Parameter error estimation method of lithium ion battery P2D model

Abstract

The invention discloses a parameter error estimation method of a lithium ion battery P2D model, and relates to the technical field of lithium ion batteries. The comso has efficient numerical calculation and symbol calculation functions, supports large-scale parallel calculation, and is suitable for complex simulation tasks; multi-dimensional modeling is supported, and the physical quantity change process at different time and different positions can be observed; rich data, curves, pictures and animation output and analysis tools are provided, and a user is helped to better understand a simulation result; battery parameter error source current sampling or battery capacity change can be determined by setting error determination; formulating original data according to current sampling or battery capacity change; an error prediction result with a confidence interval is obtained; superposing the initial estimation result by using an error prediction result with a confidence interval to obtain a final prediction result; and the data in the database is updated in real time, so that the prediction accuracy can be ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium-ion batteries, and in particular to a parameter error estimation method for a lithium-ion battery P2D model. Background Art

[0002] Lithium-ion batteries, commonly known as rocking chair batteries, are a general term for batteries with lithium-ion embedded compounds as positive electrode materials. They are a type of secondary battery (rechargeable battery) that mainly relies on the movement of lithium ions between the positive and negative electrodes to work. Lithium-ion batteries are composed of positive electrodes, negative electrodes, electrolytes and diaphragms, plus positive and negative leads, battery shells, safety devices, etc. During the charging and discharging process of lithium-ion batteries, Li+ is embedded and de-embedded back and forth between the two electrodes: when charging, Li+ is de-embedded from the positive electrode and embedded into the negative electrode through the electrolyte, and the negative electrode is in a lithium-rich state; the opposite is true during discharge. Mobile phones and laptops all use lithium-ion batteries, which are usually referred to as lithium batteries. As a power source, lithium-ion batteries are widely used in individual electronic equipment, submarines, torpedoes, and drones.

[0003] The battery management system (BMS) is used to control and regulate the battery, which is premised on an accurate description and estimation of the internal working state of the battery. In the prior art, the electrochemical model is most suitable for estimating the internal state of the battery because it can accurately reflect the working mechanism of the battery and the nature of the electrochemical reaction. Among several different electrochemical models, the pseudo-two-dimensional model (P2D model) based on the porous electrode theory and the concentrated solution theory is the most classic electrochemical model for studying the internal state of the battery, and it is also the most commonly used. For this reason, we propose a parameter error estimation method for the P2D model of lithium-ion batteries. Summary of the invention

[0004] The purpose of the present invention is to solve the problems mentioned in the above background technology. The present invention provides a parameter error estimation method for a lithium-ion battery P2D model.

[0005] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0006] A parameter error estimation method for a lithium-ion battery P2D model comprises the following steps:

[0007] S1, data collection, used to collect real-time battery data;

[0008] S2, building a database for presetting comparison data;

[0009] S3. Building a learning model to determine the type of parameters to be estimated, and the type of parameters to be estimated includes dynamic data and static data;

[0010] S4, data evaluation, used to evaluate the data of the learning model;

[0011] S5, error determination, used to determine the error;

[0012] S6: Whether the stop condition is met, if yes, proceed to the next step, if not, return to S3;

[0013] S7: Generate results, used to generate evaluation results.

[0014] Furthermore, the S1 comprises the following steps:

[0015] S11, data preprocessing, collecting battery status data and size parameters of lithium-ion batteries, and performing preprocessing after collection;

[0016] S12, data analysis, effectively analyzing the data after data preprocessing;

[0017] S13, data classification, classifying the data after data analysis;

[0018] S14. Data integration: effectively integrate the data after data classification.

[0019] Furthermore, S2 comprises the following steps:

[0020] S21, establishing a data set;

[0021] S22, establish a test set;

[0022] S23, establishing a prediction model;

[0023] S24: Generate prediction results.

[0024] Furthermore, the data set establishment in S21 is based on the external characteristic parameters and internal mechanism change data during the discharge process of the lithium-ion battery to establish data sets of different external characteristic parameters and internal mechanism change data.

[0025] Furthermore, the establishment of the test set in S22 is based on repeatedly collecting data in S21.

[0026] Furthermore, the prediction model established in S23 is used to accurately estimate the lithium-ion battery error.

[0027] Furthermore, the S3 learning model is constructed based on comsol implementation.

[0028] Furthermore, the data evaluation in S4 also includes a data storage unit for storing data.

[0029] Furthermore, the S5 comprises the following steps:

[0030] S51, determine the source of error;

[0031] S52, establishing an original error series;

[0032] S53, correlation analysis error sequence;

[0033] S54, error correction;

[0034] S55. Database update.

[0035] Furthermore, whether the stop condition is met in S6 is implemented based on a P2D model.

[0036] The beneficial effects of the present invention are as follows:

[0037] 1. The comsol of the present invention has efficient numerical and symbolic computing functions, supports large-scale parallel computing, and is suitable for complex simulation tasks; supports multi-dimensional modeling, and can observe the change process of physical quantities at different times and positions; provides rich data, curves, pictures and animation output and analysis tools to help users better understand the simulation results.

[0038] 2. The present invention can determine the source of battery parameter error, i.e., current sampling or battery capacity change, by setting the error; formulate original data according to the current sampling or battery capacity change; and obtain error prediction results with confidence intervals; superimpose the initial estimation results with the error prediction results with confidence intervals to obtain the final prediction results; and update the data in the database in real time, thereby ensuring the accuracy of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a work flow chart of the present invention;

[0040] Figure 2 It is a workflow diagram of data collection in the present invention;

[0041] Figure 3 It is a workflow diagram for constructing a database in the present invention;

[0042] Figure 4 It is a workflow diagram for error determination in the present invention. DETAILED DESCRIPTION

[0043] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0044] See also Figure 1 - Figure 4 The present invention provides a parameter error estimation method for a lithium-ion battery P2D model, comprising the following steps:

[0045] S1, data collection, used to collect real-time battery data;

[0046] S2, building a database for presetting comparison data;

[0047] S3. Building a learning model to determine the type of parameters to be estimated, and the type of parameters to be estimated includes dynamic data and static data;

[0048] S4, data evaluation, used to evaluate the data of the learning model;

[0049] S5, error determination, used to determine the error;

[0050] S6: Whether the stop condition is met, if yes, proceed to the next step, if not, return to S3;

[0051] S7: Generate results, used to generate evaluation results.

[0052] In this embodiment, preferably, the S1 includes the following steps:

[0053] S11, data preprocessing, collecting battery status data and size parameters of lithium-ion batteries, and performing preprocessing after collection;

[0054] S12, data analysis, effectively analyzing the data after data preprocessing;

[0055] S13, data classification, classifying the data after data analysis;

[0056] S14. Data integration: effectively integrate the data after data classification.

[0057] In this embodiment, preferably, S2 includes the following steps:

[0058] S21, establishing a data set;

[0059] S22, establish a test set;

[0060] S23, establishing a prediction model;

[0061] S24: Generate prediction results.

[0062] In this embodiment, preferably, the data set established in S21 is based on the external characteristic parameters and internal mechanism change data during the discharge process of the lithium-ion battery to establish data sets of different external characteristic parameters and internal mechanism change data.

[0063] In this embodiment, preferably, the establishment of the test set in S22 is based on repeatedly collecting data in S21.

[0064] In this embodiment, preferably, the prediction model established in S23 is used to accurately estimate the lithium-ion battery error.

[0065] In this embodiment, preferably, the S3 constructs the learning model based on comsol implementation.

[0066] In this embodiment, preferably, the data evaluation in S4 also includes a data storage unit for storing data.

[0067] In this embodiment, preferably, S5 includes the following steps:

[0068] S51, determining the error source, determining the battery parameter error source current sampling or battery capacity change;

[0069] S52, establishing an original error series, and formulating original data according to current sampling or battery capacity change;

[0070] S53, correlation analysis error sequence, compare the data of S52 and S53, and obtain the error prediction result with confidence interval;

[0071] S54, error correction, superimposing the initial estimation result with the error prediction result with the confidence interval to obtain the final prediction result;

[0072] S55, database update, real-time update of the data in the database to ensure the accuracy of the prediction.

[0073] In this embodiment, preferably, whether the stop condition is met in S6 is implemented based on a P2D model.

[0074] The working principle and use process of the present invention:

[0075] S1. Data collection is used to collect real-time data of the battery, which includes the following steps:

[0076] S11, data preprocessing, collecting battery status data and size parameters of lithium-ion batteries, and performing preprocessing after collection;

[0077] S12, data analysis, effectively analyzing the data after data preprocessing;

[0078] S13, data classification, classifying the data after data analysis;

[0079] S14. Data integration: effectively integrate the data after data classification.

[0080] S2, constructing a database for presetting comparison data, which includes the following steps:

[0081] S21, establishing a data set, based on the external characteristic parameters and internal mechanism change data during the discharge process of the lithium-ion battery, establishing a data set of different external characteristic parameters and internal mechanism change data;

[0082] S22, establish a test set, and repeatedly collect data based on S21;

[0083] S23, establishing a prediction model for accurately estimating lithium-ion battery errors;

[0084] S24: Generate prediction results.

[0085] S3. Construct a learning model to determine the type of parameters to be estimated, and the type of parameters to be estimated includes dynamic data and static data. Based on COMSO L, COMSO L can simulate the interaction between different physical fields, such as mechanics, electromagnetism, fluid dynamics, thermodynamics, acoustics, etc., to achieve accurate numerical simulation; it has efficient numerical and symbolic calculation functions, supports large-scale parallel calculations, and is suitable for complex simulation tasks; it has a friendly user interface and a natural language close to mathematical expressions, which is easy to learn and master; it provides professional toolboxes such as signal processing, communication, and fluid dynamics, which facilitates users to perform various processing and analysis; it supports custom partial differential equations, can easily add any physical field, and has strong scalability; it has a rich set of CAD modeling tools, supports direct import of mainstream CAD software format files, and facilitates seamless conversion from design to simulation; it supports multi-dimensional modeling, and can observe the change process of physical quantities at different times and positions; it provides a wealth of data, curves, pictures and animation output and analysis tools to help users better understand the simulation results; it supports multiple language interfaces to meet the needs of users in different countries and regions; it is widely used in engineering design and scientific research, such as thermal effect analysis of electronic equipment and the influence of fluid flow on structure.

[0086] S4, data evaluation, is used to evaluate the data of the learning model; wherein the data evaluation also includes a data storage unit for storing data.

[0087] S5, error determination, used to determine the error; it includes the following steps:

[0088] S51, determining the error source, determining the battery parameter error source current sampling or battery capacity change;

[0089] S52, establishing an original error series, and formulating original data according to current sampling or battery capacity change;

[0090] S53, correlation analysis error sequence, compare the data of S52 and S53, and obtain the error prediction result with confidence interval;

[0091] S54, error correction, superimposing the initial estimation result with the error prediction result with the confidence interval to obtain the final prediction result;

[0092] S55, database update, real-time update of the data in the database to ensure the accuracy of the prediction.

[0093] S6: Whether the stop condition is met, if so, proceed to the next step, if not, return to S3; wherein, whether the stop condition is met is implemented based on the P2D model.

[0094] S7: Generate results, used to generate evaluation results.

[0095] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A parameter error estimation method for a lithium-ion battery P2D model, characterized in that: The steps include: S1, data collection, used to collect real-time battery data; S2, building a database for presetting comparison data; S3. Building a learning model to determine the type of parameters to be estimated, and the type of parameters to be estimated includes dynamic data and static data; S4, data evaluation, used to evaluate the data of the learning model; S5, error determination, used to determine the error; S6: Whether the stop condition is met, if yes, proceed to the next step, if not, return to S3; S7: Generate results, used to generate evaluation results.

2. The parameter error estimation method of a lithium-ion battery P2D model according to claim 1, characterized in that: The S1 comprises the following steps: S11, data preprocessing, collecting battery status data and size parameters of lithium-ion batteries, and performing preprocessing after collection; S12, data analysis, effectively analyzing the data after data preprocessing; S13, data classification, classifying the data after data analysis; S14. Data integration: effectively integrate the data after data classification.

3. The parameter error estimation method of a lithium-ion battery P2D model according to claim 1, characterized in that: The S2 comprises the following steps: S21, establishing a data set; S22, establish a test set; S23, establishing a prediction model; S24: Generate prediction results.

4. The parameter error estimation method of a lithium-ion battery P2D model according to claim 3, characterized in that: The data set establishment in S21 is based on the external characteristic parameters and internal mechanism change data during the discharge process of the lithium-ion battery to establish data sets with different external characteristic parameters and internal mechanism change data.

5. The parameter error estimation method of a lithium-ion battery P2D model according to claim 3, characterized in that: The establishment of the test set in S22 is based on repeated data collection in S21.

6. The parameter error estimation method of a lithium-ion battery P2D model according to claim 3, characterized in that: The prediction model established in S23 is used to accurately estimate the lithium-ion battery error.

7. The parameter error estimation method of a lithium-ion battery P2D model according to claim 1, characterized in that: The S3 learning model is implemented based on comsol.

8. The parameter error estimation method of a lithium-ion battery P2D model according to claim 1, characterized in that: The data evaluation in S4 also includes a data storage unit for storing data.

9. The parameter error estimation method of a lithium-ion battery P2D model according to claim 1, characterized in that: The S5 comprises the following steps: S51, determine the source of error; S52, establishing an original error series; S53, correlation analysis error sequence; S54, error correction; S55. Database update.

10. The parameter error estimation method of a lithium-ion battery P2D model according to claim 1, characterized in that: Whether the stop condition is met in S6 is implemented based on the P2D model.

Images