New energy wire harness anti-interference shielding optimization method and device, equipment and storage medium

Abstract

The invention relates to the technical field of anti-interference shielding, and discloses a new energy wire harness anti-interference shielding optimization method, device and equipment and a storage medium, and the method comprises the steps: sampling voltage and current signals at the two ends of a plurality of wire harnesses in a new energy automobile, and obtaining the original data of a plurality of wire harness signals; carrying out transformation and interpolation processing to obtain multiple pieces of wire harness signal harmonic component data; carrying out calculation and adaptive noise elimination to obtain a plurality of wire harness anti-interference performance indexes; filter parameter optimization calculation is carried out, and adaptive filter optimization parameters are obtained; creating a multi-stage filtering structure and carrying out eigenvalue decomposition orthogonal optimization to obtain multi-stage filtering structure parameters; according to the method, the whole vehicle electromagnetic compatibility of the new energy vehicle is optimized in a distributed mode, a whole vehicle wire harness anti-interference shielding optimization scheme is obtained, the method can dynamically adapt to the complex electromagnetic environment, wire harness anti-interference shielding optimization with local optimization and global optimization is considered, and the whole electromagnetic compatibility of the new energy vehicle is improved.

Description

Technical Field

[0001] The present application relates to the field of anti-interference shielding technology, and in particular to a new energy wiring harness anti-interference shielding optimization method, device, equipment and storage medium. Background Art

[0002] Wiring harnesses play an important role in power transmission and signal transfer in new energy vehicles, and their anti-interference performance directly affects the electromagnetic compatibility and reliability of the entire vehicle.

[0003] Traditional wire harness anti-interference shielding methods mainly rely on physical means such as adding shielding layers and changing materials, which are difficult to cope with complex and changeable electromagnetic interference environments. At the same time, due to the widespread application of high voltage, high current and high frequency signals in new energy vehicles, the crosstalk and radiation interference problems between wire harnesses are more prominent, and traditional static shielding solutions are difficult to meet the needs of dynamic optimization. In addition, vehicle electromagnetic compatibility optimization is a complex system engineering involving multiple subsystems and a large number of wire harnesses. Traditional centralized optimization methods have high computational complexity and are difficult to effectively handle large-scale optimization problems. Summary of the invention

[0004] The present application provides a method, device, equipment and storage medium for optimizing the anti-interference shielding of new energy wiring harnesses, which can dynamically adapt to complex electromagnetic environments, optimize the anti-interference shielding of wiring harnesses while taking into account both local and global optimization, and improve the overall electromagnetic compatibility performance of new energy vehicles.

[0005] The first aspect of the present application provides a method for optimizing the anti-interference shielding of a new energy wiring harness, and the method for optimizing the anti-interference shielding of a new energy wiring harness comprises: Perform high-precision synchronous sampling of voltage and current signals at both ends of multiple wiring harnesses in new energy vehicles to obtain raw data of multiple wiring harness signals; Performing piecewise overlapping fast Fourier transform and interpolation processing on the plurality of wire beam signal original data to obtain a plurality of wire beam signal harmonic component data; Performing wire harness anti-interference performance calculation and adaptive noise elimination on the plurality of wire harness signal harmonic component data to obtain a plurality of wire harness anti-interference performance indicators; Inputting the plurality of wire harness anti-interference performance indicators into a variable step-size regularized minimum mean square error model to perform filter parameter optimization calculation to obtain adaptive filter optimization parameters; Creating a multi-stage filter structure based on the adaptive filter optimization parameters, and performing eigenvalue decomposition orthogonal optimization to obtain multi-stage filter structure parameters; Based on the multi-stage filtering structure parameters, the electromagnetic compatibility of the whole vehicle of the new energy vehicle is distributedly optimized to obtain an optimized solution for anti-interference shielding of the whole vehicle wiring harness.

[0006] A second aspect of the present application provides a new energy wiring harness anti-interference shielding optimization device, the new energy wiring harness anti-interference shielding optimization device comprising: The sampling module is used to perform high-precision synchronous sampling of voltage and current signals at both ends of multiple wiring harnesses in new energy vehicles to obtain raw data of multiple wiring harness signals; A transformation module, used for performing piecewise overlapping fast Fourier transformation and interpolation processing on the plurality of wire beam signal original data to obtain a plurality of wire beam signal harmonic component data; A processing module, used for calculating the anti-interference performance of the wiring harness and performing adaptive noise elimination on the harmonic component data of the plurality of wiring harness signals to obtain a plurality of wiring harness anti-interference performance indicators; A calculation module, used for inputting the plurality of wire harness anti-interference performance indicators into a variable step-size regularized minimum mean square error model to perform filter parameter optimization calculation to obtain adaptive filter optimization parameters; A creation module is used to create a multi-stage filter structure based on the adaptive filter optimization parameters, and perform eigenvalue decomposition orthogonal optimization to obtain multi-stage filter structure parameters; The optimization module is used to perform distributed optimization on the electromagnetic compatibility of the whole vehicle of the new energy vehicle based on the multi-stage filtering structure parameters to obtain an anti-interference shielding optimization solution for the whole vehicle wiring harness.

[0007] The third aspect of the present application provides an electronic device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the electronic device executes the above-mentioned new energy wiring harness anti-interference shielding optimization method.

[0008] The fourth aspect of the present application provides a computer-readable storage medium, which stores instructions. When the computer-readable storage medium is run on a computer, it enables the computer to execute the above-mentioned new energy wiring harness anti-interference shielding optimization method.

[0009] Compared with the prior art, the present application has the following beneficial effects: through high-precision synchronous sampling and segmented overlapping fast Fourier transform, the present method can accurately capture the harmonic characteristics of the wiring harness signal, and adopts adaptive noise elimination technology to effectively filter out the influence of environmental interference on the anti-interference performance index, thereby improving the accuracy of performance evaluation. The variable step-size regularized minimum mean square error model is introduced to realize the adaptive optimization of the filter parameters and enhance the adaptability of the method to different wiring harness characteristics. Through multi-stage filter structure design and eigenvalue decomposition orthogonal optimization, the performance and stability of the filter are improved, and interference signals of various frequency bands can be more effectively suppressed. The distributed optimization and federated learning algorithms are adopted to greatly reduce the computational complexity of the electromagnetic compatibility optimization of the whole vehicle. The optimization strategy based on the importance score of the wiring harness ensures that the key wiring harness is optimized first, thereby improving the optimization efficiency and resource utilization. The generation of the anti-interference shielding optimization scheme of the whole vehicle wiring harness provides comprehensive guidance for the electromagnetic compatibility design of new energy vehicles, including filter parameters, shielding structure and wiring optimization suggestions. The present invention can dynamically adapt to complex electromagnetic environments, take into account the anti-interference shielding optimization of the wiring harness that takes into account local and global optimization, and improve the overall electromagnetic compatibility performance of new energy vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0011] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0012] Figure 1 It is a flow chart of a new energy wiring harness anti-interference shielding optimization method provided by an embodiment of the present invention; Figure 2 It is a schematic block diagram of the structure of the new energy wiring harness anti-interference shielding optimization device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0015] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0016] It should be further understood that the term "and / or" used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. Figure 1 , an embodiment of the new energy wiring harness anti-interference shielding optimization method in the embodiment of the present application includes: Step 100: Perform high-precision synchronous sampling on voltage and current signals at both ends of multiple wiring harnesses in a new energy vehicle to obtain multiple wiring harness signal raw data; It is understandable that the execution subject of the present application can be a new energy harness anti-interference shielding optimization device, or a terminal or a server, which is not limited here. The present application embodiment is described by taking the server as the execution subject as an example.

[0017] Specifically, multiple wiring harnesses in new energy vehicles are classified and marked, and the wiring harness type data is generated by identifying and marking the function, location and electrical characteristics of each wiring harness. According to the wiring harness type data, high-precision voltage sensors and current sensors are installed at the input and output ends of each wiring harness to collect the voltage and current signals of each wiring harness. The sampling frequency is set to 100 kHz, and the sampling time window is set to 1 second to capture subtle signal fluctuations. All sampling channels are synchronized through a synchronous control system, and all sensors perform sampling operations according to a unified timing. All sensors are started according to the synchronous sampling control signal, and the ambient temperature and humidity parameters are recorded at the same time to obtain the original sampling data and environmental parameter data. The original sampling data is quantized with 16-bit precision. The analog signal output by the sensor is converted into a digital signal and stored with a 16-bit resolution to ensure the refined representation of the signal. Data integrity checks are performed based on the quantized sampling data to ensure that the sampling data is not lost or damaged during transmission and storage. At the same time, abnormal data points are eliminated, and the erroneous data that deviates from the normal range is identified and deleted through a preset algorithm to obtain the cleaned sampling data. Combine the cleaned sampling data with the environmental parameter data, and associate the environmental factors with the sampling results. According to the harness type data, the combined data is grouped and stored to ensure that the signal data of each harness can be effectively classified and managed according to type and location, and the original data of multiple harness signals is obtained.

[0018] Step 200, performing piecewise overlapping fast Fourier transform and interpolation processing on multiple wire beam signal original data to obtain multiple wire beam signal harmonic component data; Specifically, the original data of multiple beam signals are processed to remove DC bias and linear trend, eliminate DC components and linear drift in the signal, effectively reduce the influence from the external environment and the sensor itself, and obtain preprocessed signal data. The preprocessed signal data is multiplied by the Hanning window function. The Hanning window is a windowing technology that can effectively reduce spectrum leakage. The windowed signal data is segmented, and the 1-second data of each beam is divided into 8 segments, each segment overlaps 50%, and each segment contains 12,500 sampling points. This segmentation method helps to capture the frequency change of the signal in a short time, and at the same time reduces the influence of the boundary effect on the analysis result through data overlap, so as to obtain a segmented data set. A 2048-point fast Fourier transform is performed on each data segment in the segmented data set. Fast Fourier transform is a common algorithm in signal processing, which is used to convert time domain signals into frequency domain signals. By performing spectrum analysis on each data segment, the segmented spectrum data of each segment is obtained. Based on the segmented spectrum data, the amplitude average and phase average of each frequency point are calculated to obtain the average spectrum data. The average spectrum data is interpolated. A frequency point is inserted every 0.1 Hz in the range of 0 to 50 kHz to obtain high-resolution spectrum data. Based on the high-resolution spectrum data, the Goertzel algorithm is used to extract the amplitude and phase information of the 50 Hz fundamental frequency and its 2nd to 100th harmonics. The Goertzel algorithm is an efficient algorithm for harmonic analysis at a specific frequency point. It can accurately extract the amplitude and phase information of the specified frequency and is suitable for harmonic analysis. The harmonic extraction results, including the amplitude and phase of each harmonic, are obtained through this algorithm. The harmonic extraction results are grouped according to the harness type data. The amplitude ratio of each harmonic relative to the fundamental wave is calculated to obtain the harmonic component data of multiple harness signals.

[0019] Step 300, performing wire harness anti-interference performance calculation and adaptive noise elimination on multiple wire harness signal harmonic component data to obtain multiple wire harness anti-interference performance indicators; It should be noted that the transmission loss of each harness is calculated based on the harmonic component data of multiple harness signals. By comparing the voltage amplitudes at the input and output ends of the harness at each harmonic frequency, taking the logarithm of the amplitude ratio between the two and multiplying it by 20, the transmission loss data is obtained, which reflects the energy loss of each harness at different harmonic frequencies. Based on the harmonic component data of the harness signal, the near-end crosstalk between the harnesses is calculated. The near-end crosstalk is calculated by comparing the voltage amplitudes at the input end of the interfering harness and the near end of the interfered harness, taking the logarithm of the ratio and multiplying it by 20 to obtain the near-end crosstalk data. Near-end crosstalk refers to the degree to which the electromagnetic interference signal generated near the input end of the harness affects the adjacent harness. This interference usually directly affects the signal transmission quality of the harness. Calculate far-end crosstalk. Far-end crosstalk is the transmission of the interfering signal through the harness, affecting the signal amplitude at the far end of the interfered harness. The calculation method is to take the logarithm of the ratio of the voltage amplitude at the input end of the interfering wire harness and the far end of the interfered wire harness, and multiply it by 20 to obtain the far-end crosstalk data. Far-end crosstalk usually affects the far-end signal receiving part of the wire harness, reflecting the degree of influence of the interference signal in long-distance transmission. The shielding effectiveness of each wire harness is calculated based on the harmonic component data of multiple wire harness signals. Shielding effectiveness is a measure of the protection effect of wire harness shielding materials in an electromagnetic interference environment. The specific calculation method is to take the logarithm of the ratio of the electromagnetic field strength before and after shielding and multiply it by 20 to obtain the shielding effectiveness data. This data can directly reflect the anti-interference ability of the shielding material at different frequencies and help determine the effectiveness of the wire harness shielding. The transmission loss data, near-end crosstalk data, far-end crosstalk data and shielding effectiveness data are combined to obtain an initial set of anti-interference performance indicators. Based on the initial set of anti-interference performance indicators, adaptive filtering is performed to remove noise and outliers, making the anti-interference performance indicators more stable and accurate. After adaptive filtering, the anti-interference performance indicators are grouped and standardized according to the wiring harness type data to ensure that different types of wiring harnesses can obtain reasonable anti-interference performance indicators, and finally multiple wiring harness anti-interference performance indicators are obtained.

[0020] Step 400, inputting multiple wire harness anti-interference performance indicators into a variable step-size regularized minimum mean square error model to perform filter parameter optimization calculation to obtain adaptive filter optimization parameters; Specifically, the anti-interference performance indicators of multiple harnesses are normalized, and each anti-interference performance indicator is mapped to the interval [0,1], so that data of different dimensions have a unified standardized form in subsequent calculations, and the normalized anti-interference performance indicator is obtained. A 64th-order finite impulse response (FIR) filter is constructed based on the normalized anti-interference performance indicator. The order of the filter determines the degree of detail of its frequency response, and the 64th-order filter can provide high flexibility and smoothness. The filter coefficients are initialized with uniformly distributed random numbers in the initial stage to ensure the diversity of the system. The normalized anti-interference performance indicator is filtered using the initial filter parameters, and the mean square error between the filter output and the expected output is calculated. The mean square error is an important criterion for evaluating the current performance of the filter, reflecting the gap between the filter output and the target value. In order to optimize the filter parameters, a regularized minimum mean square error objective function is constructed, and the L2 norm regularization term is added to prevent excessive growth of the filter coefficients and avoid overfitting problems. The regularization coefficient is set to 0.01, which moderately suppresses the increase in complexity while retaining the accuracy of the model. The gradient of the optimization objective function is calculated, and the filter coefficients are updated using the stochastic gradient descent method. The initial step size is set to 0.1 to ensure that the filter parameters change fast enough but not too large during each iteration. During the gradient descent process, the new mean square error is continuously calculated and compared with the error of the previous iteration. By analyzing the rate of change of the error, the step size is dynamically adjusted to obtain a variable step size sequence. Ensure rapid convergence in the initial stage, and when approaching the optimal solution, the step size is gradually reduced to avoid missing the optimal solution due to excessive adjustments. The entire iterative process continues until the set maximum number of iterations (such as 1000 times) is reached or the mean square error is less than the preset threshold (such as 10 -6 ). When the filter parameters converge, the performance is evaluated and the actual effect of the filter is measured by calculating the improvement of the signal-to-noise ratio before and after filtering. The filter parameters are combined with the normalized anti-interference performance index to form the final adaptive filter optimization parameters.

[0021] Step 500: Create a multi-stage filter structure based on the adaptive filter optimization parameters, and perform eigenvalue decomposition orthogonal optimization to obtain the multi-stage filter structure parameters; Specifically, the optimization parameters of the adaptive filter are decomposed. The 64-order FIR filter is split into a 3-stage cascade structure. The orders of these three filters are 32, 16 and 16 respectively, and the initial multi-stage filter parameters are obtained. Based on the initial multi-stage filter parameters, a transfer function model of the multi-stage filter structure is constructed. The transfer function of the filter is represented by Z transform. Each filter has its own transfer function, which represents the transfer function of the 32-order, 16-order and 16-order filters respectively. The total transfer function represents the response of the entire multi-stage filter system. Through this model, the frequency response of the entire system is calculated to obtain the frequency response data. The frequency response data is analyzed, and the passband fluctuation, stopband attenuation and transition band width indicators are calculated to obtain the initial filtering performance indicators. The passband fluctuation reflects the fluctuation degree of the signal within the allowed frequency range, the stopband attenuation represents the filter's ability to suppress interference signals within the blocking frequency range, and the transition band width measures the transition smoothness between the passband and the stopband. The basic performance of the filter is evaluated through the initial filtering performance indicators. Based on the initial filtering performance index, combined with the frequency response characteristics of the wiring harness and the maximum signal distortion allowed by the system, the weight coefficients are set and a multi-objective optimization function is constructed. The coefficient matrix of the multi-objective optimization function is decomposed by eigenvalue decomposition, and the matrix is ​​decomposed into an eigenvector matrix. These eigenvectors form an orthogonal basis vector set. The orthogonal basis vector set provides a new vector space for the filter parameters. Based on the orthogonal basis vector set, the initial multi-stage filter parameters are projected, and the parameter vectors of the filter are projected into the orthogonal space spanned by the eigenvectors to obtain the orthogonalized filter parameters. The orthogonalized parameters reduce the mutual interference between different parameters and ensure the optimization effect of the filter in different frequency bands. Based on the orthogonalized filter parameters, the conjugate gradient method is used for iterative optimization. The conjugate gradient method is an effective optimization algorithm that iteratively updates the parameters so that the objective function gradually approaches the optimal value. The entire iterative process continues until the maximum number of iterations of 500 is reached, or the change in the objective function value is less than 10 -8 When , the iteration stops. The optimized multi-stage filter structure parameters are obtained, which are used to improve the anti-interference shielding effect of the new energy wiring harness.

[0022] Step 600: Based on the multi-stage filtering structure parameters, the electromagnetic compatibility of the whole vehicle of the new energy vehicle is optimized in a distributed manner to obtain an anti-interference shielding optimization solution for the whole vehicle wiring harness.

[0023] Specifically, the new energy vehicle is divided into regions. The whole vehicle is divided into four regions: power system, control system, communication system and auxiliary system. The importance of the wiring harness in each region is evaluated. The weight coefficient of each wiring harness is calculated by using the hierarchical analysis method. The hierarchical analysis method calculates the relative importance of each wiring harness in the system by comparing the wiring harness with key indicators such as the electromagnetic compatibility and anti-interference capability of the system, and obtains the wiring harness importance score, which reflects the contribution of each wiring harness to the overall electromagnetic compatibility performance. The higher the score, the more important the wiring harness. The wiring harnesses in each region are sorted based on the wiring harness importance score. In order to optimize efficiency, the top three wiring harnesses with the highest scores in each region are selected as target wiring harnesses to form a target wiring harness set. The target wiring harness is the main source of electromagnetic interference and the focus of anti-interference shielding optimization. By focusing on these key wiring harnesses, the effect of vehicle electromagnetic compatibility optimization is improved. Multi-level filtering structure parameters are applied to the target wiring harness, and a regional optimization sub-model set for local electromagnetic compatibility optimization is constructed based on the characteristics of the filter. Each sub-model corresponds to a specific region and optimizes the target wiring harness in the region. To achieve distributed optimization, a local optimization unit is assigned to each sub-model, and a central control unit is set up to build a distributed optimization network. In the distributed optimization network, based on the strategy of federated learning, each local optimization unit processes the regional sub-models for which it is responsible in parallel and updates its local optimization parameters. This process can speed up the optimization while allowing each region to be optimized independently to minimize electromagnetic interference between different systems. As the parameters of the local optimization unit are updated, the local optimization results gradually converge. When the local optimization results are generated, the central control unit collects the model parameters of each local optimization unit. The local parameters are aggregated by weighted averaging to obtain the optimization parameters of the global model. The aggregation strategy ensures that the optimization results of each region can be reasonably integrated, while giving more weight to regions with higher importance. Based on the global optimization parameters, an anti-interference shielding scheme for the entire vehicle wiring harness is generated. The scheme includes the filter parameters of each target wiring harness, and also provides suggestions for the design of the shielding structure and the optimization of the wiring harness routing.

[0024] In the embodiment of the present application, through high-precision synchronous sampling and segmented overlapping fast Fourier transform, the method can accurately capture the harmonic characteristics of the wiring harness signal, and adopts adaptive noise elimination technology to effectively filter out the influence of environmental interference on the anti-interference performance index, thereby improving the accuracy of performance evaluation. The variable step-size regularized minimum mean square error model is introduced to realize the adaptive optimization of the filter parameters and enhance the adaptability of the method to different wiring harness characteristics. Through multi-stage filter structure design and eigenvalue decomposition orthogonal optimization, the performance and stability of the filter are improved, and interference signals of various frequency bands can be more effectively suppressed. The distributed optimization and federated learning algorithms are adopted to greatly reduce the computational complexity of the electromagnetic compatibility optimization of the whole vehicle. The optimization strategy based on the importance score of the wiring harness ensures that the key wiring harness is optimized first, and the optimization efficiency and resource utilization are improved. The generation of the anti-interference shielding optimization scheme of the whole vehicle wiring harness provides comprehensive guidance for the electromagnetic compatibility design of new energy vehicles, including filter parameters, shielding structure and wiring optimization suggestions. The present invention can dynamically adapt to complex electromagnetic environments, take into account the anti-interference shielding optimization of the wiring harness that is optimized locally and globally, and improve the overall electromagnetic compatibility performance of new energy vehicles.

[0025] In a specific embodiment, the process of executing step 100 may specifically include the following steps: Classify and mark multiple wiring harnesses in new energy vehicles to obtain wiring harness type data; According to the harness type data, high-precision voltage sensors and current sensors are installed at the input and output ends of each harness, and the sampling frequency is set to 100 kHz and the sampling time window is 1 second; All sampling channels are synchronized to obtain a synchronous sampling control signal, and all sensors are started according to the synchronous sampling control signal, and the ambient temperature and humidity parameters are recorded at the same time to obtain the original sampling data and the ambient parameter data; The original sampling data is quantized with 16-bit precision to obtain quantized sampling data, and data integrity check and abnormal data point elimination are performed based on the quantized sampling data to obtain cleaned sampling data; The cleaned sampling data and environmental parameter data are combined, and stored in groups according to the wiring harness type data to obtain a plurality of wiring harness signal raw data.

[0026] Specifically, multiple wiring harnesses in new energy vehicles are classified and marked. Each wiring harness may have different functions, such as power transmission, signal control, data communication, etc. These wiring harnesses are classified. By combining the purpose, location, connected devices and other information of the wiring harness, the wiring harness is divided into different types to obtain the corresponding wiring harness type data. According to the wiring harness type data, high-precision voltage sensors and current sensors are installed for the input and output ends of each wiring harness. In new energy vehicles, the wiring harness will carry different voltage and current signals. The use of high-precision sensors can ensure the accuracy of signal acquisition. In order to capture enough details, the sampling frequency is set to 100 kHz, which means that 100,000 data points are collected per second. The sampling time window is set to 1 second, so that the signal fluctuations within a time period can be fully captured in each sampling cycle. All sampling channels are synchronized. There may be mutual interference and coupling between the wiring harnesses, and the synchronization of sampling needs to be ensured. By synchronizing the sampling control signal, ensure that all sensors start collecting data at the same time. When the sensor is started, environmental parameters such as temperature and humidity are recorded synchronously to obtain the original sampling data and environmental parameter data. Changes in temperature and humidity can affect the electrical performance and anti-interference ability of the wiring harness. The original sampled data is quantized with 16-bit precision to convert the continuous analog signal into a discrete digital signal. 16-bit precision means that each signal value can be divided into 65536 discrete levels, this accuracy can ensure that signal details are fully preserved. For each collected voltage and current signal and , and its quantitative formula is: ; ; in, and are the minimum and maximum values ​​of the voltage signal, respectively. and are the minimum and maximum values ​​of the current signal, respectively. and is the quantized signal value, Indicates a round-down operation. After quantization is completed, a data integrity check is required based on the quantized data to ensure that no data is lost or transmission errors occur during the acquisition process. At the same time, abnormal data points are eliminated, which may usually be caused by sensor failure, transient interference or external noise. By using statistical methods, such as setting upper and lower thresholds or calculating the mean and standard deviation of the data, abnormal data points that exceed the reasonable range can be detected and eliminated to obtain the cleaned sampled data. The cleaned sampled data is combined with the environmental parameter data (temperature and humidity) so that the analysis results can better explain the performance of the wiring harness under different environmental conditions. The cleaned sampled data is grouped and stored according to the wiring harness type data to help subsequent analysis tools to process and optimize operations in a targeted manner according to the types of different wiring harnesses.

[0027] In a specific embodiment, the process of executing step 200 may specifically include the following steps: Performing DC bias removal and linear trend removal processing on the original data of multiple harness signals to obtain pre-processed signal data; The preprocessed signal data is multiplied by the Hanning window function to obtain the windowed signal data, and the windowed signal data is segmented to divide the 1-second data of each harness into 8 data segments with 50% overlap, each segment contains 12,500 sampling points, and a segmented data set is obtained; Performing a 2048-point fast Fourier transform on each data segment in the segmented data set to obtain segmented spectrum data, and calculating the amplitude average and phase average of each frequency point based on the segmented spectrum data to obtain average spectrum data; Interpolate the average spectrum data, insert a frequency point every 0.1 Hz in the range of 0-50 kHz, and obtain high-resolution spectrum data; Based on high-resolution spectrum data, the amplitude and phase information of the 50 Hz fundamental frequency and its 2-100th harmonics are extracted based on the Goertzel algorithm to obtain the harmonic extraction results; The harmonic extraction results are grouped according to the wire harness type data, and the amplitude ratio of each harmonic to the fundamental wave is calculated to obtain a plurality of wire harness signal harmonic component data.

[0028] Specifically, the raw data of multiple wire harness signals are processed to remove DC bias and linear trend. The DC bias and linear trend in the wire harness signal are usually introduced by sensor errors or other external factors. The purpose of removing DC bias is to adjust the average value of the signal to zero, while removing linear trend is to remove low-frequency linear changes in the signal, making the subsequent spectrum analysis more accurate. Removing DC bias can be achieved by simply subtracting the average value of the signal, the formula is: ; in, is the original signal, is the signal after removing the DC bias, is the number of sampling points of the signal. De-linear trending can be achieved by linear regression fitting and subtracting the fitted trend line. Multiply the preprocessed signal data by the Hanning window function. The Hanning window function is a weighted function that can smooth the boundaries of the signal and reduce spectrum leakage. The expression of the Hanning window function is: ; in, It is The window function value of sampling points, is the total number of sampling points. By multiplying the preprocessed signal by the window function point by point, the windowed signal data is obtained. Windowing can reduce the spectrum leakage effect, making the spectrum analysis result more accurate. The windowed signal data is segmented. Assuming that the signal sampling time of each harness is 1 second and the sampling frequency is 100 kHz, the data per second contains 100,000 sampling points. In order to improve the stability of spectrum analysis, the 1-second signal is divided into 8 data segments, each with 50% overlap. Each segment of data contains 12,500 sampling points, ensuring that important information is not lost when the signal frequency fluctuates. The overlapping segmentation method reduces the boundary effect while ensuring the continuity of the signal. Perform a 2048-point fast Fourier transform (FFT) on each segmented data segment to convert the time domain signal to the frequency domain. The calculation formula of FFT is: ; in, It is The complex spectrum value of frequency points, is the input signal Sample values, is the number of Fourier transform points (here 2048), is an imaginary unit. Through FFT transformation, the spectrum data of each data segment can be obtained, including the amplitude and phase information of each frequency point. After obtaining the spectrum data of each data segment, the amplitude average and phase average of each frequency point are calculated based on the segmented spectrum data. The amplitude average is completed by calculating the average value of each frequency point in all data segments, which can reduce the influence of noise and random interference and obtain a smoother spectrum curve. Phase average is similar. By calculating the average value of the phase of each frequency point, the overall phase characteristics of the signal are obtained. In order to improve the resolution of spectrum analysis, the average spectrum data is interpolated. Assuming that the frequency range is 0-50 kHz, the goal of interpolation is to insert a frequency point every 0.1Hz to generate high-resolution spectrum data. Interpolation can use methods such as linear interpolation or spline interpolation. Through these methods, more frequency points are inserted on the basis of the existing spectrum data to obtain a more refined frequency distribution. Based on the high-resolution spectrum data, the Goertzel algorithm is used to extract the amplitude and phase information of the 50 Hz fundamental frequency and its 2-100 harmonics. The Goertzel algorithm is an efficient calculation algorithm for specific frequency points, suitable for extracting spectrum information of specific harmonic components. The core formula of the Goertzel algorithm is: ; in, It is The output of the iteration, is the target frequency, is the input signal. Through this algorithm, the amplitude and phase information of the 50 Hz fundamental frequency and its harmonic components are efficiently extracted. The harmonic extraction results are grouped according to the harness type data. Different harness types may carry different signal characteristics. Classifying the harmonic results helps to analyze the electromagnetic interference characteristics of various harnesses more targetedly. Based on the harmonic data, the amplitude ratio of each harmonic relative to the fundamental wave is calculated. The formula is: ; in, is the ratio of the harmonic amplitude to the fundamental amplitude, is the amplitude of the harmonic, is the amplitude of the fundamental wave. This ratio can help analyze the strength of different harmonics in the signal, thereby evaluating the anti-interference performance of the wiring harness in various frequency ranges.

[0029] In a specific embodiment, the process of executing step 300 may specifically include the following steps: Calculating the transmission loss of each harness based on the harmonic component data of the harness signals, specifically taking the logarithm of the input-end and output-end voltage amplitude ratio of each harmonic frequency and multiplying it by 20 to obtain the transmission loss data; Based on the harmonic information of adjacent wire bundles in the harmonic component data of multiple wire bundle signals, the near-end crosstalk is calculated, specifically, the logarithm of the voltage amplitude ratio between the input end of the interfering wire bundle and the near-end of the interfered wire bundle is taken and multiplied by 20 to obtain the near-end crosstalk data; Far-end crosstalk is calculated based on the harmonic component data of multiple wire harness signals, specifically, the logarithm of the voltage amplitude ratio between the input end of the interfering wire harness and the far end of the interfered wire harness is taken and multiplied by 20 to obtain the far-end crosstalk data; The shielding effectiveness of each harness is calculated based on the harmonic component data of multiple harness signals, specifically, the logarithm of the electromagnetic field intensity ratio before and after shielding is taken and multiplied by 20 to obtain the shielding effectiveness data; Combining transmission loss data, near-end crosstalk data, far-end crosstalk data and shielding effectiveness data to obtain an initial anti-interference performance indicator set; Adaptive filtering is performed based on the initial anti-interference performance index set, and the data is grouped and standardized according to the type of wiring harness, so as to obtain multiple wiring harness anti-interference performance indicators.

[0030] Specifically, the transmission loss of each harness is calculated based on the harmonic component data of multiple harnesses. The transmission loss is defined as the attenuation of the voltage amplitude at the input and output of the harness at each harmonic frequency. The ratio of the input voltage amplitude to the output voltage amplitude is taken logarithmically and multiplied by 20 to obtain the transmission loss. The formula is: ; in, Indicates the frequency The transmission loss at and are the voltage amplitudes at the input and output ends respectively. The formula is used to evaluate the degree of energy loss of each harmonic frequency during transmission. The greater the transmission loss, the more severe the attenuation of the frequency signal when passing through the harness. Based on the harmonic component data of multiple harnesses, the near-end crosstalk between adjacent harnesses is calculated. Near-end crosstalk refers to the electromagnetic signal at the input end of the interfering harness affecting the near-end voltage of the adjacent interfered harness through the coupling path. The method for calculating near-end crosstalk is similar to the transmission loss. The input voltage of the interfering harness and the voltage amplitude at the near end of the interfered harness are obtained, and then the ratio of the two is calculated, the logarithm is taken and multiplied by 20. The formula is: ; in, is at frequency Near-end crosstalk at Indicates the voltage amplitude at the input end of the interference harness, Indicates the voltage amplitude at the near end of the interfered wire bundle. The near-end crosstalk value reflects the coupling effect of the two wire bundles near the input end. The greater the crosstalk, the more serious the signal interference. Calculating far-end crosstalk means that the electromagnetic signal at the input end of the interfering wire bundle affects the voltage at the far end of the adjacent wire bundle through the transmission and coupling of the wire bundle. The calculation method of far-end crosstalk is the same as that of near-end crosstalk, but the voltage amplitude of the interfered wire bundle is taken from its far end. The formula for far-end crosstalk is: ; in, Indicates the frequency Far-end crosstalk at is the voltage amplitude at the far end of the interfered wire harness. Far-end crosstalk reflects the ability of the interference signal to affect the adjacent wire harness during transmission, especially in long-distance transmission, the value of far-end crosstalk can significantly affect the signal quality. Based on the harmonic component data of the wire harness signal, the shielding effectiveness of each wire harness is calculated. Shielding effectiveness is used to measure the ability of shielding materials to suppress electromagnetic fields. The specific calculation method is to take the logarithm of the ratio of the electromagnetic field strength before and after shielding and multiply it by 20. The formula is: ; in, is at frequency Shielding effectiveness at and The electromagnetic field strength before and after shielding, respectively. The higher the shielding effectiveness, the stronger the shielding material's ability to suppress external electromagnetic interference, which can effectively improve the anti-interference performance of the wiring harness. The data of transmission loss, near-end crosstalk, far-end crosstalk and shielding effectiveness are combined to form an initial set of anti-interference performance indicators. Based on the initial set of anti-interference performance indicators, adaptive filtering is performed. Adaptive filtering can adjust the filter parameters according to the real-time signal characteristics, which can eliminate random noise in the signal and improve the accuracy of the signal. When applying adaptive filtering, indicators such as transmission loss, crosstalk and shielding effectiveness are standardized so that they can be compared and optimized under a unified scale. The standardized data is grouped according to different wiring harness types. For example, the wiring harness of the power system and the wiring harness of the communication system may have different anti-interference requirements, so they need to be processed independently in different groups. Through group processing, the characteristics of each type of wiring harness are optimized to improve the overall anti-interference performance.

[0031] In a specific embodiment, the process of executing step 400 may specifically include the following steps: Normalizing multiple wire harness anti-interference performance indicators, mapping each wire harness anti-interference performance indicator to the interval [0,1], and obtaining a normalized anti-interference performance indicator; A 64-order FIR filter structure is constructed based on the normalized anti-interference performance index, and the filter coefficients are initialized using uniformly distributed random numbers to obtain the initial filter parameters; Perform filtering on the normalized anti-interference performance index based on the initial filter parameters, calculate the mean square error between the filter output and the expected output, and obtain the initial error value; The regularized minimum mean square error objective function is constructed based on the initial error value, the L2 norm regularization term is added, and the regularization coefficient is set to 0.01 to obtain the optimized objective function; The gradient of the optimization objective function is calculated, and the filter coefficients are updated using the stochastic gradient descent method. The initial step size is set to 0.1 to obtain the updated filter parameters. The new mean square error is calculated based on the updated filter parameters, compared with the error of the previous iteration, and the step size is dynamically adjusted according to the error change rate to obtain a variable step size sequence; Based on the variable step size sequence and the updated filter parameters, the stochastic gradient descent iteration is continued until the maximum number of iterations reaches 1000 or the mean square error is less than the preset threshold 10 -6 , get the converged filter parameters; The performance of the converged filter parameters is evaluated, the signal-to-noise ratio improvement before and after filtering is calculated, and the filter parameters and the corresponding normalized anti-interference performance indicators are combined to obtain the adaptive filter optimization parameters.

[0032] Specifically, multiple wire harness anti-interference performance indicators are normalized, and each wire harness anti-interference performance indicator is mapped to the interval [0,1] to obtain a normalized anti-interference performance indicator. Based on the normalized anti-interference performance indicator, a 64th-order finite impulse response (FIR) filter structure is constructed. The FIR filter is a linear time-invariant system with good frequency selectivity. In order to initialize the coefficients of the filter, uniformly distributed random numbers are used to generate the initial filter parameters, and the uniform distribution range is set to [-1,1] to ensure that the coefficients of the initial filter are diverse. The output signal of the filter is obtained by convolution with the input signal. The output formula of the FIR filter is: ; in, It is The filter output at time t, It is Order filter coefficients, is the normalized input signal of the anti-interference performance index, is the order of the filter (here 64). After obtaining the initial filter parameters, use these parameters to filter the normalized anti-interference performance index. The difference between the filtered output and the expected output (target anti-interference performance) can be measured by the mean square error (MSE), which is calculated as: ; in, is the total number of input data points, is the output of the filter, is the expected output. The initial mean square error value reflects the performance of the filter under the current parameters. The smaller the error, the closer the output of the filter is to the expected value. In order to optimize the performance of the filter, a regularized minimum mean square error objective function is constructed. In the original MSE objective function, the L2 norm regularization term is added to avoid the filter coefficient being too large, which leads to overfitting problems. The regularized objective function is: ; in, is the regularized objective function, is the regularization coefficient, set to 0.01. L2 regularization term is the sum of the squares of all filter coefficients, which is used to constrain the size of the coefficients to prevent overfitting. The filter coefficients are updated by using the stochastic gradient descent method by calculating the gradient of the optimization objective function. The stochastic gradient descent method updates the coefficients along the gradient direction so that the objective function is gradually reduced. The update formula of gradient descent is:

[0033] ; in, is the learning rate, and the initial step size is set to 0.1, / is the objective function with respect to the filter coefficients The gradient of . Through this formula, the coefficients of the filter are fine-tuned in each iteration, gradually approaching the optimal solution. After each iteration, a new mean square error is calculated based on the updated filter parameters and compared with the error of the previous iteration. According to the rate of change of the error, the step size is dynamically adjusted to obtain a more efficient optimization process. If the error decreases faster, the step size will remain larger; if the error changes less, the step size will gradually decrease to ensure smooth convergence. The adjustment of the step size allows the optimization process to converge quickly while avoiding oscillation or missing the optimal point. Based on the dynamically adjusted step size sequence and the updated filter parameters, continue to perform stochastic gradient descent iterations until the maximum number of iterations is 1000, or the mean square error is less than the preset threshold . When one of the conditions is met, the optimization process ends and the final converged filter parameters are obtained. In order to evaluate the performance of the converged filter, the improvement of the signal-to-noise ratio before and after filtering is calculated. The improvement of the signal-to-noise ratio can reflect the improvement of the filter's anti-interference performance. The formula for the signal-to-noise ratio is: ; By comparing the signal-to-noise ratio before and after filtering, the noise suppression effect of the filter and its fidelity to the signal are measured. The optimized filter parameters are combined with the normalized anti-interference performance index to obtain the final adaptive filter optimization parameters.

[0034] In a specific embodiment, the process of executing step 500 may specifically include the following steps: The optimization parameters of the adaptive filter are decomposed, and the 64-order FIR filter is split into a 3-stage cascade structure, with the order of each stage of the filter being 32, 16, and 16 respectively, to obtain the initial multi-stage filter parameters; The transfer function model of the multi-stage filter structure is constructed based on the initial multi-stage filter parameters, and the transfer function of each filter stage is expressed by Z transform to obtain the total transfer function of the multi-stage filter system. The frequency response of the multi-stage filtering system is calculated based on the total transfer function to obtain the frequency response data, and the frequency response data is analyzed to calculate the passband fluctuation, stopband attenuation and transition band width indicators to obtain the initial filtering performance indicators; Based on the initial filtering performance index, combined with the frequency response characteristics of the harness and the maximum allowable signal distortion, the weight coefficient is set to construct a multi-objective optimization function; Perform eigenvalue decomposition on the coefficient matrix of the multi-objective optimization function, calculate the eigenvector matrix, and obtain an orthogonal basis vector set; Based on the orthogonal basis vector set, the initial multi-stage filter parameters are projected and transformed, and the parameter vector is projected into the orthogonal space spanned by the eigenvector to obtain the orthogonalized filter parameters; Based on the orthogonalized filter parameters, the conjugate gradient method is used for iterative optimization until the maximum number of iterations reaches 500 or the objective function value changes by less than 10 -8 , and obtain the multi-stage filtering structure parameters.

[0035] Specifically, the optimization parameters of the adaptive filter are decomposed, and the 64-order FIR filter is split into a three-stage cascade structure, with the order of each stage of the filter being 32, 16, and 16, respectively. The purpose of grading is to decompose the complex 64-order filter into multiple smaller filters, thereby simplifying the design and improving the response flexibility of the system. The design of each stage of the filter can better adapt to the needs of different frequency bands. For example, the first stage 32-order filter can be used to process the main passband, while the next two stages of 16-order filters can be used to fine-tune the characteristics of the stopband or transition band. A transfer function model of the multi-stage filtering structure is constructed based on the initial multi-stage filter parameters. The transfer function of each stage of the filter can be represented by the Z transform. The Z transform is a standard tool in digital filter analysis that transforms discrete signals in the time domain to the frequency domain. For the first stage 32-order filter, its transfer function can be expressed as: ; in, is the transfer function of the first-stage filter, is the coefficient of the first-order filter. Similarly, the transfer functions of the second-order and third-order 16-order filters are: ; in, and They represent the transfer functions of the second and third stage filters respectively. The total transfer function of the entire multi-stage filter system is It can be expressed by the series connection (cascade connection) of these three filters as follows: ; The total transfer function reflects the frequency response characteristics of the entire multi-stage filter system. Based on the total transfer function, the frequency response of the entire filter system is calculated. The frequency response reflects the amplification or attenuation effect of the system on input signals of different frequencies, which can be specifically determined by In the frequency domain, the frequency The corresponding amplitude and phase response are used to obtain the frequency response data. Through the calculated frequency response data, the passband fluctuation, stopband attenuation and transition band width of the filter are analyzed. The passband fluctuation reflects the response change of the filter to the signal within the allowable frequency range. Ideally, the amplitude of the passband should be flat. Stopband attenuation is the suppression ability of the filter within the blocking frequency range. The greater the attenuation, the better. The transition band width reflects the width of the transition area between the passband and the stopband. By analyzing these performance indicators, the initial filtering performance indicators are obtained. Combined with the frequency response characteristics of the wiring harness and the maximum allowable signal distortion, the weight coefficient is set to construct a multi-objective optimization function. This optimization function is used to balance different performance indicators (such as passband fluctuation, stopband attenuation, etc.) to ensure that the comprehensive performance of the filter is optimized while meeting the system requirements. The multi-objective optimization function can be expressed as: Passband ripple Stopband Attenuation Transition zone width; in, , and is the weight coefficient of the corresponding performance index, is the filter parameter vector. The choice of weight coefficients depends on the specific requirements of the system. For example, some applications may have higher requirements for stopband attenuation, while other applications may pay more attention to the flatness of the passband. In order to optimize the multi-objective function, the coefficient matrix of the optimization function is decomposed by eigenvalue. Through eigenvalue decomposition, the original filter parameters are projected into the orthogonal space spanned by the eigenvectors, eliminating the interdependence between the filter parameters and simplifying the optimization process. The formula for eigenvalue decomposition is: ; in, is the coefficient matrix of the optimization function, is the eigenvector matrix, is a diagonal eigenvalue matrix. The eigenvector matrix An orthogonal basis vector set is provided for the filter parameters. Based on the orthogonal basis vector set, the initial multi-stage filter parameters are projected and the filter parameter vector is projected into the orthogonal space to obtain the orthogonalized filter parameters. This transformation can simplify the optimization process, reduce the coupling between different filter parameters, and improve the optimization efficiency. Based on the orthogonalized filter parameters, the conjugate gradient method is used for iterative optimization. The conjugate gradient method is an efficient iterative optimization algorithm suitable for dealing with large-scale optimization problems. By searching along the conjugate direction each time, it is possible to converge to the optimal solution more quickly. The iterative optimization process continues until the maximum number of iterations of 500 is reached, or the change in the objective function value is less than the preset threshold of 10 -8, to ensure the stability and accuracy of the optimization process.

[0036] In a specific embodiment, the process of executing step 600 may specifically include the following steps: New energy vehicles are divided into four areas: power system, control system, communication system and auxiliary system, and the vehicle area division plan is obtained; Based on the vehicle area division scheme, the importance of the wiring harness in each area is evaluated, and the weight coefficient of each wiring harness is calculated using the hierarchical analysis method to obtain the wiring harness importance score; The wire bundles in each area are sorted based on the wire bundle importance score, and the top three wire bundles with the highest scores in each area are selected as target wire bundles to obtain a target wire bundle set; Apply multi-level filtering structure parameters to each wire harness in the target wire harness set to construct a regional optimization sub-model set for local electromagnetic compatibility optimization; Distributed learning is performed based on the regional optimization sub-model set, a local optimization unit is assigned to each sub-model, and a central control unit is set up to obtain a distributed optimization network; Federated learning is performed based on a distributed optimization network. Each local optimization unit processes its own sub-model in parallel, updates local parameters, and obtains local optimization results. Aggregate the local optimization results, the central control unit collects the model parameters of each local optimization unit, and uses the weighted average method to update the global model to obtain the global optimization parameters; The anti-interference shielding scheme for the entire vehicle wiring harness is generated based on the global optimization parameters. The anti-interference shielding scheme for the entire vehicle wiring harness includes the filter parameters, shielding structure design and wiring optimization suggestions for each target wiring harness.

[0037] Specifically, the whole vehicle is divided into different areas. The whole vehicle system can be divided into four main areas: power system, control system, communication system and auxiliary system. The power system is responsible for transmitting and distributing electrical energy, the control system is responsible for accurately controlling the various functions of the vehicle, the communication system is used to process the data signal transmission inside and outside the vehicle, and the auxiliary system includes auxiliary equipment such as air conditioning and lighting. Through the regional division scheme, the wiring harnesses in each area can be organized and managed more clearly to ensure that the anti-interference design of each system meets its functional requirements. The importance of the wiring harnesses in each area is evaluated. The weight coefficient of each wiring harness is calculated by using the hierarchical analysis method to systematically evaluate the impact of each wiring harness on electromagnetic compatibility. The hierarchical analysis method calculates the weight of each wiring harness based on multiple factors such as the function of the wiring harness, the importance of the area where it is located, and the sensitivity to signal quality by constructing a series of pairwise comparison matrices. The weight calculation formula of the hierarchical analysis method is: ; in, Indicates The weight of the root bundle, It is a wiring harness and wiring harness Pairwise comparison matrix values ​​between It is a wiring harness The weight of . Through this calculation process, the importance score of each harness is obtained. The higher the score, the greater the importance of the harness in the region, and more attention should be paid to it in the subsequent optimization process. According to the importance score, the harnesses in each region are sorted, and the top three harnesses with the highest scores in each region are screened out. These harnesses have a significant impact on the overall electromagnetic compatibility of the vehicle, so they are included in the optimization design as target harnesses. Through this screening, a target harness set is obtained. Multi-stage filtering structure parameters are applied to each harness in the target harness set. Multi-stage filters can accurately suppress noise in different frequency bands and improve the anti-interference performance of the harness. A local electromagnetic compatibility optimization model is constructed for each harness to obtain a regional optimization sub-model. These sub-models can optimize the anti-interference performance of each harness separately to ensure that the electromagnetic interference control in each region achieves the best effect. In order to improve the optimization efficiency, a distributed learning framework is constructed based on these regional optimization sub-model sets. In this framework, each sub-model is assigned a local optimization unit, which can process their respective regional optimization tasks in parallel. A central control unit is set up to coordinate the work of all local optimization units to form a distributed optimization network. Federated learning is performed on the basis of the distributed optimization network. Federated learning is a distributed machine learning method that allows multiple nodes to perform model training without sharing data. Each local optimization unit is responsible for updating the local parameters of its sub-model and performing parallel anti-interference optimization based on the harness data and characteristics in its respective area. Through distributed learning, the models in each area can independently improve the local anti-interference performance while maintaining consistency with the overall electromagnetic compatibility goals. When the local optimization unit completes the update of the local model, the central control unit will collect the model parameters of all local optimization units. In order to generate global optimization parameters, these local parameters are aggregated using the weighted average method. The formula for weighted average is: ; in, is the optimization parameter of the global model, It is The optimization parameters of the local unit, is the weight of the local unit, is the total number of local units. The weighted average method ensures that the contribution ratio of the local optimization results of each area in the global model matches its actual importance. After aggregation, the global optimization parameters obtained can improve the electromagnetic compatibility performance throughout the vehicle. Based on the global optimization parameters, the anti-interference shielding scheme for the entire vehicle wiring harness is finally generated. The scheme includes the filter parameters of each target wiring harness, as well as the shielding structure design and wiring optimization suggestions. For example, in the power system, due to the large current, the shielding design may require the use of multiple shielding layers and take into account the coupling effect of high-frequency signals, while in the communication system, the focus is on crosstalk suppression and wiring isolation design between wiring harnesses. For wiring optimization, it is recommended to minimize the crossing of power lines and communication harnesses, and ensure that the harnesses of different systems are isolated from each other during wiring to reduce the propagation path of electromagnetic interference.

[0038] The above describes the anti-interference shielding optimization method for the new energy wiring harness in the embodiment of the present application. The following describes the anti-interference shielding optimization device 10 for the new energy wiring harness in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the new energy harness anti-interference shielding optimization device 10 includes: The sampling module 11 is used to perform high-precision synchronous sampling of voltage and current signals at both ends of multiple wiring harnesses in the new energy vehicle to obtain raw data of multiple wiring harness signals; A transformation module 12, used for performing segmented overlapping fast Fourier transformation and interpolation processing on the original data of multiple line beam signals to obtain multiple line beam signal harmonic component data; A processing module 13 is used to calculate the anti-interference performance of the wiring harness and perform adaptive noise elimination on the harmonic component data of the plurality of wiring harness signals to obtain a plurality of wiring harness anti-interference performance indicators; A calculation module 14 is used to input multiple wire harness anti-interference performance indicators into a variable step-size regularized minimum mean square error model to perform filter parameter optimization calculation to obtain adaptive filter optimization parameters; A creation module 15 is used to create a multi-stage filter structure based on the adaptive filter optimization parameters, and perform eigenvalue decomposition orthogonal optimization to obtain the multi-stage filter structure parameters; The optimization module 16 is used to perform distributed optimization on the electromagnetic compatibility of the whole vehicle of the new energy vehicle based on the multi-stage filtering structure parameters, and obtain the anti-interference shielding optimization plan of the whole vehicle wiring harness.

[0039] Through the cooperation of the above components, high-precision synchronous sampling and segmented overlapping fast Fourier transform, this method can accurately capture the harmonic characteristics of the wiring harness signal, and adopt adaptive noise elimination technology to effectively filter out the influence of environmental interference on anti-interference performance indicators, thereby improving the accuracy of performance evaluation. The variable step-size regularized minimum mean square error model is introduced to realize the adaptive optimization of filter parameters and enhance the adaptability of the method to different wiring harness characteristics. Through multi-stage filter structure design and eigenvalue decomposition orthogonal optimization, the performance and stability of the filter are improved, and interference signals of various frequency bands can be more effectively suppressed. The distributed optimization and federated learning algorithms are adopted to greatly reduce the computational complexity of vehicle electromagnetic compatibility optimization. The optimization strategy based on the importance score of the wiring harness ensures that the key wiring harness is optimized first, thereby improving the optimization efficiency and resource utilization. The generation of the anti-interference shielding optimization scheme for the wiring harness of the whole vehicle provides comprehensive guidance for the electromagnetic compatibility design of new energy vehicles, including filter parameters, shielding structure and wiring optimization suggestions. The present invention can dynamically adapt to complex electromagnetic environments, take into account the anti-interference shielding optimization of wiring harnesses that are optimized locally and globally, and improve the overall electromagnetic compatibility performance of new energy vehicles.

[0040] The present application also provides an electronic device, which includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the new energy wiring harness anti-interference shielding optimization method in the above-mentioned embodiments.

[0041] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the new energy wiring harness anti-interference shielding optimization method.

[0042] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0043] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable an electronic device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0044] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A new energy wiring harness anti-interference shielding optimization method, characterized in that: The method comprises: Perform high-precision synchronous sampling of voltage and current signals at both ends of multiple wiring harnesses in new energy vehicles to obtain raw data of multiple wiring harness signals; Performing piecewise overlapping fast Fourier transform and interpolation processing on the plurality of wire beam signal original data to obtain a plurality of wire beam signal harmonic component data; Performing wire harness anti-interference performance calculation and adaptive noise elimination on the plurality of wire harness signal harmonic component data to obtain a plurality of wire harness anti-interference performance indicators; Inputting the plurality of wire harness anti-interference performance indicators into a variable step-size regularized minimum mean square error model to perform filter parameter optimization calculation to obtain adaptive filter optimization parameters; Creating a multi-stage filter structure based on the adaptive filter optimization parameters, and performing eigenvalue decomposition orthogonal optimization to obtain multi-stage filter structure parameters; Based on the multi-stage filtering structure parameters, the electromagnetic compatibility of the whole vehicle of the new energy vehicle is distributedly optimized to obtain an optimized solution for anti-interference shielding of the whole vehicle wiring harness.

2. The new energy wiring harness anti-interference shielding optimization method according to claim 1 is characterized in that: The voltage and current signals at both ends of multiple wiring harnesses in the new energy vehicle are sampled synchronously with high precision to obtain multiple wiring harness signal raw data, including: Classify and mark multiple wiring harnesses in new energy vehicles to obtain wiring harness type data; According to the harness type data, a high-precision voltage sensor and a current sensor are respectively installed at the input end and the output end of each harness, and the sampling frequency is set to 100 kHz and the sampling time window is set to 1 second; Synchronize all sampling channels to obtain a synchronous sampling control signal, and start all sensors according to the synchronous sampling control signal, and simultaneously record the ambient temperature and humidity parameters to obtain original sampling data and ambient parameter data; Performing 16-bit precision quantization processing on the original sampled data to obtain quantized sampled data, and performing data integrity check and abnormal data point elimination based on the quantized sampled data to obtain cleaned sampled data; The cleaned sampling data and the environmental parameter data are combined, and stored in groups according to the wire harness type data to obtain a plurality of wire harness signal original data.

3. The new energy wiring harness anti-interference shielding optimization method according to claim 2 is characterized in that: The step of performing segmented overlapping fast Fourier transform and interpolation processing on the plurality of line beam signal original data to obtain a plurality of line beam signal harmonic component data includes: Performing DC bias removal and linear trend removal processing on the plurality of wire harness signal raw data to obtain pre-processed signal data; Multiplying the preprocessed signal data by a Hanning window function to obtain windowed signal data, and performing segmentation processing on the windowed signal data to divide 1 second of data of each harness into 8 data segments with 50% overlap, each segment containing 12,500 sampling points, to obtain a segmented data set; Performing a 2048-point fast Fourier transform on each data segment in the segmented data set to obtain segmented spectrum data, and calculating the amplitude average and phase average of each frequency point based on the segmented spectrum data to obtain average spectrum data; Performing interpolation processing on the average spectrum data, inserting a frequency point every 0.1 Hz in the range of 0-50 kHz to obtain high-resolution spectrum data; Based on the high-resolution spectrum data, the amplitude and phase information of the 50 Hz fundamental frequency and its 2-100th harmonics are extracted based on the Goertzel algorithm to obtain a harmonic extraction result; The harmonic extraction results are grouped according to the wire harness type data, and the amplitude ratio of each harmonic to the fundamental wave is calculated to obtain a plurality of wire harness signal harmonic component data.

4. The new energy wiring harness anti-interference shielding optimization method according to claim 1 is characterized in that: The wire harness anti-interference performance calculation and adaptive noise elimination are performed on the plurality of wire harness signal harmonic component data to obtain a plurality of wire harness anti-interference performance indicators, including: Calculating the transmission loss of each harness based on the harmonic component data of the plurality of harness signals, specifically taking the logarithm of the input-end and output-end voltage amplitude ratio of each harmonic frequency and multiplying it by 20 to obtain the transmission loss data; Calculate the near-end crosstalk based on the harmonic information of the adjacent wire bundles in the harmonic component data of the multiple wire bundle signals, specifically by taking the logarithm of the voltage amplitude ratio between the input end of the interfering wire bundle and the near-end of the interfered wire bundle and multiplying it by 20 to obtain the near-end crosstalk data; Calculate the far-end crosstalk based on the harmonic component data of the multiple wire harness signals, specifically by taking the logarithm of the voltage amplitude ratio between the input end of the interfering wire harness and the far-end of the interfered wire harness and multiplying it by 20 to obtain the far-end crosstalk data; Calculating the shielding effectiveness of each wire harness based on the harmonic component data of the plurality of wire harness signals, specifically taking the logarithm of the ratio of the electromagnetic field strength before and after shielding and multiplying it by 20 to obtain the shielding effectiveness data; Combining the transmission loss data, the near-end crosstalk data, the far-end crosstalk data and the shielding effectiveness data to obtain an initial anti-interference performance indicator set; Adaptive filtering is performed based on the initial anti-interference performance indicator set, and grouping and standardization processing are performed according to the wire harness type data to obtain multiple wire harness anti-interference performance indicators.

5. The new energy wiring harness anti-interference shielding optimization method according to claim 4 is characterized in that: The step of inputting the plurality of wire harness anti-interference performance indicators into a variable step-size regularized minimum mean square error model to perform filter parameter optimization calculation to obtain adaptive filter optimization parameters includes: Normalizing the plurality of wire harness anti-interference performance indicators, mapping each wire harness anti-interference performance indicator to the interval [0, 1], and obtaining a normalized anti-interference performance indicator; Based on the normalized anti-interference performance index, a 64-order FIR filter structure is constructed, and filter coefficients are initialized using uniformly distributed random numbers to obtain initial filter parameters; Filtering the normalized anti-interference performance index based on the initial filter parameters, calculating the mean square error between the filter output and the expected output, and obtaining an initial error value; A regularized minimum mean square error objective function is constructed based on the initial error value, an L2 norm regularization term is added, and the regularization coefficient is set to 0.01 to obtain an optimized objective function; Performing gradient calculation on the optimization objective function, updating the filter coefficients by using the stochastic gradient descent method, with the initial step size set to 0.1, to obtain updated filter parameters; Calculating a new mean square error based on the updated filter parameters, comparing it with the error of the previous iteration, and dynamically adjusting the step size according to the error change rate to obtain a variable step size sequence; Based on the variable step size sequence and the updated filter parameters, the random gradient descent iteration is continued until the maximum number of iterations reaches 1000 or the mean square error is less than the preset threshold 10 -6 , get the converged filter parameters; The performance of the converged filter parameters is evaluated, the signal-to-noise ratio improvement before and after filtering is calculated, and the filter parameters are combined with the corresponding normalized anti-interference performance indicators to obtain the adaptive filter optimization parameters.

6. The method for optimizing the anti-interference shielding of new energy wiring harness according to claim 5 is characterized in that: The step of creating a multi-stage filter structure based on the adaptive filter optimization parameters and performing eigenvalue decomposition orthogonal optimization to obtain multi-stage filter structure parameters includes: Decomposing the optimization parameters of the adaptive filter, splitting the 64-order FIR filter into a 3-stage cascade structure, with the order of each stage of the filter being 32, 16 and 16 respectively, to obtain initial multi-stage filter parameters; Constructing a transfer function model of a multi-stage filter structure based on the initial multi-stage filter parameters, using Z transform to represent the transfer function of each filter stage, and obtaining a total transfer function of the multi-stage filter system; Calculating the frequency response of the multi-stage filtering system based on the total transfer function to obtain frequency response data, and analyzing the frequency response data to calculate passband fluctuation, stopband attenuation and transition band width indicators to obtain initial filtering performance indicators; Based on the initial filtering performance index, combined with the frequency response characteristics of the wiring harness and the maximum allowable signal distortion, weight coefficients are set to construct a multi-objective optimization function; Performing eigenvalue decomposition on the coefficient matrix of the multi-objective optimization function, calculating the eigenvector matrix, and obtaining an orthogonal basis vector set; Performing a projection transformation on the initial multi-stage filter parameters based on the orthogonal basis vector set, projecting the parameter vectors into the orthogonal space spanned by the feature vectors, and obtaining orthogonalized filter parameters; Based on the orthogonalized filter parameters, the conjugate gradient method is used for iterative optimization until the maximum number of iterations reaches 500 or the objective function value changes by less than 10 -8 , and obtain the multi-stage filtering structure parameters.

7. The new energy wiring harness anti-interference shielding optimization method according to claim 1 is characterized in that: Based on the multi-stage filtering structure parameters, the electromagnetic compatibility of the whole vehicle of the new energy vehicle is distributedly optimized to obtain the anti-interference shielding optimization scheme of the whole vehicle wiring harness, including: Divide the new energy vehicle into four areas: a power system, a control system, a communication system, and an auxiliary system, and obtain a vehicle area division plan; Based on the vehicle area division scheme, the importance of the wiring harness in each area is evaluated, and the weight coefficient of each wiring harness is calculated using the hierarchical analysis method to obtain the importance score of the wiring harness; The wire bundles in each area are sorted based on the wire bundle importance score, and the top three wire bundles with the highest scores in each area are selected as target wire bundles to obtain a target wire bundle set; Applying the multi-stage filtering structure parameters to each wire harness in the target wire harness set to construct a regional optimization sub-model set for local electromagnetic compatibility optimization; Distributed learning is performed based on the regional optimization sub-model set, a local optimization unit is assigned to each sub-model, and a central control unit is set to obtain a distributed optimization network; Federated learning is performed based on the distributed optimization network, and each local optimization unit processes its own sub-model in parallel, updates local parameters, and obtains local optimization results; The local optimization results are aggregated, the central control unit collects the model parameters of each local optimization unit, and updates the global model using a weighted average method to obtain global optimization parameters; Based on the global optimization parameters, an anti-interference shielding solution for the entire vehicle wiring harness is generated, and the anti-interference shielding solution for the entire vehicle wiring harness includes filter parameters, shielding structure design, and wiring optimization suggestions for each target wiring harness.

8. A new energy wiring harness anti-interference shielding optimization device, characterized in that: The device is used to perform the new energy wiring harness anti-interference shielding optimization method according to any one of claims 1 to 7, and comprises: The sampling module is used to perform high-precision synchronous sampling of voltage and current signals at both ends of multiple wiring harnesses in new energy vehicles to obtain raw data of multiple wiring harness signals; A transformation module, used for performing piecewise overlapping fast Fourier transformation and interpolation processing on the plurality of wire beam signal original data to obtain a plurality of wire beam signal harmonic component data; A processing module, used for calculating the anti-interference performance of the wiring harness and performing adaptive noise elimination on the harmonic component data of the plurality of wiring harness signals to obtain a plurality of wiring harness anti-interference performance indicators; A calculation module, used for inputting the plurality of wire harness anti-interference performance indicators into a variable step-size regularized minimum mean square error model to perform filter parameter optimization calculation to obtain adaptive filter optimization parameters; A creation module is used to create a multi-stage filter structure based on the adaptive filter optimization parameters, and perform eigenvalue decomposition orthogonal optimization to obtain multi-stage filter structure parameters; The optimization module is used to perform distributed optimization on the electromagnetic compatibility of the whole vehicle of the new energy vehicle based on the multi-stage filtering structure parameters to obtain an anti-interference shielding optimization solution for the whole vehicle wiring harness.

9. An electronic device, characterized in that: The electronic device comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory so that the electronic device executes the new energy wiring harness anti-interference shielding optimization method as described in any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the new energy wiring harness anti-interference shielding optimization method as described in any one of claims 1-7 is implemented.

Images