Circuit structure based on resistance device and circuit design method

Abstract

The invention belongs to the technical field of circuit design, and discloses a circuit structure based on a resistance device and a circuit design method, which comprises the following steps: performing temperature field scanning on a battery module through a thermal imager to obtain thermal imaging data, associating resistance-temperature characteristics and thermal diffusion rate parameters of the resistance device, and calculating the resistance-temperature characteristics of the resistance device; establishing a joint calibration parameter library of the resistance device and thermal imaging, designing an embedded layout structure of the resistance device on a circuit board, arranging a heat conduction material layer on the surface of the resistance device, and integrating a digital function module; a thermal imaging sensor is integrated on a circuit board to form space alignment with a resistance device module, a multi-modal data fusion interface is designed, and temperature field data and resistance device signals are synchronously transmitted to a main control unit; optimizing the layout of the resistance device and the digital function module by optimizing the time sequence correlation analysis of the learning model on the temperature field data and the resistance device signal; through thermal imaging data joint calibration and embedded layout optimization, reliable data support is provided for an intelligent charging and discharging strategy of the BMS.

Description

Technical Field

[0001] The present invention relates to the technical field of circuit design, and particularly to a circuit structure based on a resistive device and a circuit design method. Background Art

[0002] With the rapid development of fields such as new energy vehicles and energy storage systems, the requirements for the accuracy and real-time performance of temperature monitoring in battery management systems (BMS) are becoming increasingly stringent. As the core component for temperature sensing, the integration process of thermistors (NTC) directly affects the safety and lifespan of battery packs. Under complex working conditions, the temperature distribution inside the battery module is uneven and changes violently. The traditional integration method is difficult to accurately capture local temperature mutations, leading to a risk of thermal runaway.

[0003] In the prior art, the circuit integration of thermistors mostly adopts fixed-position mounting or discrete component layout, and relies on empirical design to determine the installation area of the thermistors. For example, by evenly distributing multiple thermistors on the surface of the battery module and combining a simple linear compensation algorithm to achieve temperature acquisition. However, due to the lack of dynamic modeling of the heat diffusion path, the matching degree between the layout of the thermistors and the real temperature field is insufficient: during high-rate charging and discharging of the battery, local hot spots may deviate from the preset monitoring points, and the resistive devices cannot respond in time due to their fixed positions, resulting in temperature detection delay or missed detection. In addition, the discrete layout is easily affected by environmental heat interference, and the long signal transmission path further amplifies the temperature drift error. This defect directly limits the reliability of temperature data and is difficult to support the intelligent charging and discharging strategy optimization of the BMS.

[0004] In view of this, it is necessary to improve the circuit integration technology of thermistors in the prior art to solve the technical problems of unreasonable layout and low reliability. Summary of the Invention

[0005] The purpose of the present invention is to provide a circuit structure based on a resistive device and a circuit design method to solve the above technical problems.

[0006] To achieve this purpose, the present invention adopts the following technical solutions: A circuit design method based on a resistive device, comprising: Scanning the temperature field of the battery module with a thermal imager to obtain thermal imaging data, associating the resistance-temperature characteristics of the resistive device with the thermal diffusion rate parameters, and establishing a joint calibration parameter library of the resistive device and the thermal imaging; Based on the joint calibration parameter library, designing an embedded layout structure of the resistive device on the circuit board, and determining the high-temperature sensitive area according to the thermal imaging data to optimize the embedded layout structure of the resistive device; Setting a heat-conducting material layer on the surface of the resistive device in the embedded layout structure and integrating a digital function module; Integrate a thermal imaging sensor in the reserved area of the circuit board, form a spatial alignment with the resistor device module, design a multimodal data fusion interface, and synchronously transmit the temperature field data and the resistor device signal to the main control unit; Through optimizing the learning model for the timing correlation analysis of the temperature field data and the resistor device signal, optimize the layout of the resistor device and the digital function module.

[0007] Optionally, use a thermal imager to perform a temperature field scan on the battery module to obtain thermal imaging data, correlate the resistance-temperature characteristics of the resistor device with the thermal diffusion rate parameters, and establish a joint calibration parameter library for the resistor device and the thermal imaging, specifically including: Set the test conditions for the dynamic working conditions of the battery module, including charge and discharge rates, ambient temperature gradient, and heat dissipation boundary parameters, and construct a multi-dimensional thermal field scan reference framework; Based on the operation under the test conditions, use a thermal imager to perform a multimodal temperature field scan on the battery module, synchronously collect the thermal imaging data and the real-time resistance value change data of the resistor device, and generate an original data set of temperature-resistance correlated in time series; Extract the characteristic parameters from the original data set, and establish a resistance-temperature characteristic curve of the thermal imaging temperature field and the resistor device resistance value; the characteristic parameters include temperature gradient distribution, thermal diffusion rate, and resistor device response delay time.

[0008] Optionally, after extracting the characteristic parameters from the original data set and establishing a resistance-temperature characteristic curve of the thermal imaging temperature field and the resistor device resistance value, it further includes: Perform dynamic correlation analysis on the resistance-temperature characteristic curve of the resistor device and the thermal diffusion rate parameters, calculate the calibration coefficient matrix of the resistor device through a multivariate nonlinear regression model, and generate a dynamic calibration initial parameter table; Perform error compensation on the dynamic calibration initial parameter table based on historical working condition data, use the random forest algorithm to optimize the local hot spot response weight of the resistor device, and output a joint calibration parameter table after error correction; Bind the joint calibration parameter table with the spatial coordinate data of the thermal imaging temperature field, and construct a joint calibration parameter library including temperature-resistance-thermal diffusion rate.

[0009] Optionally, based on the joint calibration parameter library, design an embedded layout structure of the resistor device on the circuit board, and determine the high-temperature sensitive area according to the thermal imaging data, and optimize the embedded layout structure of the resistor device, specifically including: Analyze the joint calibration parameter library, use the temperature gradient distribution area exceeding the preset temperature threshold as the high-temperature sensitive area, and generate a distribution map of the dynamic thermal sensitive area of the battery module with the coordinates of the high-temperature sensitive area and the corresponding thermal diffusion rate parameters; Based on the said distribution map, a three-dimensional model of the circuit board is built using an electromagnetic-thermal coupling simulation tool, and an embedded layout structure of the resistor device is designed so that the detection end of the resistor device forms a direct heat conduction path with the high-temperature sensitive area of the battery electrode; According to the thermal diffusion rate parameter, the embedding depth and spacing of the resistor device are optimized; Perform a thermal stress test on the said embedded layout structure, simulate the temperature gradient change under the battery charge and discharge conditions, and output the embedded layout structure optimized for anti-interference by iteratively adjusting the distribution density and embedding angle of the resistor device; According to the optimized embedded layout structure, a design file containing the embedding coordinates and routing path of the resistor device is generated.

[0010] Optionally, the digital function module includes a shielding layer, a signal conditioning module, and a digital preprocessing module. Among them, the signal conditioning module and the digital preprocessing module are respectively integrated on the circuit board through the shielding layer; The signal conditioning module is used to realize the real-time conversion of resistance-temperature data, and the digital preprocessing module is used to realize the noise reduction output of the converted resistance-temperature data.

[0011] Optionally, a thermal imaging sensor is integrated in the reserved area of the circuit board, spatially aligned with the resistor device module, and a multi-modal data fusion interface is designed to synchronously transmit the temperature field data and the resistor device signal to the main control unit, specifically including: Analyze the embedding coordinates of the resistor device and the thermal imaging calibration parameter library in the embedded layout model to determine the installation position and field of view coverage range of the thermal imaging sensor in the reserved area of the circuit board; Set up a miniaturized optical window on the surface of the installation position, install the thermal imaging sensor, and adjust the lens angle of the thermal imaging sensor through a laser aligner so that the center line of its field of view forms a spatial coordinate mapping relationship with the detection end of the resistor device; Configure a dual-channel data interface through a fusion controller pre-integrated in the embedded layer of the circuit board. The dual-channel data interface is respectively connected to the I2C bus of the thermal imaging sensor and the SPI output port of the resistor device module.

[0012] Optionally, after configuring the dual-channel data interface through a fusion controller pre-integrated in the embedded layer of the circuit board, it further includes: Based on the thermal diffusion rate parameter and the timestamp synchronization protocol, set a dynamic priority allocation algorithm to perform timing alignment and data packet encapsulation on the thermal imaging temperature field data and the resistor device signal; Embed a parallel processing module in the fusion controller to realize the real-time weighted fusion of the temperature field data and the point measurement data, and generate a composite temperature data set with a confidence rating; Transmit the composite temperature dataset to the main control unit through a high - speed communication interface, and reserve a calibration instruction channel in the transmission protocol to support the collaborative control of the thermal imaging sensor and the resistive device module by the main control unit.

[0013] Optionally, the timing correlation analysis of the temperature field data and the resistive device signals by optimizing the learning model to optimize the layout of the resistive device and the digital function module specifically includes: Collect the timing data streams of the composite temperature dataset and the resistive device signals, and extract the eigenvectors of the temperature gradient change rate, the resistive device response delay, and the spatial correlation through a sliding window segmentation algorithm. Construct an optimization model based on deep reinforcement learning, take the eigenvectors and the initial parameters of the embedded layout model as inputs, and define a reward function with the temperature detection accuracy, response speed, and signal stability as the objectives. Use the historical working condition dataset to perform offline training on the optimization model, and dynamically adjust the layout density of the resistive device and the power consumption allocation weight of the digital function module through the policy gradient algorithm to generate a preliminary optimized parameter set.

[0014] Optionally, after using the historical working condition dataset to perform offline training on the optimization model, dynamically adjusting the layout density of the resistive device and the power consumption allocation weight of the digital function module through the policy gradient algorithm to generate a preliminary optimized parameter set, it further includes: Import the preliminary optimized parameter set into the digital twin simulation platform, simulate the collaborative response of the resistive device and the digital function module during the battery charge - discharge process, and output the quantitative evaluation results of the thermal field coverage efficiency and the signal - to - noise ratio. Reverse - correct the optimized parameter set according to the quantitative evaluation results, and use the genetic algorithm to iteratively update the buried coordinates of the resistive device and the shielding layer structure parameters of the digital function module to generate a final layout parameter package with enhanced anti - interference ability. Burn the final layout parameter package into the firmware of the main control unit, and perform closed - loop verification on the battery test platform, and dynamically update the policy network weights of the optimization model according to the real - time feedback data.

[0015] The present invention also provides a circuit structure based on resistive devices, adopting the circuit design method based on resistive devices as described above. The circuit structure specifically includes a circuit board and the following components integrated on the circuit board: Resistive devices, buried in the high - temperature sensitive area of the circuit board based on the embedded layout model. Thermal conduction structure, a gradient boron nitride - silica gel thermal conduction layer covering the surface of the resistive device. Integrated digital function module, used to realize real - time resistance - temperature conversion and dynamic noise reduction. A thermal imaging sensor is installed in the reserved area of the circuit board and is spatially aligned with the resistive device through a miniaturized optical window; A multimodal data fusion interface includes a dual-channel data interface and a fusion controller integrated in the embedded layer of the circuit board; A main control unit, with a processor equipped with an optimized learning model, dynamically adjusts the layout parameters of the resistive device and the charge and discharge strategy according to the results of the timing correlation analysis.

[0016] Compared with the prior art, the present invention has the following beneficial effects: First, use a thermal imager to scan the temperature field of the battery module to obtain thermal imaging data; then, establish a joint calibration parameter library of the resistive device and the thermal imaging by correlating the resistance-temperature characteristics of the resistive device and the thermal diffusion rate parameters; next, design an embedded layout structure of the resistive device on the circuit board based on this parameter library, determine the high-temperature sensitive area according to the thermal imaging data, optimize the layout structure, set a heat-conducting material layer on the surface of the resistive device, and integrate digital function modules; then, integrate a thermal imaging sensor in the reserved area of the circuit board, form a spatial alignment with the resistive device module, and design a multimodal data fusion interface to synchronously transmit the temperature field data and the resistive device signal to the main control unit; perform a timing correlation analysis on the temperature field data and the resistive device signal through the optimized learning model to further optimize the layout of the resistive device and the digital function module; this method significantly improves the detection accuracy and response speed through the joint calibration of thermal imaging data and the optimization of the embedded layout, solves the problem of local hot spot monitoring failure, and provides reliable data support for the intelligent charge and discharge strategy of the BMS. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] The structures, ratios, sizes, etc. shown in the drawings of this specification are only used to cooperate with the content disclosed in the specification for those familiar with this technology to understand and read, and are not used to limit the limited conditions under which the present invention can be implemented. Therefore, they do not have technical substantive significance. Any modification of the structure, change of the ratio relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope that the technical content disclosed by the present invention can cover.

[0019] Figure 1 It is one of the flow diagrams of the circuit design method based on resistive devices in the first embodiment; Figure 2It is the second flow schematic diagram of the circuit design method based on a resistive device in the first embodiment; Figure 3 It is the third flow schematic diagram of the circuit design method based on a resistive device in the first embodiment. Detailed implementation manners

[0020] To make the invention purpose, features, and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0021] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "upper", "lower", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. It should be noted that when a component is considered to be "connected" to another component, it may be directly connected to the other component or there may be intermediate components present at the same time.

[0022] The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and through specific implementation manners.

[0023] Embodiment 1: Combined with Figures 1 to 3 As shown, the embodiment of the present invention provides a circuit design method based on a resistive device, including: S1, performing a temperature field scan on the battery module through an infrared thermal imager to obtain thermal imaging data, associating the resistance-temperature characteristics of the resistive device with the thermal diffusion rate parameter, and establishing a joint calibration parameter library for the resistive device and the thermal imaging; it should be noted that the infrared thermal imager here is an external test device for performing a temperature field scan on the battery module. It should be noted that the resistive device in this solution is a thermistor (NTC).

[0024] Through the joint calibration of the thermal imaging data and the resistive device signal, the accuracy and response speed of temperature monitoring are significantly improved, and the local temperature change of the battery module can be captured more accurately.

[0025] S2. Based on the joint calibration parameter library, design the embedded layout structure of the resistive device on the circuit board, and determine the high-temperature sensitive area according to the thermal imaging data, and optimize the embedded layout structure of the resistive device; the scope of optimizing the embedded layout structure includes optimizing the anti-interference structure of the buried position and the signal transmission path.

[0026] S3. Set a heat-conducting material layer on the surface of the resistive device in the embedded layout structure, and integrate digital function modules; By optimizing the embedded layout of the resistive device and the heat-conducting material coating, the environmental thermal interference is effectively reduced, and the stability of temperature measurement is enhanced.

[0027] S4. Integrate a thermal imaging sensor in the reserved area of the circuit board, form a spatial alignment with the resistive device module, design a multi-modal data fusion interface, and synchronously transmit the temperature field data and the resistive device signal to the main control unit; Integrating digital function modules and multi-modal data fusion interfaces realizes the real-time processing and transmission of temperature data, and improves the intelligent level of the system.

[0028] S5. Through the optimized learning model for the timing correlation analysis of the temperature field data and the resistive device signal, optimize the layout of the resistive device and the digital function module.

[0029] Through the timing correlation analysis of the optimized learning model, the layout of the resistive device and the temperature control strategy are further optimized, the battery life is extended, and the overall reliability of the system is improved.

[0030] The working principle of the present invention is as follows: First, use a thermal imager to scan the temperature field of the battery module to obtain thermal imaging data; then, establish a joint calibration parameter library of the resistive device and the thermal imaging by correlating the resistance-temperature characteristics of the resistive device and the thermal diffusion rate parameters; then, based on this parameter library, design the embedded layout structure of the resistive device on the circuit board, determine the high-temperature sensitive area according to the thermal imaging data, optimize the layout structure, set a heat-conducting material layer on the surface of the resistive device, and integrate digital function modules; then integrate a thermal imaging sensor in the reserved area of the circuit board, form a spatial alignment with the resistive device module, and design a multi-modal data fusion interface to synchronously transmit the temperature field data and the resistive device signal to the main control unit; through the optimized learning model, perform timing correlation analysis on the temperature field data and the resistive device signal, and further optimize the layout of the resistive device and the digital function module; this method significantly improves the detection accuracy and response speed through the joint calibration of thermal imaging data and the optimization of the embedded layout, solves the problem of local hot spot monitoring failure, and provides reliable data support for the intelligent charge and discharge strategy of the BMS.

[0031] In this embodiment, specifically, step S1 specifically includes: S11. Set the test conditions for the dynamic working conditions of the battery module, including charge-discharge rate, environmental temperature gradient, and heat dissipation boundary parameters, and construct a multi-dimensional thermal field scanning reference framework; By defining the dynamic working condition test conditions of the battery module (including charge-discharge rate, environmental temperature gradient, and heat dissipation boundary parameters), simulate the extreme working states of the battery in actual applications (such as rapid charge and discharge, high and low temperature environment switching). The charge-discharge rate is set based on the ISO 12405-3 standard, covering the range from 1C to 5C to trigger different thermal power densities; the environmental temperature gradient is simulated by a PID temperature control system for a cyclic change from -20°C to 60°C to reflect the thermal shock in the real scenario; the heat dissipation boundary parameters are quantified by calculating the contact thermal resistance and convective heat transfer coefficient between the battery module and the heat dissipation structure to limit the heat dissipation capacity. The constructed multi-dimensional thermal field scanning reference framework provides a controllable and reproducible test environment for subsequent data collection, ensuring the engineering applicability of the calibrated parameters.

[0032] S12. Based on the test conditions, use a thermal imager to perform multi-modal temperature field scanning on the battery module, synchronously collect thermal imaging data and the real-time resistance value change data of the resistive device, and generate an original data set of temperature-resistance associated with time series; A high-resolution infrared thermal imager (such as FLIR A65) and a multi-channel data acquisition card work together to achieve millisecond-level synchronous acquisition of thermal imaging data and the resistance value of the resistive device. The thermal imager performs multi-modal scanning on the surface and inside of the battery module (cooperating with an infrared-transmitting window) through a preset scanning path, covering steady-state, transient, and local hot spot working conditions; the synchronous trigger signal ensures that the time stamps of the thermal imaging frame data and the resistance value of the resistive device are strictly aligned. The generated original data set includes a temperature field distribution matrix, a time series curve of the resistance value of the resistive device, and a working condition label, providing a multi-source heterogeneous data basis for subsequent feature extraction.

[0033] S13. Extract characteristic parameters from the original data set and establish a resistance-temperature characteristic curve between the thermal imaging temperature field and the resistance value of the resistive device; the characteristic parameters include temperature gradient distribution, heat diffusion rate, and response delay time of the resistive device.

[0034] Based on the wavelet transform and sliding window algorithm, three types of key features are extracted from the original data: Temperature gradient distribution: By calculating the temperature difference matrix of adjacent pixel points, identify the heat diffusion direction and rate; Heat diffusion rate: Use the Fourier heat conduction equation inversion and combine the spatial gradient of the thermal imaging data to calculate the dynamic thermal conductivity; Response delay time: Use the cross-correlation analysis method to quantify the time delay of the change in the resistance value of the resistive device lagging behind the thermal imaging temperature change.

[0035] Input the above features into the non - linear least - squares fitting model to establish the mapping relationship between the resistance value of the resistor device and the temperature field, and generate a segmented resistance - temperature characteristic curve covering the full range from - 30°C to 150°C.

[0036] S14. Perform dynamic correlation analysis on the resistance - temperature characteristic curve of the resistor device and the thermal diffusion rate parameter, calculate the calibration coefficient matrix of the resistor device through a multivariate non - linear regression model, and generate a dynamic calibration initial parameter table. To overcome the limitations of traditional linear calibration in a non - linear thermal field, a multivariate non - linear regression model is constructed, the optimal coefficient matrix is iteratively solved, and a dynamic calibration initial parameter table is generated. The initial parameter table includes the resistor device ID, calibration coefficient, effective temperature range, and confidence interval, supporting the BMS for differential compensation of different resistor devices.

[0037] S15. Perform error compensation on the dynamic calibration initial parameter table based on historical operating condition data, use the random forest algorithm to optimize the local hot - spot response weight of the resistor device, and output a jointly calibrated parameter table after error correction. Regarding the calibration deviation of the initial parameter table in the local hot - spot scenario, a random forest algorithm is introduced for error compensation. The error compensation process is as follows: Feature construction: Extract multi - dimensional features such as operating condition type, thermal diffusion direction, and battery SOC from historical data. Model training: Use the absolute value of the calibration error as the target variable, construct a forest model containing 100 decision trees, and optimize the hot - spot response weight through feature importance ranking. Parameter correction: Apply a weight compensation term to the calibration coefficient in the initial parameter table, and the corrected parameter table reduces the local hot - spot detection error to within the error tolerance range.

[0038] S16. Bind the jointly calibrated parameter table with the spatial coordinate data of the thermal imaging temperature field to construct a jointly calibrated parameter library containing temperature - resistance - thermal diffusion rate.

[0039] Encode and bind the jointly calibrated parameter table with the spatial coordinates of the thermal imaging temperature field (based on the circuit layout CAD model) to construct a hierarchical parameter library.

[0040] In this embodiment, specifically, step S2 specifically includes: S21. Analyze the jointly calibrated parameter library, use the temperature gradient distribution area exceeding the preset temperature threshold as the high - temperature sensitive area, and generate a distribution map of the dynamic thermal sensitive area of the battery module based on the coordinates of the high - temperature sensitive area and the corresponding thermal diffusion rate parameters. By analyzing the temperature gradient distribution data (based on thermal imaging scan results) in the joint calibration parameter library, a threshold determination algorithm is used to identify high-temperature sensitive areas where the temperature exceeds a preset threshold (such as 60 °C). Combining thermal diffusion rate parameters (such as thermal conductivity, heat capacity ratio), a spatial interpolation algorithm is used to generate a dynamic thermal sensitive area distribution map. This map marks the positions of high-temperature sensitive areas, the direction of heat diffusion (represented by arrow vectors), and the rate level (color gradient encoding) in a three-dimensional coordinate system, providing a quantitative basis for subsequent layout design. The distribution map is output through visualization tools such as MATLAB or Python, and supports seamless docking with the coordinate system of PCB design software (such as Altium Designer).

[0041] S22, based on the distribution map, use an electromagnetic-thermal coupling simulation tool to perform three-dimensional modeling of the circuit board, and design an embedded layout structure for the resistor device, so that the detection end of the resistor device forms a direct heat conduction path with the high-temperature sensitive area of the battery electrode; Use a multi-physics field simulation tool to perform three-dimensional modeling of the circuit board based on the distribution map, and set electromagnetic-thermal coupling boundary conditions: Thermal boundary: The contact surface of the battery electrode is set to a constant heat flux density (calculated according to the charge and discharge rate), and the thermal conductivity of the circuit board substrate (such as FR-4) is set to anisotropic parameters; Electromagnetic boundary: Set a near-field radiation interference model for frequencies below 1 GHz.

[0042] Optimize the embedded layout of the resistor device through simulation to ensure that its detection end forms a low thermal resistance path with the battery electrode through copper posts or thermal vias. In the layout design, the embedded position of the resistor device needs to avoid high-frequency signal lines.

[0043] S23, according to the thermal diffusion rate parameters, optimize the embedded depth and spacing of the resistor device; preferably, differential signal transmission lines and electromagnetic shielding layers can be embedded in the circuit board routing layer to reduce the temperature drift error of the signal transmission path.

[0044] S24, perform a thermal stress test on the embedded layout structure, simulate the temperature gradient change under the battery charge and discharge conditions, and output the embedded layout structure optimized for anti-interference by iteratively adjusting the distribution density and embedded angle of the resistor device; Simulate the battery charge and discharge conditions in a thermal cycling test chamber, and perform a thermo-mechanical stress test on the embedded layout structure. Monitor the deformation of the circuit board through digital image correlation technology, and evaluate the thermal stress distribution in combination with finite element analysis software. According to the test results, use the gradient descent algorithm to dynamically adjust the distribution density and embedded angle of the resistor device (tilted 5°-10° to match the heat flow direction) to reduce the maximum thermal stress value below the material yield strength. The stability of the optimized layout structure is verified through the anti-interference index.

[0045] S25. Generate a design file containing the buried coordinates of the resistor device and the routing path according to the optimized embedded layout structure.

[0046] Based on the optimized embedded layout structure, use EDA tools to generate a design-for-manufacturability file, including: A buried coordinate file that defines the X / Y / Z coordinates and polarity identification of the resistor device in a standard way; A routing path diagram: Adopt differential pair routing and shield ground wire surrounding design, and mark the impedance control requirements (single-ended 50Ω ± 10%); Process description: Specify the processing parameters of the buried hole (laser drilling diameter 0.3mm, depth-width ratio ≤ 8:1) and the soldering temperature curve (peak temperature 245°C ± 5°C, duration ≤ 30s).

[0047] In this embodiment, specifically, the digital function module includes a shielding layer, a signal conditioning module, and a digital preprocessing module. Among them, the signal conditioning module and the digital preprocessing module are respectively integrated on the circuit board through the shielding layer; the signal conditioning module is used to realize the real-time conversion of resistance-temperature data, and the digital preprocessing module is used to realize the noise reduction output of the converted resistance-temperature data.

[0048] In this embodiment, specifically, step S4 specifically includes: S41. Analyze the buried coordinates of the resistor device and the thermal imaging calibration parameter library in the embedded layout model to determine the installation position and field of view coverage of the thermal imaging sensor in the reserved area of the circuit board; By analyzing the spatial mapping relationship between the buried coordinates of the resistor device in the embedded layout model and the thermal imaging calibration parameter library, use the geometric projection algorithm to calculate the installation position of the thermal imaging sensor. Based on the detection end coordinates of the resistor device (3D point cloud data) and the sensor field of view angle (such as 60° × 45°), use the field of view coverage optimization model to ensure that the sensor field of view covers all high-temperature sensitive areas. The installation position coordinates are verified through the DRC module of the PCB design software to avoid conflicts with high-speed signal lines or power layers, and a 2mm edge gap is reserved to meet the thermal expansion tolerance.

[0049] S42. Set a miniaturized optical window on the surface of the installation position, install the thermal imaging sensor, and adjust the lens angle of the thermal imaging sensor through a laser aligner so that the center line of its field of view forms a spatial coordinate mapping relationship with the detection end of the resistor device; A micro-optical window is etched on the surface of the circuit board installation position using a laser micromachining process. Sapphire glass with an infrared antireflection film coating is selected as the window material. After installing a thermal imaging sensor (such as MLX90640), the pitch angle and yaw angle of the lens are adjusted through a six-degree-of-freedom laser aligner so that the center line of its field of view forms a sub-millimeter spatial mapping with the detection end of the resistive device. During the alignment process, the thermal imaging images and the coordinate data of the resistive devices are collected in real time, and the alignment parameters are optimized through a least squares fitting algorithm to ensure that each resistive device occupies at least a 5×5 pixel area in the thermal imaging image.

[0050] S43, Configure a dual-channel data interface through a fusion controller integrated in the circuit board embedded layer in advance. The dual-channel data interface is respectively connected to the I2C bus of the thermal imaging sensor and the SPI output port of the resistive device module; among them, the multi-modal data fusion interface includes a fusion controller and a dual-channel data interface.

[0051] S44, Based on the thermal diffusion rate parameter and the timestamp synchronization protocol, set a dynamic priority allocation algorithm to perform timing alignment and data packet encapsulation on the thermal imaging temperature field data and the resistive device signals; Design a dynamic priority allocation algorithm based on the thermal diffusion rate parameter (unit: mm2 / s): When the thermal diffusion rate ≥ 1.5 mm2 / s, the priority of the thermal imaging data is set to high (the sampling rate is increased to 30 Hz), and the priority of the resistive device data is set to medium (10 Hz); when the rate < 0.5 mm² / s, the priority is reversed to reduce power consumption.

[0052] Add a unified timestamp to the two types of data through a time protocol, and encapsulate the data packets in an encoded format. Each data packet contains a temperature field matrix, the ID of the resistive device, and its resistance-temperature conversion result, and supports efficient DMA (Direct Memory Access) transmission.

[0053] S45, Embed a parallel processing module in the fusion controller to achieve real-time weighted fusion of the temperature field data and the point measurement data, and generate a composite temperature data set with a confidence rating; Deploy an FPGA parallel processing module in the fusion controller to achieve real-time data fusion: Spatial alignment, interpolate the thermal imaging temperature field data to the resistive device coordinate grid (bilinear interpolation algorithm); Weighted fusion: Dynamically allocate weights according to the thermal diffusion rate (thermal imaging weight = 0.7 - 0.9, resistive device weight = 0.1 - 0.3); Confidence calculation: Generate a confidence rating (0 - 100%) based on the residual sum of squares (RSS) and the signal strength.

[0054] The fused composite temperature dataset is stored in the form of a structure, including temperature values (floating-point numbers), confidence levels (8-bit integers), and timestamps (64-bit integers), for the main control unit to call.

[0055] S46. Transmit the composite temperature dataset to the main control unit through a high-speed communication interface, and reserve a calibration instruction channel in the transmission protocol to support the coordinated control of the thermal imaging sensor and the resistance device module by the main control unit.

[0056] Use a USB 3.0 SuperSpeed interface to transmit the composite dataset to the main control unit. The data packet encapsulation follows a custom protocol. A calibration instruction channel is reserved in the protocol to support the main control unit to issue instructions: thermal imaging calibration, trigger blackbody correction; resistance device self-test, send the 0xAA instruction to start the resistance value reference calibration mode.

[0057] The coordinated control instructions are responded to in real time through an interrupt mechanism (edge trigger) to ensure the adaptive adjustment ability of the temperature monitoring system under all working conditions.

[0058] In this embodiment, specifically, step S5 specifically includes: S51. Collect the timing data streams of the composite temperature dataset and the resistance device signals, and extract the feature vectors of the temperature gradient change rate, the response delay of the resistance device, and the spatial correlation through a sliding window segmentation algorithm. Synchronously capture the composite temperature dataset (including confidence ratings) and the original signals of the resistance device through a high-speed data acquisition card, and use a dynamic sliding window algorithm to segment the timing data stream. Extract the following features: Temperature gradient change rate, calculate the temperature field gradient change (ΔT / Δt, unit: °C / s) of adjacent time segments through first-order difference. Response delay time, use the dynamic time warping algorithm to quantify the time delay of the resistance device signal lagging behind the thermal imaging data. Spatial correlation, based on the Pearson correlation coefficient matrix, analyze the temperature change correlation between the resistance device and the thermal imaging pixel points (threshold ≥ 0.7).

[0059] S52. Construct an optimization model based on deep reinforcement learning, use the feature vectors and the initial parameters of the embedded layout model as inputs, and define a reward function with temperature detection accuracy, response speed, and signal stability as the goals. Construct a deep deterministic policy gradient (DDPG) model, including: Actor network, 4-layer fully connected (256 - 128 - 64 - 32 nodes), ReLU activation, outputting layout density and power consumption weights; Critic network, dual-stream input (feature vector + layout parameters), outputting Q-value estimates.

[0060] S53. Use the historical operating condition dataset to perform offline training on the optimization model. Dynamically adjust the layout density of the resistive devices and the power consumption allocation weights of the digital functional modules through the policy gradient algorithm to generate a preliminary optimization parameter set.

[0061] Use the historical dataset (including several sets of charge and discharge operating conditions) for offline training, and update the model parameters using the Proximal Policy Optimization (PPO) algorithm: batch size, reward discount factor: γ = 0.99, GAE parameter: λ = 0.95.

[0062] After training, output the preliminary optimization parameter set, including the layout density of the resistive devices, the power consumption allocation weights of the digital modules, and the shielding layer material parameters.

[0063] S54. Import the preliminary optimization parameter set into the digital twin simulation platform to simulate the collaborative response of the resistive devices and the digital functional modules during the battery charge and discharge process, and output the quantitative evaluation results of the thermal field coverage efficiency and the signal-to-noise ratio. In the digital twin platform, import the optimization parameter set to construct a battery-circuit joint simulation model: for the thermal field coverage efficiency, calculate the area ratio of the monitoring area of the resistive devices to the high-temperature sensitive area; For the signal-to-noise ratio, extract the ratio of the signal power to the noise power within the preset bandwidth through frequency domain analysis.

[0064] S55. Perform reverse correction on the optimization parameter set according to the quantitative evaluation results, and use the genetic algorithm to iteratively update the buried coordinates of the resistive devices and the shielding layer structure parameters of the digital functional modules to generate a final layout parameter package with enhanced anti-interference ability.

[0065] S56. Burn the final layout parameter package into the firmware of the main control unit, and perform closed-loop verification on the battery test platform. Dynamically update the policy network weights of the optimization model according to the real-time feedback data. The verification results are output as the optimization model weight file and the process parameter update log.

[0066] Burn the final parameter package into the main control unit through the JTAG interface and perform closed-loop verification on the battery test platform: Real-time feedback: Collect the temperature data during the charge and discharge cycles, and update the policy network weights through the online learning mechanism; Dynamic adjustment: When it is detected that the battery capacity decay ≥ 10%, trigger the iterative update of the parameter library (the version number increases) to ensure adaptability throughout the life cycle.

[0067] Example 2: The present invention also provides a circuit structure based on resistive devices. Using the circuit design method based on resistive devices as in Example 1, the circuit structure specifically includes a circuit board, and integrated on the circuit board: Resistive devices, buried in the high-temperature sensitive area of the circuit board based on the embedded layout model.

[0068] A heat conduction structure, a gradient boron nitride-silica gel heat conduction layer covering the surface of the resistance device, and a coating material matching the non-linear heat conduction characteristics.

[0069] An integrated digital function module for realizing real-time resistance-temperature conversion and dynamic noise reduction.

[0070] A thermal imaging sensor is installed in the reserved area of the circuit board and is spatially aligned with the resistance device through a miniaturized optical window.

[0071] A multi-modal data fusion interface includes a dual-channel data interface and a fusion controller integrated in the embedded layer of the circuit board.

[0072] A main control unit, a processor equipped with an optimized learning model, dynamically adjusts the layout parameters and charge-discharge strategy of the resistance device according to the results of sequential correlation analysis.

[0073] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A circuit design method based on a resistor device, characterized in that: include: The temperature field of the battery module is scanned by a thermal imager to obtain thermal imaging data, and the resistance-temperature characteristics of the resistor device are correlated with the thermal diffusion rate parameters to establish a joint calibration parameter library of the resistor device and thermal imaging. Based on the joint calibration parameter library, the embedded layout structure of the resistor device on the circuit board is designed, and the high temperature sensitive area is determined according to the thermal imaging data to optimize the embedded layout structure of the resistor device; A thermally conductive material layer is provided on the surface of the resistor device of the embedded layout structure, and a digital functional module is integrated; Integrate a thermal imaging sensor in the reserved area of ​​the circuit board, align it with the resistor module, and design a multi-modal data fusion interface to synchronously transmit the temperature field data and the resistor signal to the main control unit. The layout of the resistor devices and digital function modules is optimized by analyzing the timing correlation between the temperature field data and the resistor device signals through an optimized learning model.

2. The circuit design method based on a resistor device according to claim 1, characterized in that: The method of scanning the temperature field of the battery module by a thermal imager to obtain thermal imaging data, correlating the resistance-temperature characteristics of the resistor device with the thermal diffusion rate parameters, and establishing a joint calibration parameter library of the resistor device and thermal imaging specifically includes: Set the test conditions of the dynamic working conditions of the battery module, including charge and discharge rate, ambient temperature gradient and heat dissipation boundary parameters, and build a multi-dimensional thermal field scanning benchmark framework; Based on the test conditions, a thermal imager is used to perform a multi-modal temperature field scan on the battery module, and thermal imaging data and real-time resistance change data of the resistor are synchronously collected to generate a time series-related temperature-resistance raw data set; Characteristic parameters are extracted from the original data set, and a resistance-temperature characteristic curve of the thermal imaging temperature field and the resistance value of the resistor device is established; the characteristic parameters include temperature gradient distribution, heat diffusion rate and resistor device response delay time.

3. The circuit design method based on a resistor device according to claim 2, characterized in that: Extract characteristic parameters of the original data set and establish a resistance-temperature characteristic curve of the thermal imaging temperature field and the resistance value of the resistor device, and then include: Dynamically correlate the resistance-temperature characteristic curve of the resistor device with the thermal diffusion rate parameter, calculate the calibration coefficient matrix of the resistor device through a multivariate nonlinear regression model, and generate a dynamic calibration initial parameter table; Based on historical working condition data, error compensation is performed on the dynamic calibration initial parameter table, a local hot spot response weight of the resistor device is optimized using a random forest algorithm, and an error-corrected joint calibration parameter table is output; The joint calibration parameter table is bound to the spatial coordinate data of the thermal imaging temperature field to construct a joint calibration parameter library including temperature-resistance-thermal diffusion rate.

4. The circuit design method based on a resistor device according to claim 3, characterized in that: The method of designing an embedded layout structure of a resistor device on a circuit board based on the joint calibration parameter library, determining a high temperature sensitive area according to thermal imaging data, and optimizing the embedded layout structure of the resistor device specifically includes: Parsing the joint calibration parameter library, taking the temperature gradient distribution area exceeding the preset temperature threshold as the high temperature sensitive area, and generating a distribution map of the dynamic heat sensitive area of ​​the battery module based on the coordinates of the high temperature sensitive area and the corresponding heat diffusion rate parameters; Based on the distribution diagram, an electromagnetic-thermal coupling simulation tool is used to perform three-dimensional modeling of the circuit board, and an embedded layout structure of the resistor device is designed so that a direct heat conduction path is formed between the detection end of the resistor device and the high temperature sensitive area of ​​the battery electrode; According to the thermal diffusion rate parameters, the buried depth and spacing of the resistor components are optimized; Performing a thermal stress test on the embedded layout structure to simulate the temperature gradient change under the battery charging and discharging conditions, and outputting an embedded layout structure after anti-interference optimization by iteratively adjusting the distribution density and embedding angle of the resistor device; According to the optimized embedded layout structure, a design file including the embedding coordinates and routing paths of the resistor device is generated.

5. The circuit design method based on resistor device according to claim 1, characterized in that: The digital function module includes a shielding layer, a signal conditioning module and a digital pre-processing module, wherein the signal conditioning module and the digital pre-processing module are respectively integrated on the circuit board through the shielding layer; The signal conditioning module is used to realize the real-time conversion of resistance-temperature data, and the digital preprocessing module is used to realize the noise reduction output of the converted resistance-temperature data.

6. The circuit design method based on resistor device according to claim 1, characterized in that: The thermal imaging sensor is integrated in the reserved area of ​​the circuit board, spatially aligned with the resistor device module, and a multi-modal data fusion interface is designed to synchronously transmit the temperature field data and the resistor device signal to the main control unit, specifically including: Parsing the embedded coordinates of the resistor device and the thermal imaging calibration parameter library in the embedded layout model to determine the installation position and field of view coverage of the thermal imaging sensor in the reserved area of ​​the circuit board; A miniaturized optical window is arranged on the surface of the installation position, a thermal imaging sensor is installed, and the lens angle of the thermal imaging sensor is adjusted by a laser alignment instrument so that a spatial coordinate mapping relationship is formed between the center line of the field of view and the detection end of the resistor device; A dual-channel data interface is configured through a fusion controller pre-integrated in the embedded layer of the circuit board, and the dual-channel data interface is respectively connected to the I2C bus of the thermal imaging sensor and the SPI output port of the resistor device module.

7. The circuit design method based on resistor device according to claim 6, characterized in that: The dual-channel data interface is configured by pre-integrating a fusion controller in the embedded layer of the circuit board, and then further includes: Based on the heat diffusion rate parameters and timestamp synchronization protocol, a dynamic priority allocation algorithm is set to perform timing alignment and data packet encapsulation on thermal imaging temperature field data and resistor device signals; A parallel processing module is embedded in the fusion controller to realize real-time weighted fusion of temperature field data and point measurement data, and generate a composite temperature data set with confidence rating; The composite temperature data set is transmitted to the main control unit through a high-speed communication interface, and a calibration instruction channel is reserved in the transmission protocol to support the main control unit's coordinated control of the thermal imaging sensor and the resistor device module.

8. The circuit design method based on resistor device according to claim 7, characterized in that: The optimization of the layout of the resistor device and the digital function module by analyzing the timing correlation between the temperature field data and the resistor device signal through the optimization learning model specifically includes: Collecting the composite temperature data set and the time series data stream of the resistor device signal, and extracting the characteristic vectors of the temperature gradient change rate, the resistor device response delay and the spatial correlation through a sliding window segmentation algorithm; An optimization model based on deep reinforcement learning is constructed, the feature vector and initial parameters of the embedded layout model are used as input, and a reward function with temperature detection accuracy, response speed, and signal stability as targets is defined; The optimization model is trained offline using a historical operating condition data set, and the layout density of resistor devices and the power consumption allocation weights of digital functional modules are dynamically adjusted through a policy gradient algorithm to generate a preliminary optimization parameter set.

9. The circuit design method based on resistor device according to claim 8, characterized in that: The optimization model is trained offline by using the historical working condition data set, and the layout density of the resistor device and the power consumption allocation weight of the digital function module are dynamically adjusted by the policy gradient algorithm to generate a preliminary optimization parameter set, and then the following is further included: Importing the preliminary optimized parameter set into the digital twin simulation platform, simulating the coordinated response of the resistor device and the digital functional module during the battery charging and discharging process, and outputting the quantitative evaluation results of the thermal field coverage efficiency and the signal-to-noise ratio; Reversely correct the optimized parameter set according to the quantitative evaluation result, iteratively update the embedding coordinates of the resistor device and the shielding layer structure parameters of the digital function module using a genetic algorithm, and generate a final layout parameter package with enhanced anti-interference; The final layout parameter package is burned into the firmware of the main control unit, and a closed-loop verification is performed on a battery test platform, and the strategy network weights of the optimization model are dynamically updated according to real-time feedback data.

10. A circuit structure based on a resistor device, characterized in that: According to any one of claims 1 to 9, the circuit design method based on a resistor device is adopted, wherein the circuit structure specifically comprises a circuit board, and integrated on the circuit board: Resistor devices are embedded in high temperature sensitive areas of the circuit board based on the embedded layout model; Thermally conductive structure, a gradient boron nitride-silicone thermal conductive layer covering the surface of the resistor device; Integrated digital function module to achieve real-time resistance-temperature conversion and dynamic noise reduction; The thermal imaging sensor is installed in the reserved area of ​​the circuit board and forms spatial alignment with the resistor device through a miniaturized optical window; A multimodal data fusion interface, including a dual-channel data interface and a fusion controller integrated in an embedded layer of a circuit board; The main control unit is equipped with a processor that optimizes the learning model and dynamically adjusts the layout parameters and charging and discharging strategies of the resistor devices according to the results of the timing correlation analysis.

Images