage interval. The closer x is to 0.05, the more complete the battery process is, and for enhancing the practicability, the greater x should be taken under the premise of meeting the requirement of extracting the aging characteristic data quantity, wherein x is 0.5. If the charging process comprises a constant voltage charging process and the condition A is met, the constant voltage charging data is enough; if the charging process includes a constant voltage charging process but the condition A is not satisfied, it indicates that the constant voltage charging data is insufficient.
(4) And inputting the charging data of the electric vehicle into the first health state prediction model or the second health state prediction model to obtain a health state prediction value.
(5) And if the charging data does not contain constant voltage charging data or constant current charging data, taking the last health state predicted value as the current health state predicted value.
In a dotted line frame I of fig. 1, the battery health prediction method selects the charging energy, the charging temperature, and the charging rate in the constant voltage charging stage as input characteristics of a first health state prediction model.
According to the difference of charging voltage curves under different aging states, the charging voltage in the constant current charging stage is divided into a plurality of characteristic voltage intervals, and in each characteristic voltage interval, the charging voltage curves corresponding to different aging states are different from the area enveloped by a coordinate axis.
In a dashed line frame II in fig. 1, the battery health prediction method selects the energy increment, the charging temperature, and the charging rate of all the characteristic voltage intervals included in the constant current charging phase as the input characteristics of the health state prediction model II.
In a dashed line frame III in fig. 1, the battery health prediction method directly reads the last predicted value of the health state as the current predicted value of the health state, which indicates that the charging process is too short and the aging characteristics included in the charging data are not sufficient to predict the health state.
In the health prediction method, the priorities of the three prediction schemes are sequentially reduced, all possible charging conditions are considered, and then the method with the highest level is selected according to the flow shown in fig. 1 and applied according to the conditions met by the current charging process. The aging characteristic extraction can be carried out by utilizing incomplete charging voltage, current and temperature data of the power battery under the charging and discharging scene which is more prone to the actual process, and the correlation between the local charging energy increment and the battery aging is analyzed, so that the current health condition of the battery is predicted. Obviously, the method is suitable for online estimation of the state of health of the battery under any charging starting and stopping point condition, and aging characteristics and estimation methods can be adjusted according to different charging working conditions.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein, and any reference signs in the claims are not intended to be construed as limiting the claim concerned.
Furthermore, it should be understood that although the specification describes embodiments, not every embodiment includes only a single embodiment, and such description is for clarity purposes only, and it will be understood by those skilled in the art that the specification as a whole and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (4)
1. A battery health prediction method based on partial energy characteristics of an incomplete charging process comprises the following steps:
collecting charging data of an electric vehicle battery, wherein the charging data comprises constant voltage charging data and constant current charging data, and respectively extracting aging characteristics of the constant voltage charging data and the constant current charging data; according to the difference of charging voltage curves under different aging states, characteristic voltage intervals can be divided for the charging voltage in the constant current charging stage;
if the charging data comprises constant voltage charging data and meets the condition A, taking the aging characteristics of the constant voltage charging data as input, taking the battery health degree as output, and establishing a health state prediction model I by utilizing a machine learning algorithm;
if the charging data only contain the constant-current charging data and meet the condition B, or the charging data contain the constant-voltage charging data and do not meet the condition A, the aging characteristics of the constant-current charging data are used as input, the battery health degree is used as output, and a health state prediction model II is established by utilizing a machine learning algorithm;
wherein the condition A is: the charging cut-off current in the constant-voltage charging process is less than or equal to x times of the charging current in the constant-current charging process, and x is between 0.05 and 1; the condition B is as follows: the constant current charging data at least comprises a complete characteristic voltage interval;
and inputting the charging data of the electric vehicle into the first health state prediction model or the second health state prediction model to obtain a health state prediction value.
2. The method of claim 1, wherein the method comprises: the aging characteristics of the constant voltage charging data include energy increment, charging temperature, and charging rate in the constant voltage charging stage.
3. The method of claim 1, wherein the method comprises: according to the difference of charging voltage curves under different aging states, dividing the charging voltage in the constant-current charging stage into a plurality of characteristic voltage intervals, wherein in each characteristic voltage interval, the charging voltage curves corresponding to different aging states are different from the areas enveloped by coordinate axes; the aging characteristics of the constant-current charging data comprise energy increment, charging temperature and charging multiplying power in each characteristic voltage interval in the constant-current charging stage as aging characteristics.
4. The method of claim 1, wherein the method comprises: and if the charging data does not contain constant voltage charging data or constant current charging data, taking the last health state predicted value as the current health state predicted value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210240487.5A CN114563713A (en) | 2022-03-10 | 2022-03-10 | Battery health prediction method based on partial energy characteristics of incomplete charging process |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210240487.5A CN114563713A (en) | 2022-03-10 | 2022-03-10 | Battery health prediction method based on partial energy characteristics of incomplete charging process |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114563713A true CN114563713A (en) | 2022-05-31 |
Family
ID=81718048
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210240487.5A Pending CN114563713A (en) | 2022-03-10 | 2022-03-10 | Battery health prediction method based on partial energy characteristics of incomplete charging process |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114563713A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115201686A (en) * | 2022-07-12 | 2022-10-18 | 中国科学技术大学 | Lithium ion battery health state assessment method under incomplete charging and discharging data |
-
2022
- 2022-03-10 CN CN202210240487.5A patent/CN114563713A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115201686A (en) * | 2022-07-12 | 2022-10-18 | 中国科学技术大学 | Lithium ion battery health state assessment method under incomplete charging and discharging data |
CN115201686B (en) * | 2022-07-12 | 2023-08-29 | 中国科学技术大学 | Lithium ion battery health state assessment method under incomplete charge and discharge data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108790893B (en) | Alternating current charging pile and charging control method | |
EP4019321A1 (en) | Battery managing method and device, and vehicle | |
CN112572233B (en) | Battery management method and device and vehicle | |
CN107132490B (en) | Method for estimating state of charge of lithium battery pack | |
CN107612076A (en) | Battery charging method, device, equipment and storage medium | |
CN115158097B (en) | Charging pile management method and system based on Internet of things and storage medium | |
CN111890985A (en) | Method for adjusting available power of battery of electric vehicle | |
CN107979119A (en) | Maas is fitted the battery charging control method and system of charging curve | |
CN109383323B (en) | Charge-discharge optimization control method for electric automobile group | |
CN103997052A (en) | A method for controlling the active power of multiple energy-storage power stations | |
CN109177805A (en) | A kind of electric car charge/discharge control method and system | |
Shi et al. | Electric vehicle battery remaining charging time estimation considering charging accuracy and charging profile prediction | |
CN108599267B (en) | Unit combination scheduling method considering electric vehicle traveling correlation | |
CN112688347A (en) | System and method for smoothing load fluctuation of power grid | |
CN114563713A (en) | Battery health prediction method based on partial energy characteristics of incomplete charging process | |
CN112670999A (en) | Low-voltage distribution network real-time voltage control method based on user-side flexible resources | |
CN113352946B (en) | Energy management method of fuel cell automobile power system | |
CN111931415A (en) | Global optimal particle filter-based life prediction method for lithium ion battery | |
CN111369159A (en) | Reliability assessment method for electric power traffic coupling network | |
Martel et al. | Economy-focused PHEV battery lifetime management through optimal fuel cell load sharing | |
CN114611595A (en) | Voltage curve self-learning method for battery of energy storage system | |
CN108110283B (en) | Flow battery energy storage system abnormal data identification method | |
CN114290959B (en) | Active life control method and system for power battery and computer readable storage medium | |
CN110293874B (en) | V2G microgrid system energy optimization control method based on fuzzy control | |
Shan et al. | Research on factors that influence the charging load of Electric Vehicles and modeling of the load |
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 |