Battery multi-physics field reconstruction method and system based on physical information radial basis function

Abstract

The present invention belongs to the field of physical field reconstruction technology and provides a battery multi-physics field reconstruction method and system based on physical information radial basis functions. Its technical solution is as follows: obtaining multi-physics field measurement data on the battery surface; mapping the obtained multi-physics field measurement data from one-dimensional length coordinates to the coordinate space where the battery surface is located; constructing a structured grid on the battery surface, introducing corresponding physical information according to the characteristics of different physical fields, combining the interpolation weights to obtain a radial basis interpolation function, and interpolating the physical field measurement data mapped to the coordinate space where the battery surface is located through the radial basis interpolation function to obtain the corresponding physical field prediction value at any grid. This multi-physics field reconstruction method has high resolution and physical consistency, solving the problems of traditional methods relying on complex physical models, insufficient interpolation accuracy, and lack of physical constraints.

Description

Technical Field

[0001] The present invention belongs to the technical field of physical field reconstruction, and in particular relates to a battery multi-physical field reconstruction method and system based on physical information radial basis functions. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Physical field reconstruction technology has important applications in battery monitoring and management, particularly in systems such as distributed fiber optic sensors and temperature-pressure thin-film sensors, enabling the measurement and analysis of multiple physical fields, including strain, temperature, and pressure. With increasing demands for battery health monitoring, thermal management, and safety performance, multi-physics field reconstruction technology has garnered widespread attention. However, existing technologies have significant shortcomings.

[0004] First, traditional methods often rely on complex physical models, such as heat conduction or mechanical equations, which require a lot of prior knowledge, have high computational costs, and have poor adaptability to non-uniform boundary conditions. Secondly, traditional reconstruction methods such as cubic interpolation are not accurate enough when dealing with non-uniformly distributed data points, and do not effectively incorporate physical constraints, resulting in distortion of the reconstruction results at the boundaries and a lack of physical consistency. In addition, single-fiber sensors or traditional strain gauges are susceptible to cross-interference between temperature and strain signals, thereby reducing measurement accuracy. At the same time, the reconstruction results of existing technologies often lack intuitive visualization capabilities, further limiting their widespread application in multi-physics field monitoring of battery surfaces. Summary of the Invention

[0005] In order to solve at least one technical problem existing in the above-mentioned background technology, the present invention provides a battery multi-physical field reconstruction method and system based on physical information radial basis function, which improves the measurement accuracy of the battery surface physical field, ensures data reliability, accurately maps the data to the battery space, and realizes efficient reconstruction of discrete data to continuous physical fields. The multi-physical field reconstruction method has high resolution and physical consistency.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A first aspect of the present invention provides a battery multi-physics field reconstruction method based on physical information radial basis function, comprising the following steps:

[0008] Acquire multi-physics measurements of battery surfaces;

[0009] Mapping the acquired multi-physics field measurement data from one-dimensional length coordinates to the coordinate space where the battery surface is located;

[0010] A structured grid is constructed on the battery surface. The corresponding physical information is introduced according to the characteristics of different physical fields. The radial basis interpolation function is obtained by combining the interpolation weights. The physical field measurement data mapped to the coordinate space where the battery surface is located is interpolated through the radial basis interpolation function to obtain the corresponding physical field prediction value at any grid.

[0011] Furthermore, after obtaining the multi-physical field measurement data of the battery surface, the physical field measurement data is also verified.

[0012] Furthermore, when mapping the acquired multi-physics field measurement data from one-dimensional length coordinates to the coordinate space where the battery surface is located, the arranged sensor measurement path is decomposed into several straight line segments and curved line segments, and mapping is performed on the straight line segments and curved line segments respectively.

[0013] Furthermore, mapping the straight line segment and the curved line segment separately includes:

[0014] When the sensor measurement path is a straight line segment, the mapping method includes: combining the coordinates of the start and end points of the straight line segment in the mapping space and the total length of the straight line segment measurement path, and calculating the spatial coordinates of any point on the measurement path by linear interpolation;

[0015] When the sensor measurement path is a curved segment, the mapping method includes: discretizing the curved segment into several points, calculating the length of the straight segment between every two adjacent points, accumulating the sum to obtain the length from the starting point to any segment, querying the straight segment described by the target measurement path length, and calculating the spatial coordinates of any point on the measurement path through linear interpolation on the straight segment.

[0016] Furthermore, corresponding physical information is introduced according to the characteristics of different physical fields, including:

[0017] Dirichlet boundary conditions are enforced by defining a set of additional boundary points on the boundaries of the battery domain and appended to the original multiphysics measurement data set.

[0018] Furthermore, the radial basis interpolation function is:

[0019] ,

[0020] in, Represents the target grid point The estimated values of physical quantities corresponding to different physical fields at , N is the number of raw sensor data points, M is the number of boundary points, is the interpolation weight, is a shape parameter that controls the range of influence of each measurement point.

[0021] A second aspect of the present invention provides a battery multi-physical field reconstruction device based on physical information radial basis functions, comprising:

[0022] A measurement data acquisition module, which is used to obtain multi-physics field measurement data of the battery surface;

[0023] A data mapping module, which is used to map the acquired multi-physics field measurement data from a one-dimensional length coordinate to a coordinate space where the battery surface is located;

[0024] The physical field reconstruction module is used to construct a structured grid on the battery surface, introduce corresponding physical information according to the characteristics of different physical fields, and obtain the radial basis interpolation function by combining the interpolation weights. The physical field measurement data mapped to the coordinate space where the battery surface is located is interpolated through the radial basis interpolation function to obtain the corresponding physical field prediction value at any grid.

[0025] A third aspect of the present invention provides a computer-readable storage medium.

[0026] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the battery multi-physical field reconstruction method based on physical information radial basis function as described above.

[0027] A fourth aspect of the present invention provides a computer device.

[0028] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the battery multi-physical field reconstruction method based on physical information radial basis function as described above are implemented.

[0029] A fifth aspect of the present invention provides a computer device.

[0030] A program product, which is a computer program product, includes a computer program. When the computer program is executed by a processor, it implements the steps in the battery multi-physical field reconstruction method based on physical information radial basis function as described above.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present invention verifies and calibrates the acquired physical field signals to ensure data reliability, and combines the characteristic curve mapping algorithm to accurately map the distributed optical fiber data from one-dimensional length coordinates to the spatial coordinates of the battery surface. It combines physical constraints such as boundary conditions with the radial basis function interpolation algorithm to achieve efficient reconstruction of discrete data into continuous physical fields, solving the problems of traditional methods relying on complex physical models, insufficient interpolation accuracy and lack of physical constraints.

[0033] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0035] Figure 1 This is a flow chart of a battery multi-physics field reconstruction method based on physical information radial basis function provided by an embodiment of the present invention;

[0036] Figure 2 Schematic diagrams of differential configurations of optical fiber sensors provided by embodiments of the present invention; (a) is a schematic diagram of differential configurations of S-shaped optical fiber sensors, and (b) is a schematic diagram of differential configurations of spiral optical fiber sensors;

[0037] Figure 3 Schematic diagram of the principle of a physical information-based radial basis function interpolation algorithm provided by an embodiment of the present invention;

[0038] Figure 4 This is a comparison of optical fiber data and strain gauge data at different charge and discharge rates provided by an embodiment of the present invention; wherein (a) shows the comparison of optical fiber data and strain gauge data at a charge and discharge rate of 0.5C; (b) shows the comparison of optical fiber data and strain gauge data at a charge and discharge rate of 1C; (c) shows the comparison of optical fiber data and strain gauge data at a charge and discharge rate of 1.5C;

[0039] Figure 5 : The spatial mapping results of strain data provided by an embodiment of the present invention; wherein (a) represents the spatial mapping results of strain data at different SOCs at a 1C charge rate, and (b) represents the spatial mapping results of strain data at different SOCs at a 1C discharge rate;

[0040] Figure 6 It is a 2 mm × 2 mm grid node provided in an embodiment of the present invention;

[0041] Figure 7 It is the strain field reconstruction result based on physical information radial basis function provided by an embodiment of the present invention; wherein, (a) represents the strain field reconstruction result of different SOC at 1C charging rate, and (b) represents the strain field reconstruction result of different SOC at 1C discharge rate. DETAILED DESCRIPTION

[0042] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0043] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0044] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0045] As mentioned in the background, traditional physical field reconstruction methods suffer from reliance on complex physical models, insufficient accuracy of traditional reconstruction methods such as cubic interpolation, a lack of physical constraints, and insufficient visualization of reconstruction results. The technical solution of the present invention enables high-precision reconstruction and visualization of multiple physical fields, including strain, temperature, and pressure fields. This paper describes the in-situ monitoring and strain field reconstruction of battery surface strain using distributed optical fiber monitoring of battery surface strain as an example. First, a dual-fiber differential configuration is used to separate temperature and strain signals, improving the accuracy of battery surface strain measurement. Second, strain gauges are used to verify and calibrate the optical fiber data to ensure data reliability. A characteristic curve mapping algorithm is then proposed to accurately map the distributed optical fiber data from one-dimensional length coordinates to two-dimensional spatial coordinates on the battery surface. Finally, physical constraints such as boundary conditions are combined with a radial basis function interpolation algorithm to achieve efficient reconstruction of discrete data into continuous physical fields. It should be noted that this method is also applicable to reconstructing battery temperature, pressure, or other physical fields based on other sensors. This multi-physics field reconstruction method, with its high resolution and physical consistency, is suitable for data processing, data enhancement, and visualization of distributed optical fiber sensing, temperature and pressure thin-film sensors, and other sensing systems in scenarios such as battery monitoring, thermal management, and health assessment.

[0046] Example 1

[0047] like Figure 1 As shown, this embodiment provides a battery multi-physics field reconstruction method based on physical information radial basis function, including the following steps:

[0048] Step 1: Acquire the first multi-physics field measurement data of the battery surface;

[0049] In this embodiment, the multi-physical measurement data of the battery surface may be multi-physical field measurement data such as strain field, temperature field or pressure field;

[0050] Taking the acquisition of strain field data on the surface of a soft-pack battery as an example, the process of acquiring strain signals is explained, which specifically includes:

[0051] In order to overcome the limitation of a single optical fiber sensor that is difficult to separate temperature and strain signals, this embodiment adopts a dual-fiber differential configuration to improve the accuracy of battery surface strain measurement. Figure 2 As shown in Figure 1, the configuration includes a reference fiber and a sensing fiber. The reference fiber is encapsulated in a perfluoroalkoxy tube, shielded from mechanical deformation by the tube wall and only sensing ambient temperature changes. The sensing fiber remains exposed and can simultaneously sense strain and temperature on the battery surface.

[0052] Taking soft-pack batteries as an example, optical fibers can be arranged in a variety of ways according to monitoring requirements: S-shaped arrangement, such as Figure 2 As shown in (a), the key positions on the battery surface are evenly covered through a simple circuitous path, which is suitable for large-area strain distribution monitoring; the spiral arrangement, such as Figure 2 As shown in (b), the battery surface is covered in a more tightly coiled manner, which is suitable for scenes with complex shapes or high precision requirements.

[0053] Regardless of the layout used, the reference fiber and the sensing fiber must be tightly fixed to the battery surface by parallel bonding, with the spacing controlled within a small range (usually less than 1 mm) to ensure that the sensing points of the two are located in the same area of the battery surface as much as possible.

[0054] This arrangement ensures that the reference and sensing fibers maintain the same sensing environment, such as the same ambient temperature and air flow conditions, ensuring the accuracy of differential processing. By comparing the signals from the reference and sensing fibers, differential processing effectively offsets temperature-induced signal variations, accurately extracting pure strain data during battery operation and providing high-quality raw monitoring data for subsequent multi-physics field reconstruction.

[0055] It should be noted that this embodiment is not only applicable to the measurement of strain field data on the surface of soft-pack batteries. In other embodiments, sensor solutions corresponding to temperature fields, pressure fields or other physical fields can be arranged on the surfaces of other types of batteries in the same manner to measure the corresponding physical fields.

[0056] Step 2: Acquire second multi-physics field measurement data on the battery surface, compare the first measurement data and the second measurement data at the same measurement position, calibrate the first measurement data according to the second measurement data, and obtain calibrated multi-physics field measurement data;

[0057] In order to ensure the reliability and accuracy of the acquired first measurement data and overcome the influence of environmental noise, installation deviation or sensor drift on the measurement result, the first measurement data can be verified using a traditional point sensor.

[0058] Still taking strain data verification as an example, the specific verification process is explained, which includes:

[0059] In order to ensure the reliability and accuracy of distributed fiber optic sensor data and overcome the influence of environmental noise, installation deviation or sensor drift on the measurement results, the fiber optic data can be verified by strain gauges and thermocouples. This step can provide reliable spatially continuous raw data for the reconstruction of multi-physical fields on the battery surface. Figure 2 As shown, the present invention arranges three strain gauges at key positions on the battery surface, which are located in the high strain area at the center of the battery and the high temperature area near the tab.

[0060] It should be noted that the number and position of strain gauges are flexibly determined according to the battery shape and strain distribution characteristics, and are usually evenly distributed to cover representative areas.

[0061] Strain gauges directly measure local deformation, acquiring independent, high-precision strain data. This data is then cross-validated with the strain measurements from fiber optic sensors, comparing the strain values at the same location to identify potential deviations in the fiber optic data. If errors are detected, the fiber optic strain values are calibrated against the strain gauge data to eliminate measurement errors and ensure data consistency.

[0062] It should be noted that the above principles can still be used for correction when reconstructing other physical fields. For example, when reconstructing the battery temperature field, a certain number of thermocouples are arranged on the battery surface, preferentially covering areas with significant temperature changes, such as the battery center and areas near the tabs, so as to compare with the temperature data monitored by optical fiber and verify the accuracy of optical fiber temperature measurement.

[0063] Step 3: Map the calibrated multi-physics field measurement data from the one-dimensional length coordinate to the coordinate space where the battery surface is located;

[0064] In this embodiment, a characteristic curve mapping algorithm is proposed. The core idea of this mapping algorithm is to decompose the arranged sensor measurement path into several straight line segments and curved line segments, and use mathematical methods to map the multi-physics field measurement data from one-dimensional length coordinates to the battery surface coordinate space;

[0065] Specifically, when the sensor measurement path is a straight line segment, the mapping method includes: combining the coordinates of the start and end points of the straight line segment in the mapping space and the total length of the straight line segment measurement path, and calculating the spatial coordinates of any point on the measurement path by linear interpolation;

[0066] When the sensor measurement path is a curved segment, the mapping method includes: discretizing the curved segment into a number of points, calculating the length of the straight segment between each two adjacent points, accumulating the sum to obtain the length from the starting point to any segment, querying the straight segment described by the target measurement path length, and obtaining the spatial coordinates of any point on the measurement path by linear interpolation on the straight segment;

[0067] Still taking the strain field data of the soft-pack battery surface as an example, the specific mapping process is explained:

[0068] The object of mapping at this time is to map the original measurement data from one-dimensional length coordinates to two-dimensional space coordinates on the battery surface. In order to accurately map the original measurement data from one-dimensional length coordinates to two-dimensional space coordinates on the battery surface,

[0069] Taking the fiber optic sensor as an example, when the fiber optic path is a straight line segment, its mapping process can be realized by interpolation using the linear interpolation formula. Assume that the coordinates of the starting point A and the end point B of the fiber optic straight line segment in two-dimensional space are and , the total length of the optical fiber along this straight segment is , the cumulative length corresponding to any point is l (in ). The two-dimensional spatial coordinates of any point on the optical fiber can be calculated through the linear interpolation formula. , expressed as:

[0070] (1),

[0071] (2),

[0072] The characteristic curve mapping algorithm described above can efficiently map the fiber length coordinates into a two-dimensional space, and is suitable for relatively regular portions of the fiber path.

[0073] For the curve segment in the optical fiber path, this embodiment proposes a discrete-cumulative interpolation mapping method. First, the curve segment is discretized into several points, whose two-dimensional coordinates are known and recorded as (in i =1, 2, …, n). Next, calculate the length of the straight line segment between each two adjacent points:

[0074] (3),

[0075] Then, the cumulative summation is obtained from the starting point to the i Total length of the segment:

[0076] (4),

[0077] For any given fiber length , by finding Index , determine the straight line segment to which it belongs. Apply the linear interpolation formula on the straight line segment to calculate the corresponding two-dimensional space coordinates and According to the principle of linear interpolation, the coordinates of any point on the optical fiber can be calculated by the following formula:

[0078] (5),

[0079] (6),

[0080] Theoretically, the discrete-accumulation interpolation mapping method is applicable to curves of arbitrary shapes, and its mapping accuracy is directly related to the number of discrete points. In practice, the number and distribution of discrete points can be flexibly adjusted according to the required accuracy.

[0081] It should be noted that the mapping principle of the above data is also applicable to other battery types and other types of physical field data.

[0082] Step 4: Construct a structured grid on the battery surface, introduce corresponding physical information based on the characteristics of different physical fields, and combine the interpolation weights to obtain the radial basis function. The physical field measurement data mapped to the coordinate space of the battery surface is interpolated through the radial basis function to obtain the corresponding physical field prediction value at any grid;

[0083] To reconstruct the distribution of physical fields from discrete measurement data, this example employs an interpolation method based on physical information radial basis functions. This method not only effectively addresses the uneven spatial distribution of discrete data points but also incorporates physical constraints (such as boundary conditions and constitutive equations) to ensure the accuracy and physical consistency of the reconstructed results.

[0084] First, if Figure 3 As shown in the figure, a high-resolution structured grid is constructed on the battery surface as the basic framework for physical field reconstruction. The grid nodes represent the locations that need to be estimated, thereby achieving a seamless transition from discrete sensor readings to continuous physical fields.

[0085] It is understandable that the size of the grid can be determined according to the spatial resolution of the sensor and the requirements of the actual application scenario.

[0086] Subsequently, the radial basis function interpolation technique is used to map the discrete data points onto the structured grid. The core idea of radial basis function interpolation is to estimate the value of the physical quantity corresponding to the physical field at the target grid point by weighted summation. Its mathematical expression is:

[0087] (7),

[0088] Where, is the number of known strain data points, is the interpolation weight, is a radial basis function. In this embodiment, a Gaussian function is selected:

[0089] (8),

[0090] Where, is a shape parameter used to control the influence range of each measurement point. The value of can flexibly adjust the smoothness of the interpolation result. Indicates the target grid point and the i The Euclidean distance between the measured points:

[0091] (9),

[0092] Interpolation weight The solution can be obtained by establishing a linear equation system based on the known strain data points. The specific solution process includes:

[0093] For each known data point , satisfying the relationship:

[0094] (10),

[0095] After substituting all the data points, the N The interpolation weights can be obtained by solving a linear system of equations ω i Once the weights are determined, the strain value at any grid point can be calculated using the radial basis function interpolation formula.

[0096] However, the traditional radial basis function interpolation method does not consider physical constraints, which may affect the accuracy of the reconstruction result. Therefore, this embodiment introduces physical information based on the characteristics of the physical field to improve the radial basis function algorithm.

[0097] If it is a strain field, such as Figure 3 As shown in the figure, a set of additional boundary points are defined on the boundary of the battery domain to enforce Dirichlet boundary conditions and append them to the original data set. Specifically, assuming that the strain value on the battery boundary is zero (i.e., the strain at the fixed boundary is zero), the boundary locations (such as x =0, x = L cell , y =0, y = H cell ) adds a set of dummy data points with strain values set to zero.

[0098] If it is a temperature field, the physical consistency of the radial basis function interpolation algorithm can be enhanced by introducing the heat conduction equation. Specifically, by introducing the steady-state heat conduction control equation in the battery domain, for example, in the two-dimensional case, (in k is the thermal conductivity, T The boundary condition treatment is similar to adding a virtual boundary point with a fixed strain value in the strain field. x =0, x= L cell , y =0, y = H cell ) introduces a set of additional boundary points and imposes Dirichlet boundary conditions (such as fixed boundary temperature or adiabatic boundary conditions) as additional constraints on the original temperature dataset. This strategy allows the RBF interpolation process to more accurately capture the spatial variations of the temperature field while maintaining numerical smoothness, effectively reconstructing sparse temperature data into a high-precision full-field temperature distribution.

[0099] The improved interpolation formula can be expressed as:

[0100] (11),

[0101] Where, Represents the target grid point Estimated values of physical quantities corresponding to different physical fields at , such as strain values, temperature values and other physical field measurement data, N is the number of raw sensor data points, M is the number of boundary points, interpolation weight Determined by sensor point data and extended data sets.

[0102] For the temperature field and other physical fields, formula (11) is still applicable, and this embodiment does not list the interpolation formulas of other physical fields one by one.

[0103] Through this method, the reconstructed physical field not only accurately reflects the discrete data characteristics acquired by the sensor, but also ensures that the results meet the constraints of the actual physical scenario, significantly improving the accuracy of strain field reconstruction. Furthermore, because the grid size can be flexibly adjusted based on the sensor's spatial resolution and specific needs, this method can adapt to various operating conditions and capture the fine features of the battery surface strain distribution.

[0104] The strain field and temperature field reconstructed by the present invention can be used to support multi-dimensional state estimation and failure warning of batteries. Among them, the temperature field can be used to identify local hot spots, monitor heat diffusion paths, and realize early warning of thermal runaway risks; the strain field can be used to capture structural response behaviors such as electrode expansion and shell deformation, and assist in judging the degree of internal mechanical damage or structural degradation. In specific applications, the two types of physical field information can be used independently for their respective functional modules, or they can be jointly constructed into a thermal-mechanical coupling analysis model. By analyzing the distribution of thermal expansion strain caused by temperature changes, potential high-risk areas can be further identified. In addition, the temperature-strain joint feature can also be used as input for data-driven state estimation algorithms to improve the comprehensive prediction capabilities of battery health status, state of charge, thermal safety boundaries, and remaining life.

[0105] To verify the effectiveness of the proposed physical field reconstruction method based on physical information radial basis functions, a specific example is provided below. This example uses a soft-pack battery as the research object and describes in detail the entire process from optical fiber data acquisition to strain field reconstruction.

[0106] like Figure 4 As shown, distributed optical fiber sensors are laid out in an S-shaped arrangement on the surface of the soft-pack battery, and the battery size is 318 mm × 96 mm × 12 mm. The reference optical fiber is encapsulated in a perfluoroalkoxy tube to shield the influence of mechanical deformation and only sense changes in ambient temperature; the sensing optical fiber remains exposed and can simultaneously sense the strain and temperature on the battery surface. The reference optical fiber and the sensing optical fiber are tightly fixed to the battery surface by parallel bonding, and the spacing is controlled to be less than 1 mm to ensure that the sensing environment of the two is as consistent as possible. In addition, to ensure the reliability and accuracy of the distributed optical fiber sensor data, strain gauges are used to verify and calibrate the optical fiber data. As shown Figure 4 As shown, three strain gauges are arranged at key positions on the battery surface.

[0107] The battery operating conditions were set at an ambient temperature of 26°C, with constant current and constant voltage charging and constant current discharge, followed by a three-hour rest period after each charge and discharge cycle. Charge and discharge rates were 0.5°C, 1°C, and 1.5°C. Under these operating conditions, the strain values obtained from the strain gauge and fiber optic sensor at the same location were compared to identify potential deviations in the fiber optic data. The strain gauge data was then used as a benchmark for verification and calibration of the fiber optic data. Similarly, the fiber optic temperature value was calibrated using thermocouple data to correct for potential errors (such as offsets caused by sensor drift or environmental interference).

[0108] Figure 4Comparisons of strain gauge and fiber optic data are shown. (a) shows the comparison of fiber optic data and strain gauge data at a 0.5C charge / discharge rate; (b) shows the comparison of fiber optic data and strain gauge data at a 1C charge / discharge rate; and (c) shows the comparison of fiber optic data and strain gauge data at a 1.5C charge / discharge rate. It can be seen that the temperature and strain data measured by the distributed fiber optic sensor agree well with those measured by traditional sensors. This demonstrates the high accuracy and reliability of the distributed fiber optic monitoring system, making it an effective alternative to traditional point sensors for continuous spatial physical field monitoring.

[0109] In order to achieve accurate mapping of distributed optical fiber strain measurement data to the spatial coordinates of the battery surface, a characteristic curve segmentation mapping algorithm is used. The curve is split into straight segments and curved segments according to the layout shape of the optical fiber, and the optical fiber length coordinates are converted into two-dimensional spatial coordinates through mathematical methods. Figure 5 As shown in the figure, taking 1C charging and discharging as an example, the optical fiber data is accurately mapped to the spatial coordinates of the battery surface, where (a) represents the spatial mapping results of strain data at different SOCs at a 1C charging rate, and (b) represents the spatial mapping results of strain data at different SOCs at a 1C discharge rate.

[0110] After completing the spatial mapping of the optical fiber data, the strain field on the battery surface is reconstructed using a physical information-based radial basis function method. The specific steps are as follows:

[0111] (1) Grid construction: According to the spatial resolution of the fiber optic sensor and actual requirements, a high-resolution structured grid is constructed. Figure 6 As shown, in this embodiment, the grid size is set to 2 mm × 2 mm, covering the entire battery surface.

[0112] (2) Boundary condition definition: In order to ensure that the reconstruction results meet the constraints in the actual physical scene, boundary conditions are introduced as additional physical information. Figure 6 As shown, at the boundary position ( x = 0 mm, x = 96 mm, y = 0 mm, y = 318 mm), and a set of virtual data points is added with their strain values set to zero. In this example, the number of boundary points on each edge is set to 50. These boundary points are merged with the original measured data to form an extended data set.

[0113] (3) Data interpolation: After combining the fiber measurement data and the boundary point data, the Gaussian radial basis function is used for interpolation calculation. According to the data distribution characteristics, the shape parameters are set to ε is 10 mm. The final strain field distribution is as follows Figure 7As shown in the figure, (a) represents the strain field reconstruction results of different SOCs at a 1C charge rate, and (b) represents the strain field reconstruction results of different SOCs at a 1C discharge rate, which clearly reflects the fine characteristics of the strain distribution on the battery surface.

[0114] The multi-physics field reconstruction method of the present invention has high resolution and physical consistency, and is suitable for data processing, data enhancement and visualization of distributed optical fiber sensing, temperature and pressure thin film sensors and other sensing systems in scenarios such as battery monitoring, thermal management and health status assessment.

[0115] Example 2

[0116] This embodiment provides a battery multi-physics field reconstruction system based on physical information radial basis functions, including:

[0117] A measurement data acquisition module, which is used to obtain multi-physics field measurement data of the battery surface;

[0118] A data mapping module, which is used to map the acquired multi-physics field measurement data from a one-dimensional length coordinate to a coordinate space where the battery surface is located;

[0119] The physical field reconstruction module is used to construct a structured grid on the battery surface, introduce corresponding physical information according to the characteristics of different physical fields, and obtain the radial basis interpolation function by combining the interpolation weights. The physical field measurement data mapped to the coordinate space where the battery surface is located is interpolated through the radial basis interpolation function to obtain the corresponding physical field prediction value at any grid.

[0120] It should be noted that the specific implementation method of the battery multi-physical field reconstruction system based on physical information radial basis function in the embodiment of the present invention is similar to the specific implementation method of the battery multi-physical field reconstruction method based on physical information radial basis function in the embodiment of the present invention. Please refer to the description of the method part for details. In order to reduce redundancy, it will not be repeated here.

[0121] Example 3

[0122] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps in the battery multi-physical field reconstruction method based on physical information radial basis function as described above are implemented.

[0123] Example 4

[0124] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the battery multi-physical field reconstruction method based on physical information radial basis function as described above are implemented.

[0125] Example 5

[0126] This embodiment provides a program product, which is a computer program product, including a computer program. When the computer program is executed by a processor, it implements the steps in the battery multi-physical field reconstruction method based on physical information radial basis function as described above.

[0127] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A battery multi-physics field reconstruction method based on physical information radial basis function, characterized in that: The steps include: Acquire multi-physics measurements of battery surfaces; Mapping the acquired multi-physics field measurement data from one-dimensional length coordinates to the coordinate space where the battery surface is located; When mapping the acquired multi-physics field measurement data from the one-dimensional length coordinate to the coordinate space where the battery surface is located, the arranged sensor measurement path is decomposed into a plurality of straight line segments and curved line segments, and mapping is performed on the straight line segments and the curved line segments respectively; The mapping is performed on the straight line segment and the curve segment respectively, including: When the sensor measurement path is a straight line segment, the mapping method includes: combining the coordinates of the start and end points of the straight line segment in the mapping space and the total length of the straight line segment measurement path, and calculating the spatial coordinates of any point on the measurement path by linear interpolation; When the sensor measurement path is a curved segment, the mapping method includes: discretizing the curved segment into a number of points, calculating the length of the straight segment between each two adjacent points, accumulating the sum to obtain the length from the starting point to any segment, querying the straight segment described by the target measurement path length, and obtaining the spatial coordinates of any point on the measurement path by linear interpolation on the straight segment; A structured grid is constructed on the battery surface. The corresponding physical information is introduced according to the characteristics of different physical fields. The radial basis interpolation function is obtained by combining the interpolation weights. The physical field measurement data mapped to the coordinate space where the battery surface is located is interpolated through the radial basis interpolation function to obtain the corresponding physical field prediction value at any grid.

2. The battery multi-physical field reconstruction method based on physical information radial basis function according to claim 1, characterized in that: After obtaining the multi-physics field measurement data of the battery surface, the physical field measurement data is also verified.

3. The battery multi-physical field reconstruction method based on physical information radial basis function according to claim 1, characterized in that: According to the characteristics of different physical fields, corresponding physical information is introduced, including: Dirichlet boundary conditions are enforced by defining a set of additional boundary points on the boundaries of the battery domain and appended to the original multiphysics measurement data set.

4. The battery multi-physics field reconstruction method based on physical information radial basis function according to claim 1, characterized in that: The radial basis interpolation function is: ， in, Represents the target grid point The estimated values of physical quantities corresponding to different physical fields at , N is the number of raw sensor data points, M is the number of boundary points, is the interpolation weight, is a shape parameter that controls the range of influence of each measurement point.

5. A battery multi-physics field reconstruction system based on physical information radial basis function, characterized by: include: A measurement data acquisition module, which is used to obtain multi-physics field measurement data of the battery surface; A data mapping module, which is used to map the acquired multi-physics field measurement data from a one-dimensional length coordinate to a coordinate space where the battery surface is located; When mapping the acquired multi-physics field measurement data from the one-dimensional length coordinate to the coordinate space where the battery surface is located, the arranged sensor measurement path is decomposed into a plurality of straight line segments and curved line segments, and mapping is performed on the straight line segments and the curved line segments respectively; The mapping is performed on the straight line segment and the curve segment respectively, including: When the sensor measurement path is a straight line segment, the mapping method includes: combining the coordinates of the start and end points of the straight line segment in the mapping space and the total length of the straight line segment measurement path, and calculating the spatial coordinates of any point on the measurement path by linear interpolation; When the sensor measurement path is a curved segment, the mapping method includes: discretizing the curved segment into a number of points, calculating the length of the straight segment between each two adjacent points, accumulating the sum to obtain the length from the starting point to any segment, querying the straight segment described by the target measurement path length, and obtaining the spatial coordinates of any point on the measurement path by linear interpolation on the straight segment; The physical field reconstruction module is used to construct a structured grid on the battery surface, introduce corresponding physical information according to the characteristics of different physical fields, and obtain the radial basis interpolation function by combining the interpolation weights. The physical field measurement data mapped to the coordinate space where the battery surface is located is interpolated through the radial basis interpolation function to obtain the corresponding physical field prediction value at any grid.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the battery multi-physical field reconstruction method based on physical information radial basis function as described in any one of claims 1 to 4 are implemented.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps in the battery multi-physical field reconstruction method based on physical information radial basis function are implemented as described in any one of claims 1 to 4.

8. A program product, wherein the program product is a computer program product, comprising a computer program, characterized in that: When the computer program is executed by a processor, the steps in the battery multi-physical field reconstruction method based on physical information radial basis function as described in any one of claims 1 to 4 are implemented.

Images