Method and System for Constructing Digital Twin Models of Battery Power Systems for New Energy Heavy Trucks

By constructing a multi-scale image model and using multi-physics coupling and real-time data-driven methods, the problem of insufficient accuracy in existing models of new energy heavy-duty truck battery power systems is solved, and high-precision dynamic prediction and management are achieved.

CN120597700BActive Publication Date: 2026-11-17INNER MONGOLIA UNIV OF TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510684245.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-05-26
Publication Date
2026-11-17
Estimated Expiration
2045-05-26

AI Technical Summary

Technical Problem

Existing digital twin models cannot accurately reflect the complex characteristics of the battery power system of new energy heavy trucks in actual operation, resulting in insufficient model prediction accuracy and system control performance.

Method used

A multi-scale imaging model of the battery power system is constructed, and multi-physics coupling modeling of electrochemical field, temperature field and stress field is carried out. The parameters are updated by real-time monitoring data, and dynamic time series modeling is carried out by graph neural network-long short-term memory neural network structure.

Benefits of technology

It achieves high-precision mapping and dynamic prediction of the operating status of the battery power system of new energy heavy trucks, enhances the adaptability and predictive ability of the model, and provides a reliable digital foundation for the full life cycle management of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120597700B_ABST
    Figure CN120597700B_ABST
Patent Text Reader

Abstract

This invention discloses a method and system for constructing a digital twin model of a battery power system for new energy heavy-duty trucks. The method includes: constructing a multi-scale image model of the battery power system, including hierarchical models from individual battery cells to modules, the entire battery pack, controllers, and drive motors; performing multi-physics field coupling modeling of the constructed multi-scale image model with electrochemical, temperature, and stress fields to obtain a coupled multi-scale image model; updating the parameters of the coupled multi-scale image model based on real-time monitoring data using a data-driven model to generate a preliminary digital twin model; and performing dynamic time-series modeling of the sensor data in the preliminary digital twin model using a graph neural network-long short-term memory neural network structure to obtain the final digital twin model. This invention can accurately reflect the complex characteristics of the battery power system of new energy heavy-duty trucks in actual operation, effectively improving the model prediction accuracy and system control performance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and in particular to a method and system for constructing a digital twin model of a new energy heavy truck battery power system. Background Technology

[0002] In the field of digital twin modeling of battery power systems for new energy heavy-duty trucks, traditional methods are mainly based on single-physics models and offline parameter calibration. A typical approach is to first establish an electrochemical model (e.g., P2D model) or a thermal model (e.g., lumped parameter model) for each battery cell, then obtain parameter values ​​under finite operating conditions through bench experiments, and apply these values ​​to system-level simulations. For example, some studies only construct a temperature field model of the battery thermal management system, simulating the coolant flow path through fluid dynamics equations, but neglect the reaction of temperature changes to battery internal resistance and electrochemical processes. This method has significant limitations in practical applications: on the one hand, a single-physics model cannot reflect the coupling effect of the electrochemical-thermal-stress field in the battery system; for example, thermal stress caused by temperature gradients can lead to electrode structural deformation, thus affecting battery internal resistance and charge / discharge efficiency. On the other hand, offline-calibrated static parameters cannot adapt to dynamic changes such as battery aging and operating condition fluctuations, resulting in a significant increase in model prediction errors over time.

[0003] Therefore, existing digital twin models cannot accurately reflect the complex characteristics of the battery power system of new energy heavy trucks in actual operation, resulting in insufficient model prediction accuracy and system control performance. Summary of the Invention

[0004] This invention provides a method and system for constructing a digital twin model of a new energy heavy-duty truck battery power system, which can accurately reflect the complex characteristics of the new energy heavy-duty truck battery power system in actual operation and effectively improve the model prediction accuracy and system control performance.

[0005] An embodiment of the present invention provides a method for constructing a digital twin model of a battery power system for new energy heavy-duty trucks, comprising:

[0006] Construct a multi-scale imaging model of the battery power system, including a hierarchical model from individual battery cells to modules, the entire pack, controllers, and drive motors;

[0007] The constructed multi-scale imaging model is coupled with electrochemical field, temperature field and stress field to obtain the coupled multi-scale imaging model.

[0008] Based on real-time monitoring data, a data-driven model is used to update the parameters of the coupled multi-scale imaging model and generate a preliminary digital twin model.

[0009] The sensor data in the preliminary digital twin model is dynamically and temporally modeled using a graph neural network-long short-term memory neural network structure to obtain the final digital twin model.

[0010] As an improvement to the above scheme, the construction of a multi-scale imaging model of the battery power system, including a hierarchical model from individual battery cells to modules, the entire battery pack, controllers, and drive motors, includes the following sub-steps:

[0011] Based on the energy conservation equation, an electrochemical-thermal model at the cell scale is constructed on top of the P2D model, resulting in a model structure that integrates electrochemical and thermal properties at the cell scale.

[0012] Based on the cell-scale basic model, considering the thermal stress and thermal expansion caused by temperature gradient during battery operation, a single-cell-scale thermo-mechanical coupling model is constructed to form a comprehensive model that includes cell and single-cell-scale characteristics.

[0013] Based on the comprehensive model, starting from the power battery module and the whole pack scale, a three-dimensional CAD model corresponding to the real shape of the physical battery is established using industrial modeling software. This accurately presents key structural parameters, constraints and positioning relationships between individual cells and modules, and obtains a three-dimensional model containing multi-scale structural information.

[0014] Pixyz Studio was used to perform surface subdivision, topology correction, extraction, and repair operations on a 3D model containing multi-scale structural information. Then, combined with Hertz theory and extended finite element method, an overall digital physical model was established, the semantic information annotation objects were clarified, and the construction of a multi-scale image model of the battery power system was completed.

[0015] As an improvement to the above scheme, the step of performing multi-physics coupling modeling of the constructed multi-scale imaging model with electrochemical field, temperature field, and stress field to obtain the coupled multi-scale imaging model includes the following sub-steps:

[0016] For the completed multi-scale imaging model of the battery power system, we comprehensively sort out the parameter information related to the electrochemical field, temperature field and stress field, and obtain the parameter set of each physical field at different scales;

[0017] Based on the parameter set, we analyze the transmission paths and methods of each physical field parameter between different levels of models, clarify the rules of parameter transmission, and obtain a specific description of the rules of parameter transmission.

[0018] Based on the obtained parameter transfer rule description, the interaction mechanism between different physical fields and different level models is studied, the influencing factors of the interaction are determined, and a list of key factors affecting the interaction is obtained.

[0019] Based on the parameter set, parameter transfer rule description, and list of key factors affecting interaction, the multi-scale imaging model is coupled and integrated with multi-physics fields to obtain the coupled multi-scale imaging model.

[0020] As an improvement to the above scheme, the step of updating the parameters of the coupled multi-scale imaging model based on real-time monitoring data and using a data-driven model to generate a preliminary digital twin model includes the following sub-steps:

[0021] By pre-deploying sensors in the battery power system, real-time monitoring data during system operation is collected to obtain a dataset containing the real-time status of the battery power system.

[0022] Data-driven methods are used to extract information from datasets containing the real-time status of battery power systems, and effective information related to the parameters of multi-scale imaging models is filtered out to form an effective information set.

[0023] The information in the effective information set is compared and matched with the parameters of the coupled multi-scale imaging model to determine the parameters that need to be updated and their update values, thus obtaining a parameter update list.

[0024] Based on the parameter update list, the parameters of the coupled multi-scale imaging model are updated to generate a preliminary digital twin model.

[0025] As an improvement to the above scheme, the step of dynamically and temporally modeling the sensor data in the preliminary digital twin model using a graph neural network-long short-term memory neural network structure to obtain the final digital twin model includes the following sub-steps:

[0026] The sensor data in the preliminary digital twin model is abstracted into a network topology diagram of the sensor data time-series accumulation process according to the time series and spatial relationship, so as to obtain a topology structure that reflects the spatiotemporal characteristics of sensor data;

[0027] A graph neural network-long short-term memory neural network structure is used to process the topological structure that reflects the spatiotemporal characteristics of sensor data, and integrate the structural information and temporal information to form intermediate data that fuses spatiotemporal characteristics.

[0028] By utilizing the designed hierarchical feature aggregation method, different feature aggregators are learned on intermediate data with fused spatiotemporal features at different depths of neighborhood, thereby obtaining aggregated data containing multi-scale features.

[0029] Dynamic time-series modeling of sensor data is performed, and aggregated data containing multi-scale features is incorporated into the model to obtain the final digital twin model.

[0030] Another embodiment of the present invention provides a digital twin model construction system for a new energy heavy-duty truck battery power system, comprising:

[0031] The building module is used to construct a multi-scale image model of the battery power system, including a hierarchical model from individual battery cells to modules, the entire pack, the controller, and the drive motor.

[0032] The first modeling module is used to perform multi-physics field coupling modeling of electrochemical field, temperature field and stress field on the constructed multi-scale image model to obtain the coupled multi-scale image model.

[0033] The update module is used to update the parameters of the coupled multi-scale imaging model based on real-time monitoring data and a data-driven model to generate a preliminary digital twin model.

[0034] The second modeling module is used to perform dynamic time-series modeling of sensor data in the preliminary digital twin model using a graph neural network-long short-term memory neural network structure, in order to obtain the final digital twin model.

[0035] As an improvement to the above solution, the building module is specifically used for:

[0036] Based on the energy conservation equation, an electrochemical-thermal model at the cell scale is constructed on top of the P2D model, resulting in a model structure that integrates electrochemical and thermal properties at the cell scale.

[0037] Based on the cell-scale basic model, considering the thermal stress and thermal expansion caused by temperature gradient during battery operation, a single-cell-scale thermo-mechanical coupling model is constructed to form a comprehensive model that includes cell and single-cell-scale characteristics.

[0038] Based on the comprehensive model, starting from the power battery module and the whole pack scale, a three-dimensional CAD model corresponding to the real shape of the physical battery is established using industrial modeling software. This accurately presents key structural parameters, constraints and positioning relationships between individual cells and modules, and obtains a three-dimensional model containing multi-scale structural information.

[0039] Pixyz Studio was used to perform surface subdivision, topology correction, extraction, and repair operations on a 3D model containing multi-scale structural information. Then, combined with Hertz theory and extended finite element method, an overall digital physical model was established, the semantic information annotation objects were clarified, and the construction of a multi-scale image model of the battery power system was completed.

[0040] As an improvement to the above solution, the first modeling module is specifically used for:

[0041] For the completed multi-scale imaging model of the battery power system, we comprehensively sort out the parameter information related to the electrochemical field, temperature field and stress field, and obtain the parameter set of each physical field at different scales;

[0042] Based on the parameter set, we analyze the transmission paths and methods of each physical field parameter between different levels of models, clarify the rules of parameter transmission, and obtain a specific description of the rules of parameter transmission.

[0043] Based on the obtained parameter transfer rule description, the interaction mechanism between different physical fields and different level models is studied, the influencing factors of the interaction are determined, and a list of key factors affecting the interaction is obtained.

[0044] Based on the parameter set, parameter transfer rule description, and list of key factors affecting interaction, the multi-scale imaging model is coupled and integrated with multi-physics fields to obtain the coupled multi-scale imaging model.

[0045] As an improvement to the above solution, the update module is specifically used for:

[0046] By pre-deploying sensors in the battery power system, real-time monitoring data during system operation is collected to obtain a dataset containing the real-time status of the battery power system.

[0047] Data-driven methods are used to extract information from datasets containing the real-time status of battery power systems, and effective information related to the parameters of multi-scale imaging models is filtered out to form an effective information set.

[0048] The information in the effective information set is compared and matched with the parameters of the coupled multi-scale imaging model to determine the parameters that need to be updated and their update values, thus obtaining a parameter update list.

[0049] Based on the parameter update list, the parameters of the coupled multi-scale imaging model are updated to generate a preliminary digital twin model.

[0050] As an improvement to the above solution, the second modeling module is specifically used for:

[0051] The sensor data in the preliminary digital twin model is abstracted into a network topology diagram of the sensor data time-series accumulation process according to the time series and spatial relationship, so as to obtain a topology structure that reflects the spatiotemporal characteristics of sensor data;

[0052] A graph neural network-long short-term memory neural network structure is used to process the topological structure that reflects the spatiotemporal characteristics of sensor data, and integrate the structural information and temporal information to form intermediate data that fuses spatiotemporal characteristics.

[0053] By utilizing the designed hierarchical feature aggregation method, different feature aggregators are learned on intermediate data with fused spatiotemporal features at different depths of neighborhood, thereby obtaining aggregated data containing multi-scale features.

[0054] Dynamic time-series modeling of sensor data is performed, and aggregated data containing multi-scale features is incorporated into the model to obtain the final digital twin model.

[0055] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0056] First, a multi-scale mapping model from individual battery cells to the drive motor is constructed, comprehensively covering all levels of the system through a hierarchical architecture to ensure accurate representation of characteristics from the microscopic to the macroscopic. Next, multi-physics coupled modeling of electrochemical, temperature, and stress fields is performed, changing the traditional single-physics modeling mode and recreating the real-world interaction of various physical fields during system operation, enhancing the model's simulation capability for complex operating conditions. Then, real-time monitoring data is used to update the parameters of the coupled multi-scale mapping model through data-driven modeling, enabling the model to respond in real-time to dynamic factors such as battery aging and changes in operating conditions, maintaining synchronization with the actual system. Finally, a graph neural network-long short-term memory neural network structure is used; the former mines the spatial correlation of sensor data, and the latter captures time-series features, achieving spatiotemporal dynamic analysis of the system's operating state. Compared to existing technologies with problems such as isolated multi-scale modeling, insufficient physical field coupling, data-driven lag, and limitations of time-series modeling, this invention, through hierarchical multi-scale modeling, deep physical field coupling, dynamic data-driven approach, and spatiotemporal feature mining, achieves high-precision mapping and dynamic prediction of the operating state of the new energy heavy-duty truck battery power system, providing a solid and reliable digital foundation for efficient management and optimization throughout the system's entire lifecycle. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating a method for constructing a digital twin model of a new energy heavy-duty truck battery power system according to an embodiment of the present invention;

[0058] Figure 2 This is a schematic diagram of the structure of a digital twin model construction system for a new energy heavy truck battery power system provided in an embodiment of the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] See Figure 1 This is a flowchart illustrating a method for constructing a digital twin model of a new energy heavy-duty truck battery power system according to an embodiment of the present invention. The method for constructing a digital twin model of a new energy heavy-duty truck battery power system includes the following steps:

[0061] S10, construct a multi-scale image model of the battery power system, including a hierarchical model from individual battery cells to modules, packs, controllers, and drive motors;

[0062] S11, perform multi-physics coupling modeling of electrochemical field, temperature field and stress field on the constructed multi-scale imaging model to obtain the coupled multi-scale imaging model;

[0063] S12, based on real-time monitoring data, a data-driven model is used to update the parameters of the coupled multi-scale imaging model to generate a preliminary digital twin model;

[0064] S13, Dynamic temporal modeling of sensor data in the preliminary digital twin model is performed using a graph neural network-long short-term memory neural network structure to obtain the final digital twin model.

[0065] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0066] First, a multi-scale mapping model from individual battery cells to the drive motor is constructed, comprehensively covering all levels of the system through a hierarchical architecture to ensure accurate representation of characteristics from the microscopic to the macroscopic. Next, multi-physics coupled modeling of electrochemical, temperature, and stress fields is performed, changing the traditional single-physics modeling mode and recreating the real-world interaction of various physical fields during system operation, enhancing the model's simulation capability for complex operating conditions. Then, real-time monitoring data is used to update the parameters of the coupled multi-scale mapping model through data-driven modeling, enabling the model to respond in real-time to dynamic factors such as battery aging and changes in operating conditions, maintaining synchronization with the actual system. Finally, a graph neural network-long short-term memory neural network structure is used; the former mines the spatial correlation of sensor data, and the latter captures time-series features, achieving spatiotemporal dynamic analysis of the system's operating state. Compared to existing technologies with problems such as isolated multi-scale modeling, insufficient physical field coupling, data-driven lag, and limitations of time-series modeling, this invention, through hierarchical multi-scale modeling, deep physical field coupling, dynamic data-driven approach, and spatiotemporal feature mining, achieves high-precision mapping and dynamic prediction of the operating state of the new energy heavy-duty truck battery power system, providing a solid and reliable digital foundation for efficient management and optimization throughout the system's entire lifecycle.

[0067] As one example, the construction of a multi-scale imaging model of the battery power system includes a hierarchical model from individual battery cells to modules, the entire battery pack, the controller, and the drive motor, comprising the following sub-steps:

[0068] Based on the energy conservation equation, an electrochemical-thermal model at the cell scale is constructed on top of the P2D model, resulting in a model structure that integrates electrochemical and thermal properties at the cell scale.

[0069] Based on the cell-scale basic model, considering the thermal stress and thermal expansion caused by temperature gradient during battery operation, a single-cell-scale thermo-mechanical coupling model is constructed to form a comprehensive model that includes cell and single-cell-scale characteristics.

[0070] Based on the comprehensive model, starting from the power battery module and the whole pack scale, a three-dimensional CAD model corresponding to the real shape of the physical battery is established using industrial modeling software. This accurately presents key structural parameters, constraints and positioning relationships between individual cells and modules, and obtains a three-dimensional model containing multi-scale structural information.

[0071] Pixyz Studio was used to perform surface subdivision, topology correction, extraction, and repair operations on a 3D model containing multi-scale structural information. Then, combined with Hertz theory and extended finite element method, an overall digital physical model was established, the semantic information annotation objects were clarified, and the construction of a multi-scale image model of the battery power system was completed.

[0072] In this embodiment, a high-precision multi-scale imaging model of the battery power system is constructed through a hierarchical modeling strategy that integrates micro- and macro-scale and multi-physical characteristics. First, based on the energy conservation equation, an electrochemical-thermal model at the cell scale is built on the P2D model, coupling the electrochemical reactions and heat transfer processes inside the battery to accurately characterize the core characteristics at the microscopic level of the cell. Next, based on the cell model, thermal stress and thermal expansion factors induced by temperature gradients are introduced to construct a thermo-mechanical coupling model at the single-cell scale, realizing the extension of physical characteristics from the micro-cell to the single-cell battery, enabling the model to reflect more complex physical phenomena. Then, using the comprehensive model as a blueprint, three-dimensional CAD models are constructed at the power battery module and overall pack scales using industrial modeling software, accurately presenting the structural parameters of the battery entity and the constraint relationships between components, completing the structural modeling from the single-cell to the system level. Finally, the three-dimensional model is optimized using Pixyz Studio, and a holistic digital physical model is established by combining Hertzian theory and the extended finite element method, clarifying semantic annotations, and achieving a deep integration of multi-scale structure and physical characteristics. In summary, this embodiment effectively solves the problems of fragmented multi-scale characteristics and insufficient physical field coupling in traditional modeling by adopting a progressive modeling approach. It not only achieves accurate modeling of the full-level structure and physical characteristics from the cell to the entire battery pack, but also ensures that the model can realistically reflect the complex state of the battery power system in actual operation. This provides a solid and reliable foundation for subsequent multi-physics coupling modeling, parameter updates, and dynamic time series analysis, and significantly improves the mapping accuracy and simulation capability of the digital twin model for the battery power system.

[0073] As an example, the working process of this embodiment is as follows:

[0074] First, the electrochemical-thermal model at the cell scale is constructed based on the energy conservation equation and built upon a P2D (two-dimensional pseudo-) model. During the cell's charging and discharging process, the heat generated by the electrochemical reaction, as well as the conduction and convection of heat within the cell, are considered. An improved heat transfer algorithm formula is introduced: Among them, Q total This represents the total heat generated by the battery cell; n is the number of micro-elements within the battery cell; α i T is the convective heat transfer coefficient between the i-th micro-element and the environment. This coefficient can be obtained by experimentally measuring the heat transfer between the battery cell and the surrounding air under different material and surface conditions; i The temperature of the i-th micro-element is acquired in real time by high-precision temperature sensors pre-positioned at different locations within the cell; T amb β represents the ambient temperature, measured by an ambient temperature sensor; β is a coefficient related to the cell material and structure, calculated through analysis of the cell material properties and internal structural parameters; I represents the current flowing through the cell, monitored in real time by the battery management system; and R represents the cell's internal resistance, obtained by correcting for factory parameters using an aging model. By combining the electrochemical heat generation and heat transfer processes of the cell in this way, a model structure integrating electrochemical and thermal characteristics at the cell scale is obtained.

[0075] Next, based on the aforementioned cell-scale basic model, a thermo-mechanical coupling model at the single-cell scale is constructed. The thermal stress and thermal expansion caused by temperature gradients during battery operation are considered. For the calculation of thermal stress, an improved stress calculation model is adopted: σ thermal =γ·ΔT·E·(1-v). Among them, σ thermal γ represents thermal stress; γ is the coefficient of thermal expansion of the cell material, determined through material property experiments; ΔT is the temperature difference at different locations within the cell, calculated from the temperatures of each micro-element in the previously constructed cell-scale model; E is the elastic modulus of the material, which can be modified by referring to the material handbook and considering actual production processes; v is the Poisson's ratio of the material, obtained through experimental measurement. Thermal stress is calculated using this formula and combined with the cell-scale model, while also considering the structural changes caused by the thermal expansion of individual battery cells, forming a comprehensive model that incorporates both cell-scale and single-cell-scale characteristics.

[0076] Then, based on the comprehensive model, starting from the dimensions of the power battery module and the entire pack, a 3D CAD model corresponding to the actual shape of the physical battery is created using industrial modeling software (such as CATIA). During the modeling process, based on the battery design drawings and actual production process parameters, the dimensions, shapes, and relative positions of the battery modules and the entire pack are precisely set, presenting key structural parameters such as the arrangement of individual cells, the connection structure between modules, and the position of fasteners. Simultaneously, utilizing the software's constraint and positioning functions, the constraint and positioning relationships between individual cells and modules are accurately simulated, obtaining a 3D model containing multi-scale structural information.

[0077] Finally, Pixyz Studio was used to perform surface subdivision, topology correction, extraction, and repair operations on the 3D model containing multi-scale structural information. Based on this, a global digital physical model was established using Hertzian theory and the extended finite element method. During the model building process, according to Hertzian theory, the contact mechanics analysis between battery cells was performed using the following formula: Among them, F contact δ represents the contact force between individual cells; k is the contact stiffness coefficient related to material properties and contact surface geometry, obtained through analysis and calculation of battery cell materials and surface treatment processes; δ is the contact deformation between individual cells, calculated through simulation using finite element analysis software. The extended finite element method is used to simulate and analyze potential defects such as cracks in the battery. After the model is built, semantic information annotation objects are defined, such as key battery components and parameter measurement points, completing the construction of a multi-scale imaging model of the battery power system.

[0078] As one example, the process of performing multi-physics coupling modeling of the constructed multi-scale imaging model with electrochemical field, temperature field, and stress field to obtain the coupled multi-scale imaging model includes the following sub-steps:

[0079] For the completed multi-scale imaging model of the battery power system, we comprehensively sort out the parameter information related to the electrochemical field, temperature field and stress field, and obtain the parameter set of each physical field at different scales;

[0080] Based on the parameter set, we analyze the transmission paths and methods of each physical field parameter between different levels of models, clarify the rules of parameter transmission, and obtain a specific description of the rules of parameter transmission.

[0081] Based on the obtained parameter transfer rule description, the interaction mechanism between different physical fields and different level models is studied, the influencing factors of the interaction are determined, and a list of key factors affecting the interaction is obtained.

[0082] Based on the parameter set, parameter transfer rule description, and list of key factors affecting interaction, the multi-scale imaging model is coupled and integrated with multi-physics fields to obtain the coupled multi-scale imaging model.

[0083] In this embodiment, deep coupling of multiple physics fields in a multi-scale imaging model of a battery power system is achieved through systematic parameter analysis, transfer rule mining, and interaction analysis. First, the relevant parameters of the electrochemical field, temperature field, and stress field in the multi-scale imaging model are comprehensively analyzed, and parameter sets at different scales are constructed to provide a data foundation for subsequent coupling. Next, the transfer paths and methods of each physics field parameter between hierarchical models are analyzed in depth, clarifying the parameter transfer rules and establishing parameter transfer rule descriptions to solve the problems of isolated parameters and lack of correlation in traditional models. Then, the interaction mechanisms between different physics fields and hierarchical models are studied, identifying key factors affecting interactions and forming a list of key factors to guide coupling modeling. Finally, by integrating the parameter sets, transfer rules, and key factors, the multiple physics fields are integrated into the multi-scale imaging model, completing the multiphysics field coupling modeling. In summary, this embodiment overcomes the shortcomings of traditional modeling, such as the fragmentation and insufficient coupling of multiple physics fields, through a systematic study of physical field parameters, transfer rules, and interaction mechanisms. It achieves collaborative modeling of electrochemical fields, temperature fields, and stress fields at different scales, significantly improving the simulation accuracy of digital twin models for complex physical phenomena in battery power systems. This provides a more realistic and reliable model basis for system performance analysis and optimization, and enhances the model's adaptability and predictive ability to actual working conditions.

[0084] As an example, the working process of this embodiment is as follows:

[0085] 1. For the established multi-scale imaging model, a comprehensive review of all physical field parameters is first conducted. For the electrochemical field, at the cell scale, the exchange current density and lithium-ion diffusion coefficient of the electrode materials are measured using an electrochemical workstation; at the module and package scales, internal resistance distribution and SOC (state of charge) distribution are calculated using multi-point voltage monitoring combined with an equivalent circuit model. For the temperature field, temperature distribution data are acquired using thermocouples and infrared thermal imagers distributed at key locations in the battery, and thermal conductivity and specific heat capacity are calculated using thermal imaging analysis software. For the stress field, fiber optic sensors are embedded during cell manufacturing to monitor stress changes in real time during charging and discharging, and a set of stress field parameters is constructed by combining the elastic modulus and Poisson's ratio determined by materials mechanics experiments. By establishing a parameter mapping table, each physical field parameter is classified and stored according to four scale levels: cell, single cell, module, and package, forming a complete parameter matrix.

[0086] 2. Based on the obtained parameter set, the transmission paths and modes of each physical field parameter between different levels of the model are analyzed. Taking the coupling of the temperature field and the electrochemical field as an example, temperature changes significantly affect the electrochemical reaction rate and the lithium-ion diffusion coefficient. This invention improves the traditional Arrhenius equation, proposes a temperature correction factor ξ, and establishes a dynamic correlation model between temperature and electrochemical parameters: Among them, C Li For lithium ion concentration, D eff For the effective diffusion coefficient, For the Laplace operator, The temperature change rate is represented by ξ. The temperature correction factor ξ, obtained through fitting experimental data, reflects the influence of different temperature ranges on lithium-ion diffusion. This model accurately describes the transfer process from temperature field parameters to electrochemical field parameters. Similarly, a similar method is used to establish transfer rules for parameters between other physical fields, ultimately forming a complete description of parameter transfer rules.

[0087] 3. Based on parameter transfer rules, this invention delves into the interaction mechanisms between different physical fields and different levels of models. Taking the interaction between temperature and stress fields as an example, temperature changes cause thermal expansion and contraction of battery materials, generating thermal stress; while stress changes affect the thermal conductivity of the materials, forming a two-way coupling. This invention proposes a thermal stress calculation model: σ thermal =γ·ΔT·E·(1-v), where σ thermal Let γ be the thermal stress, ΔT be the temperature difference, E be the elastic modulus, and v be Poisson's ratio. This model allows for accurate calculation of thermal stress caused by temperature changes. Furthermore, research has revealed that stress changes alter the material's internal microstructure, thus affecting its thermal conductivity. A stress-thermal conduction coupling model was established: λ(σ)=λ0·(1+α·σ), where λ(σ) is the thermal conductivity under stress, λ0 is the initial thermal conductivity, α is the stress influence coefficient, and σ is the stress value. These two models clarify the key influencing factors of the interaction between the temperature and stress fields, forming a complete list of key factors affecting this interaction.

[0088] 4. Based on the parameter set, parameter transfer rules, and list of key interaction factors, multi-physics coupling integration is performed on the multi-scale imaging model. This invention proposes a comprehensive coupling weight algorithm to unify the modeling of electrochemical, temperature, and stress fields: Among them, S total S represents the combined state parameters after multiphysics coupling. i Let w be the state parameters of each physical field (i = 1, 2, 3 represent the electrochemical field, temperature field, and stress field, respectively). i λ represents the weighting coefficients for each physical field. ijThis represents the coupling coefficient between physical fields. The weighting coefficient w i The coupling coefficient λ is dynamically adjusted based on the battery's operating state. For example, the electrochemical field weight increases during fast charging, while the temperature field weight increases under high-temperature conditions. ij The model, obtained by fitting experimental data, reflects the interaction strength between different physical fields. This integrated coupling model achieves the organic integration of the three physical fields, resulting in a complete coupled multi-scale mapping model.

[0089] As one example, the process of updating the parameters of the coupled multi-scale imaging model based on real-time monitoring data and using a data-driven model to generate a preliminary digital twin model includes the following sub-steps:

[0090] By pre-deploying sensors in the battery power system, real-time monitoring data during system operation is collected to obtain a dataset containing the real-time status of the battery power system.

[0091] Data-driven methods are used to extract information from datasets containing the real-time status of battery power systems, and effective information related to the parameters of multi-scale imaging models is filtered out to form an effective information set.

[0092] The information in the effective information set is compared and matched with the parameters of the coupled multi-scale imaging model to determine the parameters that need to be updated and their update values, thus obtaining a parameter update list.

[0093] Based on the parameter update list, the parameters of the coupled multi-scale imaging model are updated to generate a preliminary digital twin model.

[0094] In this embodiment, a dynamic parameter update mechanism based on real-time data-driven principles is constructed to ensure the synchronization between the digital twin model and the physical system. First, real-time monitoring data is collected by sensors pre-deployed in the battery power system to obtain the system's original operational status information, providing a data foundation for model updates. Second, a data-driven approach is used to deeply mine the original dataset, filtering out effective information related to the parameters of the multi-scale imaging model, eliminating redundant data, and improving data utilization efficiency. Then, the effective information is compared and matched with the model parameters to accurately locate the parameters that need updating and their values, forming a parameter update list and clarifying the update direction. Finally, the parameters of the coupled multi-scale imaging model are adjusted according to the update list, enabling the model to reflect the real-time changes in the operating status of the battery power system and generating a preliminary digital twin model. In summary, this implementation breaks through the limitations of static parameter setting in traditional models by real-time data acquisition, effective information extraction, precise parameter matching, and dynamic updates. It solves the problem of the model being out of sync with the actual system operating state, significantly improving the dynamic adaptability and accuracy of the digital twin model. This enables the model to respond promptly to dynamic factors such as battery aging and changes in operating conditions, providing more reliable digital model support for real-time monitoring, fault diagnosis, and optimized control of the battery power system.

[0095] As an example, the working process of this embodiment is as follows:

[0096] Step 1, Real-time Data Acquisition Monitoring: A comprehensive and sophisticated sensor monitoring network is constructed within the battery power system of new energy heavy-duty trucks. At the cell level, high-precision miniature thermocouple sensors are installed at the positive and negative tabs of each cell, achieving a measurement accuracy of ±0.1℃, enabling precise capture of subtle local temperature changes during charging and discharging. Specialized miniature solid-state electrolyte lithium-ion concentration sensors are embedded on both sides of the cell separator. These sensors can acquire lithium-ion concentration data in real time, with a fast response speed and a measurement range of 0.1-2 mol / m³. 3At the module level, high-precision Hall current sensors (accuracy ±0.2%) are used to monitor the current distribution of each module. Strain gauges are also attached to key stress-bearing areas of the modules to measure changes in mechanical stress caused by battery expansion and contraction. At the battery pack level, pressure sensors are strategically placed at key locations such as the top, bottom, and sides of the pack to monitor internal pressure fluctuations. A high-resolution infrared thermal imager is installed above the pack to scan the temperature field distribution in real time, with a resolution of 640×480. All sensors are connected to the data acquisition unit via a CAN bus. The data acquisition unit employs a dynamic sampling strategy based on parameter change characteristics: parameters with slow changes, such as voltage and temperature, are sampled at a frequency of 1Hz; while parameters with rapid changes, such as current and stress, are sampled at a high frequency of 100Hz. The acquired data includes timestamps, sensor IDs, measured values, and data quality identifiers, ultimately forming a multidimensional dataset D covering the real-time status of the battery power system.

[0097] Step 2, Data-Driven Effective Information Extraction: When filtering effective information related to the parameters of the multi-scale mapping model from the original dataset D, an improved feature selection method is adopted. First, Principal Component Analysis (PCA) is used to reduce the dimensionality of the original data. PCA maps the original data to a new low-dimensional space through linear transformation, removing redundant information and reducing data dimensionality while retaining the main feature information. After dimensionality reduction, an improved Recursive Feature Elimination (RFE) algorithm is introduced, combined with the model parameter sensitivity index for feature selection. The parameter sensitivity formula is defined as:

[0098]

[0099] Among them, S i The sensitivity of the i-th parameter is used to measure the degree of influence of that parameter on the model output; M represents the output of the multi-scale mapping model, such as the predicted voltage and temperature of the battery; P i Let be the i-th parameter in the model; n is the total number of model parameters. This formula calculates the sensitivity of each data feature to the model parameters, prioritizing the retention of features with high sensitivity, and integrating these features to form an effective information set I. eff This improved algorithm significantly enhances the accuracy and relevance of information filtering by quantifying the impact of features on model output, enabling it to more precisely identify data features valuable for updating model parameters.

[0100] Step 3, Parameter Comparison and Matching: The valid information set I... effThe parameters are compared with those of the coupled multi-scale image model to determine the parameters that need to be updated and their updated values. An improved dynamic threshold matching algorithm is used here, and the parameter bias evaluation formula is constructed as follows:

[0101]

[0102] Where, Δ k This represents the deviation ratio of the k-th parameter, used to intuitively reflect the degree of deviation between the measured value and the model prediction value; It is the measured value of the kth parameter, which is obtained by the sensor in step 1 and after data processing; The model prediction value for the k-th parameter is calculated by the multi-scale imaging model based on the current parameter state. This serves as a reference value for the parameter, typically the design nominal value or a stable value obtained through historical data statistics. Set the dynamic threshold θ. k =θ0×(1+β×SOH+γ×SOC), where θ0 is the basic threshold, an initial judgment standard set based on experience and experiments; β and γ are adjustment coefficients, optimized and determined through extensive experimental data and actual operation; SOH represents the battery health state, reflecting the degree of battery aging; SOC is the state of charge, representing the current remaining capacity of the battery. When Δ k >θ k When the parameter needs to be updated, a parameter update list L is generated. This list records in detail the name of the parameter to be updated, its current value, the target update value, and the priority identifier, providing a clear basis for subsequent model parameter updates.

[0103] Step 4, Multi-scale Image Model Parameter Update: Based on the parameter update list L, a hierarchical iterative update strategy is used to correct the parameters of the coupled multi-scale image model. For independent parameters, such as the cell internal resistance, since they are less affected by other parameters, the updated values ​​can be directly used to replace the original parameters to complete the parameter update. For parameters with coupling relationships, such as electrochemical-thermal-stress coupling parameters, a collaborative update algorithm is used.

[0104]

[0105] in, This represents the updated value of the i-th parameter; This is the original value of the parameter; ΔP i λ represents the parameter update amount calculated in step 3. iThis involves updating the weighting coefficients, which are determined based on the parameter's hierarchy and coupling relationship. Generally, cell-level parameters have a more direct and critical impact on battery performance, resulting in relatively higher weights, while pack-level parameters have lower weights. During the update process, through continuous iterative calculations, the model parameters are constantly adjusted to ensure that the error between the model output and the measured data gradually converges to a pre-set threshold range, ultimately generating a preliminary digital twin model that accurately reflects the real-time state of the battery power system.

[0106] As one example, the process of dynamically and temporally modeling sensor data in the initial digital twin model using a graph neural network-long short-term memory neural network structure to obtain the final digital twin model includes the following sub-steps:

[0107] The sensor data in the preliminary digital twin model is abstracted into a network topology diagram of the sensor data time-series accumulation process according to the time series and spatial relationship, so as to obtain a topology structure that reflects the spatiotemporal characteristics of sensor data;

[0108] A graph neural network-long short-term memory neural network structure is used to process the topological structure that reflects the spatiotemporal characteristics of sensor data, and integrate the structural information and temporal information to form intermediate data that fuses spatiotemporal characteristics.

[0109] By utilizing the designed hierarchical feature aggregation method, different feature aggregators are learned on intermediate data with fused spatiotemporal features at different depths of neighborhood, thereby obtaining aggregated data containing multi-scale features.

[0110] Dynamic time-series modeling of sensor data is performed, and aggregated data containing multi-scale features is incorporated into the model to obtain the final digital twin model.

[0111] In this embodiment, firstly, real-time operational data is collected by sensors deployed in the battery power system to obtain raw information about the system's true state, thus building a data foundation. Next, a data-driven approach is used to deeply process the raw dataset, filtering out effective information related to the parameters of the multi-scale imaging model from the massive data, removing redundancy, and improving data quality. Then, the effective information is precisely compared with the coupled multi-scale imaging model parameters to identify parameter differences, determine the parameters that need updating and their corresponding values, and form a parameter update list. Finally, the model parameters are adjusted according to the list to ensure the model closely matches the system's operating state in real time, generating a preliminary digital twin model. In summary, this embodiment effectively solves the problem of traditional model parameters being fixed and unable to reflect dynamic changes in the system. Through real-time data acquisition and precise parameter updates, it ensures that the digital twin model can dynamically adjust with the operating status of the battery power system, significantly improving the matching degree and synchronization between the model and the actual system. This provides dynamic and accurate model support for the monitoring, analysis, and optimization of battery power systems based on digital twins, enhancing the reliability of system prediction and decision-making.

[0112] As an example, the working process of this embodiment is as follows:

[0113] Step 1, Sensor Data Topology Construction: In the initial digital twin model, spatiotemporal characteristics are abstracted for the sensor data of the battery power system. Taking the battery module as an example, devices such as cell temperature sensors, voltage sensors, and current sensors are defined as network nodes according to their physical location and connection relationships; the physical connection lines between sensors (such as CAN bus connections, internal circuit connections within the module) or data interaction logic relationships are used as network edges. For time series data, a 1-minute time window is used, and the data collected by the sensors within each window is used as node attributes. For example, node v i Includes the average temperature T of the i-th cell within the current time window. i Voltage V i Charging and discharging current I i Attributes, forming an attribute vector X i =[T i V i I i In this way, sensor data is constructed into a directed weighted network topology graph with attributes. in Let ε be the set of nodes, ε be the set of edges, and X be the attribute matrix of all nodes. This topology graph fully reflects the spatial distribution (node ​​locations and connections) and time-series characteristics (data aggregation within a window) of the sensor data.

[0114] Step 2, Spatiotemporal Feature Fusion Processing: An improved Graph Neural Network-Long Short-Term Memory Neural Network (GNN-LSTM) fusion structure is used for data processing. First, the topological graph $\mathcal{G}$ is input into the graph neural network layer (GNNLayer), and the node features are updated using an improved message passing algorithm:

[0115]

[0116] in, Let i be the hidden state of node i in the l-th layer of the GNN; It is the set of neighboring nodes of node i; e ij The weight of the edge between nodes i and j (set according to the physical connection strength or data correlation); W (l) and b (l) σ represents the weight matrix and bias vector of the l-th layer; σ is the activation function (such as ReLU). This formula enhances the extraction of spatial structural features by weighted aggregation of neighbor node information.

[0117] Subsequently, the node feature sequences output by the GNN are input into a Long Short-Term Memory (LSTM) neural network layer. To adapt to the multi-scale characteristics of sensor data, the forget gate, input gate, and output gate of the LSTM are improved by introducing a spatiotemporal weight coefficient α. t :

[0118] f t =σ(W f ·[h t-1 x t ]+b f )·α t

[0119] i t =σ(W i ·[h t-1 x t ]+b i )·(1-α t )

[0120] o t =σ(W o ·[h t-1 x t ]+b o )·α t

[0121] Among them, f t i t o t The outputs of the forget gate, input gate, and output gate are respectively; h t-1 The hidden state from the previous moment; x t Input for the current time; Wf W i W o and b f b i b o For the weights and biases; α t The spatiotemporal weighting coefficients are dynamically adjusted based on the fluctuation of sensor data within the current time window (calculated through regression of historical data). After joint processing by GNN and LSTM, intermediate data Z, which integrates spatiotemporal features, is generated.

[0122] Step 3, Hierarchical Feature Aggregation: Design a hierarchical feature aggregation method to achieve multi-scale feature extraction. The intermediate data Z is divided into three categories according to node level: cell level, module level, and package level. An improved aggregator is used in each level. Perform feature learning:

[0123]

[0124] in, The output of node i in the k-th layer aggregator; Let z be the set of nodes in the k-th layer neighborhood of node i; j Let be the feature vector of node j. Taking the cell level as an example, Using a weighted average aggregator:

[0125]

[0126] Among them, w ij The relevance weights between nodes i and j are calculated based on the thermal conductivity and electrical connection tightness between cells. At the module and package levels, attention-based aggregators and pooling aggregators are used respectively to highlight the characteristics of key nodes. Through three-layer aggregation processing, aggregated data Y containing micro-scale features of cells, meso-scale features of modules, and macro-scale features of the entire package is generated.

[0127] Step 4, Dynamic Time Series Modeling and Final Model Generation: Based on the aggregated data Y, a dynamic time series prediction model is constructed. An improved version of the Autoregressive Integrated Moving Average (ARIMA) model is adopted, introducing multi-scale feature weighting coefficients ω. k :

[0128]

[0129] in, ω is the predicted value at time t; k φ represents the weighting coefficients of the k-th scale features (optimized through cross-validation); i,k and θ i,k These are the autoregressive and moving average coefficients, respectively; ∈ tThe input is a white noise sequence; p and q are the model orders. This formula enhances the model's adaptability to complex dynamic changes by fusing multi-scale features. The prediction results are fed back to the digital twin model to update the model parameters, ultimately generating a final digital twin model that accurately reflects the dynamic behavior of the battery power system.

[0130] It should be noted that the relevant algorithmic technical details not elaborated above can be found in existing related algorithms and will not be repeated here.

[0131] See Figure 2 This is a schematic diagram of a digital twin model construction system for a new energy heavy-duty truck battery power system according to an embodiment of the present invention. The digital twin model construction system for the new energy heavy-duty truck battery power system includes:

[0132] Module 10 is used to build a multi-scale image model of the battery power system, including a hierarchical model from individual battery cells to modules, the entire pack, the controller, and the drive motor.

[0133] The first modeling module 11 is used to perform multi-physics field coupling modeling of electrochemical field, temperature field and stress field on the constructed multi-scale image model to obtain the coupled multi-scale image model.

[0134] Update module 12 is used to update the parameters of the coupled multi-scale imaging model based on real-time monitoring data and a data-driven model to generate a preliminary digital twin model.

[0135] The second modeling module 13 is used to perform dynamic time-series modeling of sensor data in the preliminary digital twin model through a graph neural network-long short-term memory neural network structure to obtain the final digital twin model.

[0136] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0137] First, a multi-scale mapping model from individual battery cells to the drive motor is constructed, comprehensively covering all levels of the system through a hierarchical architecture to ensure accurate representation of characteristics from the microscopic to the macroscopic. Next, multi-physics coupled modeling of electrochemical, temperature, and stress fields is performed, changing the traditional single-physics modeling mode and recreating the real-world interaction of various physical fields during system operation, enhancing the model's simulation capability for complex operating conditions. Then, real-time monitoring data is used to update the parameters of the coupled multi-scale mapping model through data-driven modeling, enabling the model to respond in real-time to dynamic factors such as battery aging and changes in operating conditions, maintaining synchronization with the actual system. Finally, a graph neural network-long short-term memory neural network structure is used; the former mines the spatial correlation of sensor data, and the latter captures time-series features, achieving spatiotemporal dynamic analysis of the system's operating state. Compared to existing technologies with problems such as isolated multi-scale modeling, insufficient physical field coupling, data-driven lag, and limitations of time-series modeling, this invention, through hierarchical multi-scale modeling, deep physical field coupling, dynamic data-driven approach, and spatiotemporal feature mining, achieves high-precision mapping and dynamic prediction of the operating state of the new energy heavy-duty truck battery power system, providing a solid and reliable digital foundation for efficient management and optimization throughout the system's entire lifecycle.

[0138] As one example, the building module is specifically used for:

[0139] Based on the energy conservation equation, an electrochemical-thermal model at the cell scale is constructed on top of the P2D model, resulting in a model structure that integrates electrochemical and thermal properties at the cell scale.

[0140] Based on the cell-scale basic model, considering the thermal stress and thermal expansion caused by temperature gradient during battery operation, a single-cell-scale thermo-mechanical coupling model is constructed to form a comprehensive model that includes cell and single-cell-scale characteristics.

[0141] Based on the comprehensive model, starting from the power battery module and the whole pack scale, a three-dimensional CAD model corresponding to the real shape of the physical battery is established using industrial modeling software. This accurately presents key structural parameters, constraints and positioning relationships between individual cells and modules, and obtains a three-dimensional model containing multi-scale structural information.

[0142] Pixyz Studio was used to perform surface subdivision, topology correction, extraction, and repair operations on a 3D model containing multi-scale structural information. Then, combined with Hertz theory and extended finite element method, an overall digital physical model was established, the semantic information annotation objects were clarified, and the construction of a multi-scale image model of the battery power system was completed.

[0143] As one example, the first modeling module is specifically used for:

[0144] For the completed multi-scale imaging model of the battery power system, we comprehensively sort out the parameter information related to the electrochemical field, temperature field and stress field, and obtain the parameter set of each physical field at different scales;

[0145] Based on the parameter set, we analyze the transmission paths and methods of each physical field parameter between different levels of models, clarify the rules of parameter transmission, and obtain a specific description of the rules of parameter transmission.

[0146] Based on the obtained parameter transfer rule description, the interaction mechanism between different physical fields and different level models is studied, the influencing factors of the interaction are determined, and a list of key factors affecting the interaction is obtained.

[0147] Based on the parameter set, parameter transfer rule description, and list of key factors affecting interaction, the multi-scale imaging model is coupled and integrated with multi-physics fields to obtain the coupled multi-scale imaging model.

[0148] As one example, the update module is specifically used for:

[0149] By pre-deploying sensors in the battery power system, real-time monitoring data during system operation is collected to obtain a dataset containing the real-time status of the battery power system.

[0150] Data-driven methods are used to extract information from datasets containing the real-time status of battery power systems, and effective information related to the parameters of multi-scale imaging models is filtered out to form an effective information set.

[0151] The information in the effective information set is compared and matched with the parameters of the coupled multi-scale imaging model to determine the parameters that need to be updated and their update values, thus obtaining a parameter update list.

[0152] Based on the parameter update list, the parameters of the coupled multi-scale imaging model are updated to generate a preliminary digital twin model.

[0153] As one example, the second modeling module is specifically used for:

[0154] The sensor data in the preliminary digital twin model is abstracted into a network topology diagram of the sensor data time-series accumulation process according to the time series and spatial relationship, so as to obtain a topology structure that reflects the spatiotemporal characteristics of sensor data;

[0155] A graph neural network-long short-term memory neural network structure is used to process the topological structure that reflects the spatiotemporal characteristics of sensor data, and integrate the structural information and temporal information to form intermediate data that fuses spatiotemporal characteristics.

[0156] By utilizing the designed hierarchical feature aggregation method, different feature aggregators are learned on intermediate data with fused spatiotemporal features at different depths of neighborhood, thereby obtaining aggregated data containing multi-scale features.

[0157] Dynamic time-series modeling of sensor data is performed, and aggregated data containing multi-scale features is incorporated into the model to obtain the final digital twin model.

[0158] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0159] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for constructing a digital twin model of a battery power system for new energy heavy-duty trucks, characterized in that, include: Construct a multi-scale imaging model of the battery power system, including a hierarchical model from individual battery cells to modules, the entire pack, controllers, and drive motors; The constructed multi-scale imaging model is coupled with electrochemical field, temperature field and stress field to obtain the coupled multi-scale imaging model. Based on real-time monitoring data, a data-driven model is used to update the parameters of the coupled multi-scale imaging model, generating a preliminary digital twin model. The parameter update includes: collecting real-time monitoring data from sensors pre-deployed in the battery power system to obtain a multidimensional dataset containing the real-time state of the battery power system; performing dimensionality reduction on the multidimensional dataset, and then combining it with parameter sensitivity indices for feature filtering to form a set of effective information related to the parameters of the multi-scale imaging model. The parameter sensitivity indices are used to measure the degree of influence of the parameters on the model output. The information in the effective information set is compared and matched with the parameters of the coupled multi-scale imaging model; wherein, a dynamic threshold matching algorithm is used to determine the parameters that need to be updated and their update values, specifically including: constructing the deviation ratio Δk of the k-th parameter: , Indicates the first The deviation ratio of each parameter, It is the first Measured values ​​of each parameter For the first The model prediction values ​​for each parameter. This is the reference value for this parameter; Set the dynamic threshold θk: , Based on the threshold, and The adjustment coefficient is SOH, which represents the battery health state, and SOC, which represents the state of charge. The dynamic threshold is dynamically adjusted according to the battery health state and the state of charge. When the parameter needs to be updated, a parameter update list is generated, which records the name of the parameter to be updated, its current value, the target update value, and the priority identifier. Based on the parameter update list, a hierarchical iterative update strategy is used to correct the parameters of the coupled multi-scale imaging model. Update weights are assigned according to the parameter's level, with cell layer parameters having a greater update weight than the entire package layer parameters. For independent parameters, the updated values ​​directly replace the original parameters. For electrochemical-thermal-stress coupled parameters with a coupling relationship, a collaborative update algorithm is used for iterative correction. , This represents the updated value of the i-th parameter. This is the original value of the parameter. For the calculated parameter update amount, To update the weight coefficients, the updated weight coefficients are determined based on the parameter's level and coupling relationship; during the iterative correction process, the calculation is continuously iterated and the model parameters are continuously adjusted until the error between the model output and the measured data converges to a preset threshold range; The sensor data in the preliminary digital twin model is dynamically and temporally modeled using a graph neural network-long short-term memory neural network structure to obtain the final digital twin model.

2. The method for constructing a digital twin model of a new energy heavy-duty truck battery power system as described in claim 1, characterized in that, The construction of the multi-scale imaging model of the battery power system includes a hierarchical model from individual battery cells to modules, the entire battery pack, the controller, and the drive motor, and includes the following sub-steps: Based on the energy conservation equation, an electrochemical-thermal model at the cell scale is constructed on top of the P2D model, resulting in a model structure that integrates electrochemical and thermal properties at the cell scale. Based on the cell-scale basic model, considering the thermal stress and thermal expansion caused by temperature gradient during battery operation, a single-cell-scale thermo-mechanical coupling model is constructed to form a comprehensive model that includes cell and single-cell-scale characteristics. Based on the comprehensive model, starting from the power battery module and the whole pack scale, a three-dimensional CAD model corresponding to the real shape of the physical battery is established using industrial modeling software. This accurately presents key structural parameters, constraints and positioning relationships between individual cells and modules, and obtains a three-dimensional model containing multi-scale structural information. Pixyz Studio was used to perform surface subdivision, topology correction, extraction, and repair operations on a 3D model containing multi-scale structural information. Then, combined with Hertz theory and extended finite element method, an overall digital physical model was established, the semantic information annotation objects were clarified, and the construction of a multi-scale image model of the battery power system was completed.

3. The method for constructing a digital twin model of a new energy heavy-duty truck battery power system as described in claim 1, characterized in that, The process of performing multi-physics coupling modeling of the constructed multi-scale imaging model with electrochemical field, temperature field, and stress field to obtain the coupled multi-scale imaging model includes the following sub-steps: For the completed multi-scale imaging model of the battery power system, we comprehensively sort out the parameter information related to the electrochemical field, temperature field and stress field, and obtain the parameter set of each physical field at different scales; Based on the parameter set, we analyze the transmission paths and methods of each physical field parameter between different levels of models, clarify the rules of parameter transmission, and obtain a specific description of the rules of parameter transmission. Based on the obtained parameter transfer rule description, the interaction mechanism between different physical fields and different level models is studied, the influencing factors of the interaction are determined, and a list of key factors affecting the interaction is obtained. Based on the parameter set, parameter transfer rule description, and list of key factors affecting interaction, the multi-scale imaging model is coupled and integrated with multi-physics fields to obtain the coupled multi-scale imaging model.

4. The method for constructing a digital twin model of a new energy heavy-duty truck battery power system as described in claim 1, characterized in that, The process of updating the parameters of the coupled multi-scale imaging model based on real-time monitoring data and using a data-driven model to generate a preliminary digital twin model includes the following sub-steps: By pre-deploying sensors in the battery power system, real-time monitoring data during system operation is collected to obtain a dataset containing the real-time status of the battery power system. Data-driven methods are used to extract information from datasets containing the real-time status of battery power systems, and effective information related to the parameters of multi-scale imaging models is filtered out to form an effective information set. The information in the effective information set is compared and matched with the parameters of the coupled multi-scale imaging model to determine the parameters that need to be updated and their update values, thus obtaining a parameter update list. Based on the parameter update list, the parameters of the coupled multi-scale imaging model are updated to generate a preliminary digital twin model.

5. The method for constructing a digital twin model of a new energy heavy-duty truck battery power system as described in claim 1, characterized in that, The process of dynamically and temporally modeling the sensor data in the preliminary digital twin model using a graph neural network-long short-term memory neural network structure to obtain the final digital twin model includes the following sub-steps: The sensor data in the preliminary digital twin model is abstracted into a network topology diagram of the sensor data time-series accumulation process according to the time series and spatial relationship, so as to obtain a topology structure that reflects the spatiotemporal characteristics of sensor data; A graph neural network-long short-term memory neural network structure is used to process the topological structure that reflects the spatiotemporal characteristics of sensor data, and integrate the structural information and temporal information to form intermediate data that fuses spatiotemporal characteristics. By utilizing the designed hierarchical feature aggregation method, different feature aggregators are learned on intermediate data with fused spatiotemporal features at different depths of neighborhood, thereby obtaining aggregated data containing multi-scale features. Dynamic time-series modeling of sensor data is performed, and aggregated data containing multi-scale features is incorporated into the model to obtain the final digital twin model.

6. A digital twin model construction system for a new energy heavy-duty truck battery power system, characterized in that, include: The building module is used to construct a multi-scale image model of the battery power system, including a hierarchical model from individual battery cells to modules, the entire pack, the controller, and the drive motor. The first modeling module is used to perform multi-physics coupling modeling of electrochemical field, temperature field and stress field on the constructed multi-scale image model to obtain the coupled multi-scale image model. An update module is used to update the parameters of the coupled multi-scale imaging model based on real-time monitoring data and a data-driven model to generate a preliminary digital twin model. The parameter update includes: collecting real-time monitoring data of the system operation using sensors pre-deployed in the battery power system to obtain a multidimensional dataset containing the real-time state of the battery power system; performing dimensionality reduction on the multidimensional dataset, and then combining it with a parameter sensitivity index for feature filtering to form a set of effective information related to the parameters of the multi-scale imaging model. The parameter sensitivity index is used to measure the degree of influence of the parameters on the model output. The information in the effective information set is compared and matched with the parameters of the coupled multi-scale imaging model; wherein, a dynamic threshold matching algorithm is used to determine the parameters that need to be updated and their update values, specifically including: constructing the deviation ratio Δk of the k-th parameter: , Indicates the first The deviation ratio of each parameter, It is the first Measured values ​​of each parameter For the first The model predictions of the parameters. This is the reference value for this parameter; Set the dynamic threshold θk: , Based on the threshold, and The adjustment coefficient is SOH, which represents the battery health state, and SOC, which represents the state of charge. The dynamic threshold is dynamically adjusted according to the battery health state and the state of charge. When the parameter needs to be updated, a parameter update list is generated, which records the name of the parameter to be updated, its current value, the target update value, and the priority identifier. Based on the parameter update list, a hierarchical iterative update strategy is used to correct the parameters of the coupled multi-scale imaging model. Update weights are assigned according to the parameter's level, with cell layer parameters having a greater update weight than the entire package layer parameters. For independent parameters, the updated values ​​directly replace the original parameters. For electrochemical-thermal-stress coupled parameters with a coupling relationship, a collaborative update algorithm is used for iterative correction. , This represents the updated value of the i-th parameter. This is the original value of the parameter. For the calculated parameter update amount, To update the weight coefficients, the updated weight coefficients are determined based on the parameter's hierarchical level and coupling relationship; during the iterative correction process, the calculation is continuously iterated and the model parameters are continuously adjusted until the error between the model output and the measured data converges to a preset threshold range; The second modeling module is used to perform dynamic time-series modeling of sensor data in the preliminary digital twin model using a graph neural network-long short-term memory neural network structure, in order to obtain the final digital twin model.

7. The digital twin model construction system for the battery power system of new energy heavy trucks as described in claim 6, characterized in that, The building module is specifically used for: Based on the energy conservation equation, an electrochemical-thermal model at the cell scale is constructed on top of the P2D model, resulting in a model structure that integrates electrochemical and thermal properties at the cell scale. Based on the cell-scale basic model, considering the thermal stress and thermal expansion caused by temperature gradient during battery operation, a single-cell-scale thermo-mechanical coupling model is constructed to form a comprehensive model that includes cell and single-cell-scale characteristics. Based on the comprehensive model, starting from the power battery module and the whole pack scale, a three-dimensional CAD model corresponding to the real shape of the physical battery is established using industrial modeling software. This accurately presents key structural parameters, constraints and positioning relationships between individual cells and modules, and obtains a three-dimensional model containing multi-scale structural information. Pixyz Studio was used to perform surface subdivision, topology correction, extraction, and repair operations on a 3D model containing multi-scale structural information. Then, combined with Hertz theory and extended finite element method, an overall digital physical model was established, the semantic information annotation objects were clarified, and the construction of a multi-scale image model of the battery power system was completed.

8. The digital twin model construction system for the battery power system of new energy heavy trucks as described in claim 6, characterized in that, The first modeling module is specifically used for: For the completed multi-scale imaging model of the battery power system, we comprehensively sort out the parameter information related to the electrochemical field, temperature field and stress field, and obtain the parameter set of each physical field at different scales; Based on the parameter set, we analyze the transmission paths and methods of each physical field parameter between different levels of models, clarify the rules of parameter transmission, and obtain a specific description of the rules of parameter transmission. Based on the obtained parameter transfer rule description, the interaction mechanism between different physical fields and different level models is studied, the influencing factors of the interaction are determined, and a list of key factors affecting the interaction is obtained. Based on the parameter set, parameter transfer rule description, and list of key factors affecting interaction, the multi-scale imaging model is coupled and integrated with multi-physics fields to obtain the coupled multi-scale imaging model.

9. The digital twin model construction system for the battery power system of new energy heavy trucks as described in claim 6, characterized in that, The update module is specifically used for: By pre-deploying sensors in the battery power system, real-time monitoring data during system operation is collected to obtain a dataset containing the real-time status of the battery power system. Data-driven methods are used to extract information from datasets containing the real-time status of battery power systems, and effective information related to the parameters of multi-scale imaging models is filtered out to form an effective information set. The information in the effective information set is compared and matched with the parameters of the coupled multi-scale imaging model to determine the parameters that need to be updated and their update values, thus obtaining a parameter update list. Based on the parameter update list, the parameters of the coupled multi-scale imaging model are updated to generate a preliminary digital twin model.

10. The digital twin model construction system for the battery power system of new energy heavy trucks as described in claim 6, characterized in that, The second modeling module is specifically used for: The sensor data in the preliminary digital twin model is abstracted into a network topology diagram of the sensor data time-series accumulation process according to the time series and spatial relationship, so as to obtain a topology structure that reflects the spatiotemporal characteristics of sensor data; A graph neural network-long short-term memory neural network structure is used to process the topological structure that reflects the spatiotemporal characteristics of sensor data, and integrate the structural information and temporal information to form intermediate data that fuses spatiotemporal characteristics. By utilizing the designed hierarchical feature aggregation method, different feature aggregators are learned on intermediate data with fused spatiotemporal features at different depths of neighborhood, thereby obtaining aggregated data containing multi-scale features. Dynamic time-series modeling of sensor data is performed, and aggregated data containing multi-scale features is incorporated into the model to obtain the final digital twin model.

Citation Information

Patent Citations

  • Multi-source heterogeneous twin data fusion method and system based on factory production process

    CN114595773A

  • Method for constructing digital twin model of power battery

    CN117574762A