A battery system temperature prediction method based on virtual sensor

Abstract

The present invention discloses a method for predicting the temperature of a battery system based on a virtual sensor, including: establishing a battery simulation model based on the heating and cooling mechanism of battery charging and discharging to simulate the temperature rise of the battery under different environments and working conditions; collecting temperature rise change data during the charging and discharging process of the battery system, verifying and improving the simulation model, and indirectly obtaining the output results of the virtual sensor based on the simulation data; fusing the measured temperature and the virtual measured temperature, extracting key characteristic parameters, and combining a deep neural network to establish a dynamic prediction model for temperature rise, so as to achieve rapid prediction of the temperature rise of the battery system after the working condition changes. This method is suitable for predicting the temperature changes during the charging and discharging process of the battery system. By fusing the measurement and virtual measurement data to construct a prediction model, the prediction of the battery temperature change can be achieved at a low cost.

Description

Technical Field

[0001] The present invention relates to the technical field of battery management, and in particular to a battery system temperature prediction method based on a virtual sensor. Background Art

[0002] With the implementation of my country's dual-carbon strategy, the development of new power systems and transportation electrification, energy storage systems, especially battery energy storage systems, have received more and more attention. Battery systems generate a lot of heat during charging and discharging. If they cannot be cooled in time and effectively, the internal temperature of the battery will rise rapidly, and the temperature consistency of the system will deteriorate, affecting its working performance, shortening its service life, and may cause serious accidents. Therefore, in order to ensure the long-term safe and efficient operation of the battery energy storage system, it is necessary to predict its temperature changes during charging and discharging under different working conditions and environments.

[0003] However, current battery system temperature prediction requires data measured by a large number of sensors. Too many sensors not only increase installation costs, but also increase the risk of failure. In addition, it is difficult to place sensors inside the battery. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above existing problems, a battery system temperature prediction method based on virtual sensor is proposed.

[0006] A method for predicting battery system temperature based on a virtual sensor, the method comprising the following steps: S1: establishing a battery simulation model based on the heating and cooling mechanism of battery charging and discharging, to simulate the temperature rise of the battery under different environments and working conditions; S2: collecting temperature rise change data during the charging and discharging process of the battery system, verifying and improving the simulation model, and indirectly obtaining the output result of the virtual sensor based on the simulation data; S3: fusing the measured temperature and the virtual measured temperature, extracting key characteristic parameters, and establishing a temperature rise dynamic prediction model in combination with a recurrent neural network, to achieve rapid prediction of the temperature rise of the battery system after the working condition changes.

[0007] The battery system temperature prediction method provided by the present invention establishes a battery numerical simulation model through the analysis of the heating and heat transfer mechanism, can simulate and calculate the heat generation and temperature of the battery before actual operation, and can simulate and analyze different working conditions and different environments, especially the conditions that are difficult to test such as extreme environmental conditions; further use the limited data measured by the temperature acquisition system using various means to verify and refine the simulation model, including key parameters such as heat source and heat dissipation coefficient and grid segmentation refinement strategy, to improve the calculation accuracy of the model, and then indirectly obtain the output result of the virtual sensor from the simulation data; based on a large number of virtual measured temperatures calculated and generated by the simulation model, especially the internal temperatures that are difficult to measure directly, the measured temperatures are integrated, and further combined with the heating and cooling mechanisms, the key parameters used as input feature quantities and outputs can be extracted, and a dynamic prediction model is established through a recurrent neural network to accurately predict the temperature changes of the battery system under different environmental conditions with a small number of sensors. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0009] Figure 1 A basic flow chart of a battery system temperature prediction method based on a virtual sensor provided by an embodiment of the present invention;

[0010] Figure 2 A schematic diagram of a prediction model of a battery system temperature prediction method based on a virtual sensor provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0011] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0012] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0013] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0014] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0015] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0016] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0017] Example

[0018] The present invention provides a battery system temperature prediction method based on a virtual sensor, comprising:

[0019] S1: Establish a battery simulation model based on the heating and cooling mechanism of battery charging and discharging to simulate the temperature rise of the battery under different environments and working conditions;

[0020] S2: Collect temperature rise data during the charging and discharging process of the battery system, verify and improve the simulation model, and indirectly obtain the output results of the virtual sensor based on the simulation data;

[0021] S3: The measured temperature and virtual measured temperature are integrated to extract key characteristic parameters, and a temperature rise dynamic prediction model is established in combination with a recurrent neural network to achieve rapid prediction of the temperature rise of the battery system after changes in operating conditions.

[0022] Reference Figure 1, steps S1-S3 specifically include:

[0023] Based on the working principle and heat generation mechanism of the battery system, the heat generated by the battery operation is determined by determining several main parts of the battery heat source, including the heat generated by the electrochemical reaction, the Joule heat of the battery internal resistance, and the heat generated by the battery polarization; based on the cooling heat transfer mechanism and combined with the actual heat dissipation conditions, the heat dissipation boundary conditions of the battery system are determined; based on the battery system structure, especially the internal structure, its geometric model is established, and the corresponding thermophysical parameters of each part of the material are set respectively; based on the computational fluid dynamics theory combined with the flow field properties, a numerical simulation model for the heat dissipation analysis of the battery system is established; the temperature change of the battery system is calculated, and the simulation parameters can be adjusted according to the operating environment and working conditions, and then the temperature change of the battery system under different operating environments and working conditions is analyzed.

[0024] Furthermore, the numerical simulation of battery heating includes: meshing of the geometric model of the battery system, and discretization of the control equations contained in its numerical simulation model in the time and space domains, including the conservation of mass, momentum, energy and turbulent transport equations, combining the thermal physical properties of the material and the convection and radiation boundary conditions to form a set of algebraic equations, and solving them to obtain the specific distribution of temperature.

[0025] The temperature acquisition system composed of an infrared thermal imager and a multi-channel temperature tester is used to comprehensively measure the outer surface and internal temperature of the battery under different working conditions. The infrared thermal imager can obtain the temperature distribution of the outer surface of the battery system through infrared thermal images, and the multi-channel temperature tester introduces the temperature measuring line through the hole in the end cover to measure the temperature of the corresponding position inside the battery. Based on the measured temperature rise results, the simulation data is extracted in combination with the temperature rise measurement position for comparison and verification, and the key parameters such as the heat source and heat dissipation coefficient used in the simulation are corrected, and the grid division and refinement strategy is adjusted accordingly to improve the calculation accuracy of the numerical model. Based on the improved numerical simulation model, the temperature changes of the battery system under different environments and working conditions are further calculated, and the output results of the virtual sensor placed at a fixed position are indirectly obtained.

[0026] A large number of virtual measured temperatures under different working conditions of the battery system generated based on simulation calculations are an effective supplement to the measurement data, especially the internal temperature which is difficult to measure directly and poses safety risks. Further combined with the heating and cooling mechanism of the battery system, the principal component analysis method is used to extract the key parameters used as input feature quantities. The input feature quantity data is preprocessed in combination with the characteristics of the recurrent neural network. A dynamic prediction model for temperature rise of the battery system is established through data training. The recurrent neural network is combined with the fully connected neural network to enhance the nonlinear expression ability of the model. The key hyperparameters in the prediction model, including the number of time steps, hidden layer parameters, and dropout ratio, are determined. The temperature correlation information between different times is learned. With the use of a small number of sensors, the temperature changes of the battery system under different environmental conditions can be accurately predicted at a low cost.

[0027] Furthermore, data preprocessing includes: using the principal component analysis method to extract physical quantities that affect the temperature rise of each screening, such as load size, duration, heat dissipation conditions, location information, etc., and using them as input feature quantities of the prediction model; preprocessing sample inputs with different dimensional value range distributions to improve the training speed of the model; using a sliding time window to construct the preprocessed sample inputs into a form that adapts to the model framework, which means that the number of time steps obtained by synthesizing the information of the previous two moments at most is equivalent to the number of recurrent neural network units.

[0028] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A battery system temperature prediction method based on virtual sensor, characterized in that: include: A battery simulation model is established based on the heating and cooling mechanism of battery charging and discharging to simulate the temperature rise of the battery under different environments and working conditions; Collect temperature rise change data during the charging and discharging process of the battery system, verify and improve the simulation model, and indirectly obtain the output results of the virtual sensor based on the simulation data; the temperature acquisition system composed of an infrared thermal imager and a multi-channel temperature tester comprehensively measures the outer surface and internal temperature of the battery under different working conditions, wherein the infrared thermal imager can obtain the temperature distribution of the outer surface of the battery system through the infrared thermal image, and the multi-channel temperature tester introduces the temperature measuring line through the hole in the end cover to measure the temperature at the corresponding position inside the battery; based on the measured temperature rise results, the simulation data is extracted in combination with the temperature rise measurement position for comparison and verification, and the key parameters such as the heat source and heat dissipation coefficient used in the simulation are corrected, and the grid division and refinement strategy is adjusted accordingly to improve the calculation accuracy of the numerical model; based on the improved numerical simulation model, the temperature changes of the battery system under different environments and working conditions are further calculated, and the output results of the virtual sensor placed in a fixed position are indirectly obtained; The measured temperature and virtual measured temperature are integrated, key characteristic parameters are extracted, and a temperature rise dynamic prediction model is established in combination with a recurrent neural network to achieve rapid prediction of the temperature rise of the battery system after changes in operating conditions.

2. The method according to claim 1, characterized in that: Based on the heat generation mechanism of the battery and the actual measurement data, the sources of various heat are determined, mainly including electrochemical reaction heat, Joule heat, polarization heat and side reaction heat; based on the battery cooling and heat transfer mechanism, the battery thermal physical properties and boundary conditions are determined; based on the computational fluid dynamics theory, a battery numerical simulation model is established to calculate the temperature changes of the battery system under different environments and working conditions.

3. The method according to claim 2, characterized in that: The battery numerical simulation model is governed by the law of conservation of mass, momentum, energy and turbulent transport equations. It includes the discretization of continuous problems in the time and space domains and the solution of the system of equations formed after discretization, including different boundary conditions, heat source reaction environment and changes in operating conditions.

4. The method according to claim 1, characterized in that: The temperature acquisition system composed of an infrared thermal imager and a multi-channel temperature tester comprehensively measures the external surface and internal temperature of the battery. The internal temperature is measured by a temperature sensor placed at a fixed position inside the battery and transmitted to the multi-channel temperature tester through a temperature measuring line.

5. The method according to claim 1, characterized in that: Based on the measured temperature, virtual measured temperature and operating parameters of the battery system temperature rise, the key parameters are analyzed and extracted as input feature quantities. A dynamic prediction model of battery system temperature rise is established by combining the recurrent neural network. The optimal hyperparameters are screened and determined to improve the model training speed and prediction accuracy, so as to realize the prediction of battery system temperature changes at low cost.

6. The method according to claim 5, characterized in that: The main features of the training data are extracted based on the principal component analysis method, and the data is preprocessed in combination with the characteristics of the recurrent neural network. The key hyperparameters in the neural network, including the number of time steps, hidden layer parameters, and dropout ratio, are determined to effectively learn the temperature correlation information at different times.

Images