An intelligent high-temperature resistance testing method for electronic wiring harnesses based on dynamic temperature monitoring

Abstract

This invention discloses an intelligent high-temperature resistance testing method for electronic wiring harnesses based on dynamic temperature monitoring. This method uses a sensor array to collect temperature data from multiple regions and dynamically adjusts the measurement point density and sampling frequency for areas with high temperature change rates. For localized high-temperature areas, the invention adds sampling points and triggers a high-temperature micro-change monitoring mechanism to record small temperature changes in real time. Based on the temperature gradient trend, the invention extracts dynamic adjustment parameters for key areas and optimizes the measurement point density distribution. Furthermore, the invention considers the optimal utilization of computing resources and appropriately reduces the measurement point density in temperature-stable areas. Through continuous monitoring and algorithm parameter optimization, the invention ultimately achieves stable high-temperature evaluation, keeping the temperature evaluation error within 1°C. This dynamic, adaptive temperature monitoring method significantly improves the accuracy and efficiency of temperature monitoring during electronic wiring harness operation.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to an intelligent high-temperature resistance testing method for an electronic wiring harness based on dynamic temperature monitoring. Background Art

[0002] As an indispensable core component in modern industry and technology, electronic wiring harnesses' performance in high-temperature environments directly impacts system reliability and safety. With the increasing demand for high-temperature resistance in aerospace, automotive electronics, and energy equipment, the development of accurate and efficient high-temperature resistance evaluation methods has become a critical issue. The stability and lifespan of electronic wiring harnesses in extreme temperatures are not only fundamental to technological innovation but also essential for ensuring safe equipment operation.

[0003] However, current evaluation methods often have difficulty balancing accuracy and efficiency, limiting their applicability in complex application scenarios. Existing high-temperature resistance evaluation methods often rely on the traditional model of fixed measurement point intervals, usually using evenly distributed measurement points for temperature collection. This method is prone to insufficient data or waste of resources in local high-temperature areas when faced with uneven distribution or dynamic changes in the temperature field. Especially in scenarios with significant high-temperature gradients, fixed intervals cannot flexibly adapt to the severity of temperature changes, resulting in evaluation results that are either too rough, ignoring minor changes, or too redundant, adding unnecessary computational burdens. These limitations make it difficult for existing methods to meet the dual demands of high accuracy and high efficiency.

[0004] In this area, the core challenge lies in how to achieve dynamic matching of measurement point density and temperature changes. Specifically, the uncertainty of the temperature change rate, the ability to adjust the measurement point interval in real time, and the optimal allocation of computing resources have become technical bottlenecks that need to be broken through. Because the temperature change rate varies significantly in different regions and time periods, the unresolved dynamic adjustment problem will make it difficult for the evaluation system to accurately capture local high-temperature micro-changes, and it will also be impossible to effectively reduce resource consumption when the temperature is stable. In addition, the intelligent regulation of measurement point density involves the coordination of algorithms and real-time data processing, which increases the complexity of technical implementation. These difficulties directly affect the accuracy and practicality of the evaluation.

[0005] Therefore, the key issue in this research is how to adaptively adjust the measurement point interval based on the temperature change rate of the electron beam to strike a balance between capturing micro-variations at high temperatures and optimizing computing resources. Solving this problem will provide a new technical path to improve the accuracy and efficiency of high-temperature resistance evaluation. Summary of the Invention

[0006] To solve the above technical problems, the present invention provides an intelligent high-temperature resistance testing method for electronic wiring harnesses based on dynamic temperature monitoring, comprising:

[0007] S1, collects temperature values ​​of multiple regions during the operation of the electronic beam through the sensor array, marks the initial measurement point position for the region where the temperature change rate exceeds 1.5 times the average value, and obtains the basic temperature distribution data set;

[0008] S2, calculating the temperature change rate of each area from the temperature distribution basic data set. If the temperature change rate exceeds the first change rate threshold, the measurement point density in the corresponding area is increased by 50% to determine the dynamic measurement point density distribution plan;

[0009] S3, adjusting the measurement point interval according to the dynamic measurement point density distribution plan, obtaining a high-frequency acquisition instruction once per second for the area where the temperature change rate exceeds the first change rate threshold, and obtaining real-time adjusted measurement point layout data;

[0010] S4, performing data collection based on the real-time adjusted measurement point layout data, doubling the number of sampling points for local high-temperature areas where the temperature exceeds a first temperature threshold. If the temperature exceeds a second temperature threshold, a high-temperature micro-change monitoring mechanism is triggered to record micro-changes once per second to obtain high-temperature micro-change characteristic data;

[0011] S5, analyzing the temperature gradient change trend based on the high-temperature micro-change characteristic data, extracting key area dynamic adjustment parameters from the area where the temperature gradient is greater than the temperature gradient threshold, and determining the optimized measurement point density distribution;

[0012] S6, reconfiguring computing resources based on the optimized measurement point density distribution. If the computing resource usage is lower than the resource threshold, reducing the measurement point density in the stable area where the temperature change rate is lower than the second change rate threshold by 30%, thereby obtaining a resource-optimized acquisition plan.

[0013] S7, continuously monitor the temperature field changes according to the resource-optimized acquisition plan, determine the stable state from the area where the high temperature slight change amplitude is lower than the amplitude threshold, and obtain the final measurement point interval adjustment data;

[0014] S8, adjust the data from the final measurement point interval to update the sensor array acquisition frequency, and collaboratively optimize the algorithm parameters based on the temperature change rate and local high temperature characteristics to make the temperature evaluation error less than 1°C and obtain stable high temperature evaluation data.

[0015] Optionally, step S1, collecting temperature values ​​of multiple regions during operation of the electronic beam through a sensor array, marking the initial measurement point positions for regions where the temperature change rate exceeds 1.5 times the average value, and obtaining a basic temperature distribution data set, includes:

[0016] Step S11, collecting temperature values ​​of multiple regions during operation through a sensor array to obtain a basic temperature distribution data set;

[0017] Step S12, calculating the temperature change rate of each region based on the temperature distribution basic data set, and determining the region where the change rate exceeds 1.5 times the average value;

[0018] Step S13: If the temperature change rate exceeds 1.5 times the average value, the initial measurement point position is marked in the exceeded area to obtain a marked data set;

[0019] Step S14: extracting temperature distribution features based on the labeled data set and dividing the temperature anomaly area using the K-means clustering algorithm;

[0020] Step S15, analyzing the temporal and spatial variation trend of the operating temperature through the temperature anomaly area to obtain the variation trend distribution;

[0021] Step S16, performing smoothing processing on the change trend distribution to obtain an optimized temperature distribution model;

[0022] Step S17: Use the optimized temperature distribution model to predict potential abnormal areas and determine the distribution pattern of the abnormal areas.

[0023] Optionally, step S2, calculating the temperature change rate of each region from the temperature distribution basic data set, and if the temperature change rate exceeds a first change rate threshold, increasing the measurement point density in the corresponding region by 50%, and determining a dynamic measurement point density distribution scheme, includes:

[0024] Step S21, obtaining temperature distribution data through the basic data set, processing it by region, and calculating the temperature change rate value of each region;

[0025] Step S22: if the temperature change rate exceeds the first change rate threshold, performing an increase operation on the density of the measuring points in the corresponding area to obtain adjusted density data;

[0026] Step S23, using the adjusted density data to update the dynamic distribution and generate a preliminary distribution plan;

[0027] Step S24, extracting the change trend from the preliminary distribution scheme, performing data smoothing processing on the change trend, and obtaining a smoothed distribution result;

[0028] Step S25, detecting abnormal areas through the smoothed distribution results, and if abnormal areas exist, recalculating the temperature change rate value for the abnormal areas and updating the dynamic distribution;

[0029] Step S26, obtaining updated dynamic distribution data and determining a final distribution plan based on regional division;

[0030] Step S27: output the measurement point density adjustment result according to the final distribution plan, and complete the dynamic measurement point density distribution processing.

[0031] Optionally, step S3, adjusting the measurement point interval according to the dynamic measurement point density distribution scheme, obtaining a high-frequency acquisition instruction once per second for an area where the temperature change rate exceeds a first change rate threshold, and obtaining real-time adjusted measurement point layout data, includes:

[0032] Step S31, acquiring temperature change data through a sensor network, and determining an area where a change rate exceeds a first change rate threshold;

[0033] Step S32, generating a high-frequency acquisition instruction for the area where the rate of change exceeds the limit, and obtaining sampling data once per second;

[0034] Step S33, using the K-means clustering algorithm to process the sampled data and determine the need to adjust the measurement point interval;

[0035] Step S34: updating the measurement point density distribution plan according to the adjustment requirements, and obtaining real-time updated measurement point layout data;

[0036] Step S35, extracting regional targeted features from the real-time layout data to determine the temperature change trend;

[0037] Step S36, optimizing the acquisition instructions through change trend analysis to obtain the sampling frequency adjustment data for the next cycle;

[0038] Step S37: Use a decision tree algorithm to process the adjusted data and determine the final layout of the measurement point density distribution plan.

[0039] Optionally, step S4 performs data collection based on the real-time adjusted measurement point layout data, doubles the number of sampling points for a local high-temperature area where the temperature exceeds a first temperature threshold, and triggers a high-temperature micro-change monitoring mechanism to record micro-changes once per second if the temperature exceeds a second temperature threshold, thereby obtaining high-temperature micro-change characteristic data, including:

[0040] Step S41, acquiring initial data through a pre-established measurement point layout, performing an adjustment to double the number of sampling points in a local high temperature area, and obtaining an expanded data acquisition result;

[0041] Step S42: extracting the temperature value from the expanded data collection result. If the temperature value exceeds the second temperature threshold, activating the micro-change monitoring mechanism to obtain micro-change data recorded every second.

[0042] Step S43, using a support vector machine algorithm to classify the slight change data and determine the high temperature slight change characteristic data;

[0043] Step S44, calculating the change trend through the high temperature slight change characteristic data to obtain the trend direction and amplitude information;

[0044] Step S45, adjusting the distribution of sampling points according to the trend direction and amplitude information to obtain optimized measurement point layout data;

[0045] Step S46, re-collecting data from the optimized measurement point layout data, determining whether there are still areas with excessive temperatures, and obtaining an updated high-temperature area distribution;

[0046] Step S47 : Repeat the above-mentioned subtle change monitoring and feature extraction for the updated high-temperature area distribution to obtain final high-temperature subtle change feature data.

[0047] Optionally, step S5, analyzing the temperature gradient variation trend based on the high-temperature slight variation characteristic data, extracting key area dynamic adjustment parameters from areas where the temperature gradient is greater than the temperature gradient threshold, and determining the optimized measurement point density distribution, includes:

[0048] Step S51, obtaining the temperature gradient value from the high temperature slight change characteristic data, and generating the change trend distribution using the gradient calculation method;

[0049] Step S52: Based on the change trend distribution, a threshold judgment method is used to extract areas where the temperature gradient is greater than the temperature gradient threshold, and determine the key area range;

[0050] Step S53: adjusting parameters using the Kriging interpolation method for the key area to generate a dynamically adjusted parameter set;

[0051] Step S54: obtaining the distribution characteristics of the measuring points from the dynamically adjusted parameter set, determining whether the density of measuring points is greater than 10 per square kilometer, and determining a preliminary density distribution;

[0052] Step S55, optimizing the measurement point layout using the Thiessen polygon method based on the preliminary density distribution to generate an optimized measurement point density distribution;

[0053] Step S56, extracting a distribution scheme from the optimized measurement point density distribution, judging whether the scheme covers more than 90% of the high-temperature slight change characteristic data area, and determining the final distribution scheme;

[0054] Step S57: Based on the final distribution scheme, ArcGIS is used to generate visualization results of the high-temperature micro-variation characteristic data to obtain a complete analysis output.

[0055] Optionally, step S6, reconfiguring computing resources based on the optimized measurement point density distribution, and if the computing resource usage is lower than the resource threshold, reducing the measurement point density in the stable region where the temperature change rate is lower than the second change rate threshold by 30%, thereby obtaining a resource-optimized acquisition plan, includes:

[0056] Step S61, obtaining computing resource occupancy data through the optimized measurement point density distribution, and determining whether the resource occupancy is lower than the resource threshold;

[0057] Step S62: if the resource occupancy is lower than the resource threshold, extracting temperature change rate data from the measurement point density distribution, and using a temperature change rate analysis tool to determine a stable region where the temperature change rate is lower than a second change rate threshold;

[0058] Step S63: using a density adjustment tool to reduce the measurement point density by 30% for the measurement point density data in the stable area, to obtain an adjusted density distribution;

[0059] Step S64: Using the adjusted density distribution, recalculate resource usage using a resource monitoring tool to determine whether resource optimization requirements are met.

[0060] Step S65: Generate a new acquisition plan using a rule engine based on the optimized resource status to obtain the measurement point configuration after resource optimization;

[0061] Step S66: extract the temperature change rate and resource occupancy data from the new acquisition plan, use the optimization judgment tool to determine whether further adjustments are still needed, and obtain the final acquisition plan;

[0062] Step S67: According to the final acquisition plan, use the density adjustment tool to adjust the density distribution of the measurement points and determine the operating parameters after resource optimization.

[0063] Optionally, step S7, continuously monitoring the temperature field changes according to the resource-optimized acquisition plan, judging the stable state from the area where the high-temperature slight change amplitude is lower than the amplitude threshold, and obtaining the final measurement point interval adjustment data, includes:

[0064] Step S71, obtaining temperature field change data through an acquisition scheme, and extracting micro-change amplitude features for high temperature areas;

[0065] Step S72, judging whether the high temperature slight change amplitude data is lower than the amplitude threshold, if it is lower than the amplitude threshold, determining that the region has reached a stable state;

[0066] Step S73, calculating a preliminary adjustment value of the measurement point interval based on the temperature distribution characteristics of the stable state area;

[0067] Step S74, using the continuous monitoring data to verify the measurement point interval adjustment value, and determining the accuracy of the adjustment value by comparing the temperature change amplitude before and after the adjustment;

[0068] Step S75, performing resource optimization processing on the adjustment value to generate an optimized measurement point distribution plan;

[0069] Step S76: updating the acquisition plan according to the optimized measurement point distribution plan to obtain the final temperature field monitoring data;

[0070] Step S77: extract the temperature change trend characteristics from the final monitoring data to determine the continuity of the temperature field stable state.

[0071] Optionally, step S8, adjusting the data based on the final measurement point interval to update the sensor array acquisition frequency, collaboratively optimizing algorithm parameters based on the temperature change rate and local high temperature characteristics, so that the temperature evaluation error is less than 1°C, and obtaining stable high temperature evaluation data, includes:

[0072] Step S81, obtaining initial acquisition frequency data through the sensor array, adjusting the acquisition frequency according to the measurement point interval, and obtaining a preliminary frequency distribution;

[0073] Step S82, extracting temperature data from the preliminary frequency distribution, calculating the temperature change rate using a differential method, extracting a change trend from the temperature change rate, and determining a basis for dynamic adjustment;

[0074] Step S83: If the change trend exceeds the change threshold, the algorithm parameters are adjusted by the gradient descent method to obtain an optimized parameter set;

[0075] Step S84, using the optimized parameter set to process the local high temperature data, extracting the high temperature distribution from the local high temperature features, and obtaining feature enhanced data;

[0076] Step S85, updating the acquisition frequency through feature enhancement data, recalculating the change value based on the temperature change rate, and determining whether the evaluation error is reduced;

[0077] Step S86: If the evaluation error does not meet the standard, the high temperature feature is processed by the random forest algorithm to obtain stable high temperature data;

[0078] Step S87: adjusting the operating frequency of the sensor array according to the stable high temperature data, and generating final high temperature evaluation data from the adjusted frequency.

[0079] The present invention discloses an intelligent high-temperature resistance testing method for an electronic wiring harness based on dynamic temperature monitoring. The method collects temperature data from multiple regions through a sensor array, and dynamically adjusts the measurement point density and sampling frequency for regions with higher temperature change rates. For local high-temperature regions, the present invention increases sampling points and triggers a high-temperature micro-change monitoring mechanism to record small temperature changes in real time. According to the temperature gradient change trend, the present invention extracts dynamic adjustment parameters in key areas and optimizes the distribution of measurement point density. At the same time, the present invention also considers the optimal utilization of computing resources and appropriately reduces the measurement point density in temperature-stable regions. Through continuous monitoring and algorithm parameter optimization, the present invention ultimately achieves stable high-temperature evaluation and controls the temperature evaluation error within 1°C. This dynamic and adaptive temperature monitoring method significantly improves the temperature monitoring accuracy and efficiency during the operation of the electronic wiring harness. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 The present invention is a flow chart of an intelligent high-temperature resistance testing method for an electronic wiring harness based on dynamic temperature monitoring. DETAILED DESCRIPTION

[0081] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0082] like Figure 1 The present invention provides an intelligent high-temperature resistance testing method for an electronic wiring harness based on dynamic temperature monitoring, comprising:

[0083] S1, collects the temperature values ​​of multiple areas when the electronic beam is running through the sensor array, marks the initial measurement point position for the area where the temperature change rate exceeds 1.5 times the average value, and obtains the basic temperature distribution data set.

[0084] Optionally, this step also includes:

[0085] Step S11 , collecting temperature values ​​of multiple regions during operation through a sensor array to obtain a basic temperature distribution data set.

[0086] Step S12: Calculate the temperature change rate of each region based on the temperature distribution basic data set, and determine the region where the change rate exceeds 1.5 times the average value.

[0087] Step S13: If the temperature change rate exceeds 1.5 times the average value, the initial measurement point position is marked in the exceeded area to obtain a marked data set.

[0088] Step S14: extract temperature distribution features based on the labeled data set, and use the K-means clustering algorithm to divide the temperature anomaly area.

[0089] Optionally, calculate the temperature anomaly index using the following formula: ;

[0090] Where R represents the temperature anomaly index, represents the temperature value of the i-th point, μ represents the average temperature of the region, σ represents the standard deviation of the temperature, and n represents the total number of sampling points.

[0091] Step S15: analyzing the spatiotemporal variation trend of the operating temperature through the temperature anomaly area to obtain the variation trend distribution.

[0092] Step S16: performing smoothing processing on the change trend distribution to obtain an optimized temperature distribution model.

[0093] Specifically, based on the temporal and spatial trends of temperature anomaly areas, the system optimizes the temperature distribution model using a Gaussian smoothing algorithm. This smoothing applies a Gaussian kernel function with a sliding window size of five time points to the temperature data. For example, if the temperature in a certain area is 25°C, 26°C, 27°C, 28°C, and 29°C at five consecutive time points, the temperature in that area will stabilize at around 27°C after Gaussian smoothing, reducing data fluctuations. The system then uses time series analysis to forecast the smoothed temperature data using the Autoregressive Integrated Moving Average (ARIMA) model. For example, after training the model using historical temperature data, it predicts that the temperature trend in a certain area will gradually increase from 28°C to 32°C over the next 10 seconds. Based on this, the system classifies the temperature anomaly areas using a clustering algorithm. The K-means clustering algorithm groups areas with similar temperature trends into the same category. For example, areas with a temperature change rate of 2°C / second to 4°C / second are grouped into one category, while areas with a temperature change rate of 5°C / second or more are grouped into another. Based on the clustering results, the system further analyzes the distribution patterns of potential abnormal areas. For example, it found that temperature anomalies are mostly concentrated in the middle of the electronic wiring harness. The temperature changes in these anomaly areas are periodic, with significant temperature fluctuations occurring every 30 seconds. Finally, the system integrates these analysis results into a conclusion on the distribution patterns of potential abnormal areas, providing a basis for subsequent monitoring of the electronic wiring harness's operating status.

[0094] According to the labeled temperature distribution data set, the K-means clustering algorithm is used to divide the temperature anomaly area. By analyzing the spatiotemporal change trend of the abnormal area, the change trend distribution is obtained and smoothed to obtain the optimized temperature distribution prediction model, thereby determining the distribution law of the abnormal area.

[0095] Step S17: Use the optimized temperature distribution model to predict potential abnormal areas and determine the distribution pattern of the abnormal areas.

[0096] Specifically, the sensor array is used to collect temperature values ​​in multiple areas of the electronic harness during operation, and distributed temperature sensors are arranged in multiple key areas of the electronic harness. For example, a sensor is set every 10 cm to ensure comprehensive temperature monitoring. The sensor array collects temperature data at a frequency of once per second and uploads it to the data processing center in real time through a wireless transmission module. The data processing center first pre-processes the collected temperature data, including removing noise and outliers. The filtering algorithm used is Kalman filtering, which makes the data smoother and more reliable. Next, the temperature change rate of each area is calculated, and the temperature difference between adjacent time points is calculated by the difference method, and then divided by the time interval to obtain the change rate.

[0097] For example, if the temperature of an area is 23°C and 28°C at two consecutive time points, with a time interval of 1 second, the temperature change rate is 5°C / second. Then, calculate the average temperature change rate for all areas, assuming it is 0°C / second. Any area with a temperature change rate exceeding 1.5 times the average (i.e., a rate of change greater than 5°C / second) is marked as the initial measurement point.

[0098] For example, a region with a temperature change rate of 8°C / second, exceeding 1.5 times the average, is marked as the initial measurement point. Ultimately, the temperature data from all marked regions is integrated to form a basic temperature distribution dataset for subsequent analysis and optimization. Through these steps, the system can automatically identify areas of abnormal temperature, providing data support for the safe operation of the electronic wiring harness.

[0099] According to the temporal and spatial variation trend distribution of temperature anomaly areas, the smoothing method is used to optimize the temperature distribution model and obtain the distribution law judgment results of potential anomaly areas.

[0100] Specifically, based on the labeled temperature distribution dataset, the system used the K-means clustering algorithm to segment temperature anomaly regions, setting the number of cluster centers to three. Using Euclidean distance to calculate the distance from each data point to a cluster center, the temperature data was divided into three categories: low-temperature (18°C to 22°C), medium-temperature (23°C to 27°C), and high-temperature (28°C to 32°C). The system then performed a spatiotemporal trend analysis of the high-temperature regions, extracting the temperature change rate every 10 seconds. The system found that the high-temperature regions exhibited a temporal trend of first increasing and then decreasing. For example, the temperature change rate for a particular region within 10 consecutive seconds was 5°C / second, 8°C / second, 2°C / second, 9°C / second, and 6°C / second. To reduce the impact of data fluctuations on trend analysis, the system used a moving average algorithm to smooth the temperature change rate. Setting the window size to 3, the smoothed temperature change rates were 7°C / second, 0°C / second, and 9°C / second. Based on the smoothed data, the system used exponential smoothing to construct a temperature distribution prediction model, setting a smoothing coefficient of 3. It predicted that the temperature in the high-temperature area would rise from 29°C to 31°C within the next 5 seconds. Analysis of the prediction results revealed that the high-temperature areas were mostly concentrated near the equipment heat sinks, and that the temperature change trend was positively correlated with the equipment's operating load. For example, for every 10% increase in equipment load, the temperature in the high-temperature area rose by approximately 5°C.

[0101] S2: Calculate the temperature change rate of each area from the temperature distribution basic data set. If the temperature change rate exceeds the first change rate threshold of 5°C / minute, increase the measurement point density in the corresponding area by 50% to determine the dynamic measurement point density distribution plan.

[0102] Optionally, this step also includes:

[0103] Step S21 : obtaining temperature distribution data through the basic data set, processing it by region, and calculating the temperature change rate value of each region.

[0104] Step S22: If the temperature change rate exceeds the first change rate threshold, an increase operation is performed on the density of the measuring points in the corresponding area to obtain adjusted density data.

[0105] Step S23: Use the adjusted density data to update the dynamic distribution and generate a preliminary distribution plan.

[0106] Step S24: extracting the change trend from the preliminary distribution scheme, performing data smoothing processing on the change trend, and obtaining a smoothed distribution result.

[0107] Step S25: detecting abnormal areas through the smoothed distribution results. If abnormal areas exist, recalculating the temperature change rate value for the abnormal areas and updating the dynamic distribution.

[0108] Step S26: Obtain updated dynamic distribution data and determine the final distribution plan based on regional division.

[0109] Step S27: output the measurement point density adjustment result according to the final distribution plan, and complete the dynamic measurement point density distribution processing.

[0110] Specifically, the temperature values ​​of each area are extracted from the temperature distribution basic data set. Assuming that the initial temperature of a certain area is 25°C and the temperature is 32°C after one minute, the temperature change rate is (32-25) / 1=7°C / minute. If the preset first change rate threshold is 5°C / minute, then 7°C / minute exceeds the first change rate threshold. At this time, the system automatically increases the measurement point density in the area by 50%. Assuming that the original measurement point density is 10 per square meter, it will be increased to 15 per square meter. By dynamically adjusting the measurement point density in this way, areas with large temperature changes can be monitored more accurately, ensuring the accuracy and real-time nature of the data. The system will dynamically adjust the measurement point density of each area according to the real-time temperature change rate to form an optimized measurement point distribution plan, thereby improving the overall monitoring effect.

[0111] S3, adjusting the measurement point interval according to the dynamic measurement point density distribution plan, obtaining a high-frequency acquisition instruction once per second for the area where the temperature change rate exceeds the first change rate threshold of 5°C / minute, and obtaining real-time adjusted measurement point layout data.

[0112] Optionally, this step also includes:

[0113] Step S31 : acquiring temperature change data through a sensor network, and determining an area where a change rate exceeds a first change rate threshold.

[0114] Step S32: generating a high-frequency acquisition instruction for the area where the rate of change exceeds the limit, and obtaining sampling data once per second.

[0115] Step S33: Use the K-means clustering algorithm to process the sampled data and determine the need for adjusting the measurement point interval.

[0116] Step S34: updating the measurement point density distribution plan according to the adjustment requirements, and obtaining real-time updated measurement point layout data.

[0117] Step S35: extracting regional targeted features from the real-time layout data to determine the temperature change trend.

[0118] Step S36: Optimize the acquisition instructions through change trend analysis to obtain the sampling frequency adjustment data for the next cycle.

[0119] Step S37: Use a decision tree algorithm to process the adjusted data and determine the final layout of the measurement point density distribution plan.

[0120] Specifically, based on the dynamic measurement point density distribution scheme, the system first uses an adaptive grid division algorithm to preliminarily divide the monitoring area according to the changing characteristics of the temperature field.

[0121] For example, in the initial stage, the system covers the entire monitoring area based on a 10m x 10m grid. Then, the system calculates the temperature changes within each grid by analyzing the historical data of the temperature sensor in real time. When the temperature change within a grid exceeds 5°C, the system automatically subdivides the grid into smaller subgrids, such as 2m x 2m, to more accurately capture temperature fluctuations. In the refined grid, the system generates a high-frequency acquisition instruction once per second to ensure the real-time and accuracy of the temperature data. At the same time, the system merges adjacent high-change rate grids through optimization algorithms, such as the K-means clustering algorithm, to form a reasonable measurement point layout.

[0122] For example, if the temperature variation between two adjacent grids exceeds 5°C, the system merges them into a larger measurement area to reduce redundant data and improve monitoring efficiency. Ultimately, the system outputs real-time adjusted measurement point layout data, including measurement point coordinates, acquisition frequency, and other information, providing reliable data support for subsequent temperature monitoring and analysis.

[0123] S4, perform data collection from the measurement point layout data adjusted in real time, double the sampling points for local high temperature areas where the temperature exceeds the first temperature threshold of 80°C, and if the temperature exceeds the preset second temperature threshold of 90°C, trigger the high temperature micro-change monitoring mechanism to record small changes once per second to obtain high temperature micro-change characteristic data.

[0124] Optionally, this step also includes:

[0125] Step S41 , obtaining initial data through a pre-established measurement point layout, performing an adjustment to double the number of sampling points in a local high temperature area, and obtaining an expanded data acquisition result.

[0126] Step S42: extracting the temperature value from the expanded data collection result. If the temperature value exceeds the second temperature threshold, activating the micro-change monitoring mechanism to obtain micro-change data recorded every second.

[0127] Step S43: using a support vector machine algorithm to classify the slight change data and determine high temperature slight change feature data.

[0128] Step S44: Calculate the change trend through the high-temperature slight change characteristic data to obtain the trend direction and amplitude information.

[0129] Step S45 , adjusting the distribution of sampling points according to the trend direction and amplitude information to obtain optimized measurement point layout data.

[0130] Step S46 , re-collecting data from the optimized measurement point layout data, determining whether there are still areas with excessive temperatures, and obtaining an updated high-temperature area distribution.

[0131] Step S47 : Repeat the above-mentioned subtle change monitoring and feature extraction for the updated high-temperature area distribution to obtain final high-temperature subtle change feature data.

[0132] Specifically, data collection is performed based on the real-time adjusted measurement point layout data. First, the current temperature distribution is obtained through the sensor network. Assume that the initial number of measurement points is 100, distributed on the surface of the equipment. When the temperature in a local area is detected to exceed 80°C, the system automatically doubles the number of sampling points in that area. For example, in a high-temperature area, the original number of measurement points is 10, and after the increase, it becomes 20, ensuring that the data collection density is high enough to capture the details of temperature changes. If the temperature further rises to above 90°C, the system immediately triggers the high-temperature micro-change monitoring mechanism, recording small changes once a second. The temperature data is smoothed using a time series analysis method combined with the Kalman filter algorithm to extract the characteristics of the temperature change.

[0133] For example, at a certain moment, the temperature rises from 91°C to 93°C. The system records this tiny change and uses an FFT (Fast Fourier Transform) algorithm to analyze its frequency characteristics and identify whether there are periodic fluctuations. This high-temperature micro-variation characteristic data is used for subsequent fault prediction and health management, helping engineers accurately determine the operating status of the equipment and provide early warning of potential failure risks.

[0134] S5, analyze the temperature gradient change trend based on the high-temperature micro-change characteristic data, extract the key area dynamic adjustment parameters from the area where the temperature gradient is greater than the temperature gradient threshold of 2°C / cm, and determine the optimized measurement point density distribution.

[0135] Optionally, this step also includes:

[0136] Step S51 , obtaining a temperature gradient value from the high-temperature slight change characteristic data, and generating a change trend distribution using a gradient calculation method.

[0137] Step S52 : Based on the change trend distribution, a threshold judgment method is used to extract areas where the temperature gradient is greater than the temperature gradient threshold, and determine the range of the key area.

[0138] Step S53: For the key area, use the Kriging interpolation method to adjust the parameters and generate a dynamically adjusted parameter set.

[0139] Step S54 , obtaining the distribution characteristics of the measuring points from the dynamically adjusted parameter set, judging whether the density of the measuring points is greater than 10 per square kilometer, and determining a preliminary density distribution.

[0140] Step S55 : optimizing the measurement point layout by using the Thiessen polygon method based on the preliminary density distribution to generate an optimized measurement point density distribution.

[0141] Step S56: extract a distribution scheme from the optimized measurement point density distribution, determine whether the scheme covers more than 90% of the high-temperature slight change characteristic data area, and determine the final distribution scheme.

[0142] Step S57: Based on the final distribution scheme, ArcGIS is used to generate visualization results of the high-temperature micro-variation characteristic data to obtain a complete analysis output.

[0143] Specifically, in the analysis of high-temperature micro-variation data, temperature data was first collected from sensors. The raw data was smoothed using a Gaussian filter algorithm to remove noise, with the filter parameter set to σ = 5. Next, a spatial interpolation method (such as Kriging) was used to generate a temperature gradient distribution map, with an interpolation step size of 1 cm. In the temperature gradient analysis, a temperature gradient threshold of 2°C / cm was set, and a region growing algorithm was used to identify key areas with gradients greater than 2°C / cm. A seed point growing threshold of 5°C was used to ensure regional connectivity. Within these key areas, the K-means clustering algorithm was used to dynamically adjust the measurement point distribution based on the calculated dispersion of temperature changes, with a set number of cluster centers of 5 and 100 iterations. The measurement point density was optimized by calculating the temperature standard deviation within the area, increasing the measurement point density to 4 points per square centimeter in high-gradient areas and reducing it to 1 point per square centimeter in low-gradient areas. Finally, sensors were redeployed based on the optimized measurement point density to ensure the accuracy and representativeness of the monitoring data.

[0144] S6: Reconfigure computing resources based on the optimized measurement point density distribution. If the computing resource usage is lower than the resource threshold of 70%, reduce the measurement point density in the stable area where the temperature change rate is lower than the second change rate threshold of 1°C / minute by 30%, and obtain a resource-optimized acquisition plan.

[0145] Optionally, this step also includes:

[0146] Step S61 : Obtain computing resource occupancy data through the optimized measurement point density distribution, and determine whether the resource occupancy is lower than a resource threshold of 70%.

[0147] Step S62: If the resource occupancy is lower than the resource threshold of 70%, the temperature change rate data is extracted from the measurement point density distribution, and a temperature change rate analysis tool is used to determine a stable area where the temperature change rate is lower than 1°C / minute.

[0148] In step S63 , for the measurement point density data in the stable area, a density adjustment tool is used to reduce the measurement point density by 30% to obtain an adjusted density distribution.

[0149] Step S64 , using the adjusted density distribution, and using a resource monitoring tool to recalculate resource occupancy to determine whether resource optimization requirements are met.

[0150] Step S65: Generate a new acquisition plan using a rule engine according to the optimized resource status, and obtain the measurement point configuration after resource optimization.

[0151] Step S66: extract the temperature change rate and resource occupancy data from the new acquisition plan, use an optimization judgment tool to determine whether further adjustments are still needed, and obtain a final acquisition plan.

[0152] Step S67: According to the final acquisition plan, use the density adjustment tool to adjust the density distribution of the measurement points and determine the operating parameters after resource optimization.

[0153] Specifically, in the optimized measurement point density distribution, reconfiguration is performed based on the occupancy of computing resources. First, the computing resource usage rate of each node is collected in real time through the monitoring system. If the computing resource occupancy rate of a certain area is found to be lower than the resource threshold of 70%, the measurement point density adjustment algorithm is started.

[0154] For example, in a temperature field simulation, the system detected that the computing resource utilization rate in a certain area was 65%, which was lower than the resource threshold. At the same time, the temperature change rate analysis module found that the temperature change rate in this area was stable below 8°C / minute, which met the definition of a stable area. The system then called the density optimization algorithm to reduce the measurement point density in this area by 30%, from the original 100 measurement points to 70 measurement points. At the same time, the interpolation algorithm was used to supplement the missing data to ensure that the simulation accuracy was not significantly affected. The specific algorithm uses a gradient descent-based optimization method to determine the optimal measurement point distribution through iterative calculations. While reducing the density, it dynamically adjusts the allocation of computing resources, allocating the released computing resources to areas with high change rates, such as hot spots with temperature change rates higher than 2°C / minute, thereby improving the overall simulation efficiency. Through this process, the system achieves dynamic optimization of computing resources, which not only reduces resource waste but also ensures data collection accuracy in key areas.

[0155] S7, according to the acquisition plan after resource optimization, continuously monitor the temperature field changes, judge the stable state from the area where the high temperature slight change amplitude is lower than the amplitude threshold of 0.5℃, and obtain the final measurement point interval adjustment data.

[0156] Optionally, this step also includes:

[0157] Step S71: Acquire temperature field change data through an acquisition scheme, and extract slight change amplitude features for high temperature areas.

[0158] Step S72: determine whether the high temperature slight change amplitude data is lower than the amplitude threshold. If it is lower than the preset threshold, it is determined that the region has reached a stable state.

[0159] Step S73: Calculate a preliminary adjustment value of the measurement point interval based on the temperature distribution characteristics of the stable state area.

[0160] Step S74: Verify the measurement point interval adjustment value using the continuous monitoring data, and determine the accuracy of the adjustment value by comparing the temperature change amplitude before and after the adjustment.

[0161] Step S75: performing resource optimization processing on the adjustment value to generate an optimized measurement point distribution plan.

[0162] Step S76: Update the acquisition plan according to the optimized measurement point distribution plan to obtain the final temperature field monitoring data.

[0163] Step S77: extract the temperature change trend characteristics from the final monitoring data to determine the continuity of the temperature field stable state.

[0164] Specifically, the resource-optimized data acquisition solution first monitors temperature field changes in real time through a distributed temperature sensor network, sampling once every minute to ensure high-precision data collection. The collected temperature data is preprocessed using a sliding average algorithm with a 10-minute window size to eliminate the influence of random noise. Next, based on the preprocessed data, the temperature field is smoothed using a Gaussian filter algorithm with a standard deviation of 2 to further highlight the overall trend of the temperature field. By comparing temperature field data at consecutive time points, the temperature change amplitude of each region is calculated. When the temperature change amplitude in a region is less than 5°C for 30 consecutive minutes, the region is considered to have reached a stable state. For stable regions, a dynamic adjustment algorithm is used to optimize the measurement point spacing. Specifically, the initial measurement point spacing is adjusted from 5 meters to 10 meters. An adaptive adjustment mechanism based on the temperature gradient is also introduced. When the temperature gradient is less than 1°C / meter, the measurement point spacing is further increased to 15 meters. Finally, simulations verify that the optimized measurement point spacing scheme significantly reduces resource consumption while maintaining monitoring accuracy, improving overall data acquisition efficiency by approximately 30%. Throughout the entire process, all algorithms and parameters are executed through an automated system to ensure the real-time and accuracy of the data.

[0165] S8, adjust the data from the final measurement point interval to update the sensor array acquisition frequency, and collaboratively optimize the algorithm parameters based on the temperature change rate and local high temperature characteristics to make the temperature evaluation error less than 1°C and obtain stable high temperature evaluation data.

[0166] Optionally, this step also includes:

[0167] Step S81 : acquiring initial acquisition frequency data through the sensor array, adjusting the acquisition frequency according to the measurement point interval, and obtaining a preliminary frequency distribution.

[0168] Step S82: extract temperature data from the preliminary frequency distribution, calculate the temperature change rate using the differential method, extract the change trend from the temperature change rate, and determine the basis for dynamic adjustment.

[0169] Step S83: If the change trend exceeds the change threshold, the algorithm parameters are adjusted by the gradient descent method to obtain an optimized parameter set.

[0170] Step S84: Process the local high temperature data using the optimized parameter set, extract the high temperature distribution from the local high temperature features, and obtain feature enhanced data.

[0171] Step S85: Update the acquisition frequency through feature enhancement data, recalculate the change value based on the temperature change rate, and determine whether the evaluation error is reduced.

[0172] Step S86: If the evaluation error does not meet the standard, the high temperature feature is processed by the random forest algorithm to obtain stable high temperature data.

[0173] Step S87: adjusting the operating frequency of the sensor array according to the stable high temperature data, and generating final high temperature evaluation data from the adjusted frequency.

[0174] Specifically, during the temperature monitoring process, initial data is first collected through the sensor array. The measurement point spacing is set at 10 cm, and the acquisition frequency is set at 10 times per second to ensure high-density data coverage. Based on this initial data, the temperature change rate is analyzed, and the temperature change curve is fitted using the least squares method. It is found that the temperature change rate in local areas reaches 5°C per second. To optimize the acquisition frequency, a dynamic adjustment algorithm is introduced to adjust the acquisition frequency in real time based on the temperature change rate. When the temperature change rate exceeds 3°C per second, the acquisition frequency is increased to 20 times per second to ensure data accuracy. Simultaneously, a Gaussian filter algorithm is used to smooth the data to eliminate noise interference and extract the core temperature value of the high-temperature area. By collaboratively optimizing the algorithm parameters, the standard deviation of the Gaussian filter is set to 2°C, ensuring that the temperature evaluation error in the local high-temperature area is controlled within 8°C. Ultimately, through multiple iterative optimizations, the stability of the high-temperature evaluation data output by the system has been significantly improved, with the evaluation error consistently below 1°C, meeting the requirements of high-precision temperature monitoring.

[0175] This invention discloses an intelligent high-temperature resistance testing method for electronic harnesses based on dynamic temperature monitoring. This method uses a sensor array to collect temperature data from multiple regions and dynamically adjusts the measurement point density and sampling frequency for areas with high temperature change rates. For localized high-temperature areas, the invention adds sampling points and triggers a high-temperature micro-change monitoring mechanism to record minute temperature changes in real time. Based on the temperature gradient trend, the invention extracts dynamic adjustment parameters for key areas and optimizes the measurement point density distribution. Furthermore, the invention considers the optimal utilization of computing resources and appropriately reduces the measurement point density in temperature-stable areas. Through continuous monitoring and algorithm parameter optimization, the invention ultimately achieves stable high-temperature evaluation, keeping the temperature evaluation error within 1°C. This dynamic, adaptive temperature monitoring method significantly improves the accuracy and efficiency of temperature monitoring during electronic harness operation.

[0176] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. An intelligent high temperature resistance testing method for electronic wiring harnesses based on dynamic temperature monitoring, characterized in that: The method comprises: S1, collects temperature values ​​of multiple regions during the operation of the electronic beam through the sensor array, marks the initial measurement point position for the region where the temperature change rate exceeds 1.5 times the average value, and obtains the basic temperature distribution data set; S2, calculating the temperature change rate of each area from the temperature distribution basic data set. If the temperature change rate exceeds a first change rate threshold, the measurement point density in the corresponding area is increased by 50% to determine a dynamic measurement point density distribution plan; S3, adjusting the measurement point interval according to the dynamic measurement point density distribution plan, obtaining a high-frequency acquisition instruction once per second for the area where the temperature change rate exceeds the first change rate threshold, and obtaining real-time adjusted measurement point layout data; S4, performing data collection based on the real-time adjusted measurement point layout data, doubling the number of sampling points for local high-temperature areas where the temperature exceeds a first temperature threshold. If the temperature exceeds a second temperature threshold, a high-temperature micro-change monitoring mechanism is triggered to record micro-changes once per second to obtain high-temperature micro-change characteristic data; S5, analyzing the temperature gradient change trend based on the high-temperature micro-change characteristic data, extracting key area dynamic adjustment parameters from the area where the temperature gradient is greater than the temperature gradient threshold, and determining the optimized measurement point density distribution; S6, reconfiguring computing resources based on the optimized measurement point density distribution. If the computing resource usage is lower than the resource threshold, reducing the measurement point density in the stable area where the temperature change rate is lower than the second change rate threshold by 30%, thereby obtaining a resource-optimized acquisition plan. S7, continuously monitor the temperature field changes according to the resource-optimized acquisition plan, determine the stable state from the area where the high temperature slight change amplitude is lower than the amplitude threshold, and obtain the final measurement point interval adjustment data; This step also includes: Step S71, obtaining temperature field change data through an acquisition scheme, and extracting micro-change amplitude features for high temperature areas; Step S72, judging whether the high temperature slight change amplitude data is lower than the amplitude threshold, if it is lower than the amplitude threshold, determining that the region has reached a stable state; Step S73, calculating a preliminary adjustment value of the measurement point interval based on the temperature distribution characteristics of the stable state area; Step S74, using the continuous monitoring data to verify the measurement point interval adjustment value, and determining the accuracy of the adjustment value by comparing the temperature change amplitude before and after the adjustment; Step S75, performing resource optimization processing on the adjustment value to generate an optimized measurement point distribution plan; Step S76: updating the acquisition plan according to the optimized measurement point distribution plan to obtain the final temperature field monitoring data; Step S77, extracting temperature change trend characteristics from the final monitoring data to determine the persistence of the temperature field stable state; In the resource-optimized acquisition solution, changes in the temperature field are first monitored in real time through a distributed temperature sensor network, with a sampling frequency set to once per minute. The collected temperature data is preprocessed using a sliding average algorithm. Then, based on the preprocessed data, the temperature field is smoothed using a Gaussian filter algorithm. By comparing the temperature field data at consecutive time points, the temperature variation in each area is calculated. When the temperature variation in a certain area is less than 5°C for 30 consecutive minutes, it is determined that the area has reached a stable state. For stable areas, a dynamic adjustment algorithm is used to optimize the measurement point interval. The specific method is as follows: the initial measurement point interval is adjusted from 5 meters to 10 meters, and an adaptive adjustment mechanism based on the temperature gradient is introduced. When the temperature gradient is less than 1°C / meter, the measurement point interval is further expanded to 15 meters. S8, adjust the data from the final measurement point interval to update the sensor array acquisition frequency, collaboratively optimize the algorithm parameters based on the temperature change rate and local high temperature characteristics, set the standard deviation of the Gaussian filter to 2°C, and through multiple iterative optimizations, reduce the temperature evaluation error to less than 1°C, and obtain stable high temperature evaluation data.

2. The method according to claim 1, characterized in that The step S1 collects temperature values ​​of multiple regions during the operation of the electronic beam through a sensor array, marks the initial measurement point position for the region where the temperature change rate exceeds 1.5 times the average value, and obtains a basic temperature distribution data set, including: Step S11, collecting temperature values ​​of multiple regions during operation through a sensor array to obtain a basic temperature distribution data set; Step S12, calculating the temperature change rate of each region based on the temperature distribution basic data set, and determining the region where the change rate exceeds 1.5 times the average value; Step S13: If the temperature change rate exceeds 1.5 times the average value, the initial measurement point position is marked in the exceeded area to obtain a marked data set; Step S14: extracting temperature distribution features based on the labeled data set and dividing the temperature anomaly area using the K-means clustering algorithm; Step S15, analyzing the temporal and spatial variation trend of the operating temperature through the temperature anomaly area to obtain the variation trend distribution; Step S16, performing smoothing processing on the change trend distribution to obtain an optimized temperature distribution model; Step S17: Use the optimized temperature distribution model to predict potential abnormal areas and determine the distribution pattern of the abnormal areas.

3. The method according to claim 1, characterized in that The step S2 calculates the temperature change rate of each region from the temperature distribution basic data set. If the temperature change rate exceeds a first change rate threshold, the measurement point density of the corresponding region is increased by 50%, and a dynamic measurement point density distribution scheme is determined, including: Step S21, obtaining temperature distribution data through the basic data set, processing it by region, and calculating the temperature change rate value of each region; Step S22: if the temperature change rate exceeds the first change rate threshold, performing an increase operation on the density of the measuring points in the corresponding area to obtain adjusted density data; Step S23, using the adjusted density data to update the dynamic distribution and generate a preliminary distribution plan; Step S24, extracting the change trend from the preliminary distribution scheme, performing data smoothing processing on the change trend, and obtaining a smoothed distribution result; Step S25, detecting abnormal areas through the smoothed distribution results, and if abnormal areas exist, recalculating the temperature change rate value for the abnormal areas and updating the dynamic distribution; Step S26, obtaining updated dynamic distribution data and determining a final distribution plan based on regional division; Step S27: Output the measurement point density adjustment result according to the final distribution plan to complete the dynamic measurement point density distribution processing; The system will dynamically adjust the measurement point density in each area according to the real-time temperature change rate to form an optimized measurement point distribution plan.

4. The method according to claim 1, wherein The step S3, adjusting the measurement point interval according to the dynamic measurement point density distribution scheme, obtaining a high-frequency acquisition instruction once per second for the area where the temperature change rate exceeds the first change rate threshold, and obtaining the measurement point layout data after real-time adjustment, includes: Step S31, acquiring temperature change data through a sensor network, and determining an area where a change rate exceeds a first change rate threshold; Step S32, generating a high-frequency acquisition instruction for the area where the rate of change exceeds the limit, and obtaining sampling data once per second; Step S33, using the K-means clustering algorithm to process the sampled data and determine the need to adjust the measurement point interval; Step S34: updating the measurement point density distribution plan according to the adjustment requirements, and obtaining real-time updated measurement point layout data; Step S35, extracting regional targeted features from the real-time layout data to determine the temperature change trend; Step S36, optimizing the acquisition instructions through change trend analysis to obtain the sampling frequency adjustment data for the next cycle; Step S37, using a decision tree algorithm to process the adjusted data and determine the final layout of the measurement point density distribution plan; In the initial stage, the system is based on a 10m x 10m grid, covering the entire monitoring area. Then, the system calculates the temperature changes in each grid by real-time analysis of the historical data of the temperature sensor. When the temperature change in a grid exceeds 5°C, the system automatically subdivides the grid into smaller sub-grids to more accurately capture temperature fluctuations. In the refined grid, the system generates a high-frequency acquisition instruction once per second. At the same time, the system merges adjacent high-change rate grids through the K-means clustering algorithm to form a reasonable measurement point layout. When the temperature change of two adjacent grids exceeds 5°C, the system merges them into a larger measurement point area. Finally, the system outputs the real-time adjusted measurement point layout data, including measurement point coordinates and acquisition frequency information.

5. The method according to claim 1, wherein Step S4 performs data collection based on the real-time adjusted measurement point layout data, doubles the number of sampling points for a local high-temperature area where the temperature exceeds a first temperature threshold, and triggers a high-temperature micro-change monitoring mechanism to record micro-changes once per second if the temperature exceeds a second temperature threshold, thereby obtaining high-temperature micro-change characteristic data, including: Step S41, acquiring initial data through a pre-established measurement point layout, performing an adjustment to double the number of sampling points in a local high temperature area, and obtaining an expanded data acquisition result; Step S42: extracting the temperature value from the expanded data collection result. If the temperature value exceeds the second temperature threshold, activating the micro-change monitoring mechanism to obtain micro-change data recorded every second. Step S43, using a support vector machine algorithm to classify the slight change data and determine the high temperature slight change characteristic data; Step S44, calculating the change trend through the high temperature slight change characteristic data to obtain the trend direction and amplitude information; Step S45, adjusting the distribution of sampling points according to the trend direction and amplitude information to obtain optimized measurement point layout data; Step S46, re-collecting data from the optimized measurement point layout data, determining whether there are still areas with excessive temperatures, and obtaining an updated high-temperature area distribution; Step S47 : Repeat the above-mentioned subtle change monitoring and feature extraction for the updated high-temperature area distribution to obtain final high-temperature subtle change feature data.

6. The method according to claim 1, characterized in that Step S6, reconfiguring computing resources based on the optimized measurement point density distribution, and if the computing resource usage is lower than the resource threshold, reducing the measurement point density in the stable area where the temperature change rate is lower than the second change rate threshold by 30%, thereby obtaining a resource-optimized acquisition plan, includes: Step S61, obtaining computing resource occupancy data through the optimized measurement point density distribution, and determining whether the resource occupancy is lower than the resource threshold; Step S62: if the resource occupancy is lower than the resource threshold, extracting temperature change rate data from the measurement point density distribution, and using a temperature change rate analysis tool to determine a stable region where the temperature change rate is lower than a second change rate threshold; Step S63: using a density adjustment tool to reduce the measurement point density by 30% for the measurement point density data in the stable area, to obtain an adjusted density distribution; Step S64: Using the adjusted density distribution, recalculate resource usage using a resource monitoring tool to determine whether resource optimization requirements are met. Step S65: Generate a new acquisition plan using a rule engine based on the optimized resource status to obtain the measurement point configuration after resource optimization; Step S66: extract the temperature change rate and resource occupancy data from the new acquisition plan, use the optimization judgment tool to determine whether further adjustments are still needed, and obtain the final acquisition plan; Step S67: According to the final acquisition plan, use the density adjustment tool to adjust the density distribution of the measurement points and determine the operating parameters after resource optimization.

7. The method according to claim 1, characterized in that The step S8, which updates the sensor array acquisition frequency based on the final measurement point interval adjustment data, collaboratively optimizes the algorithm parameters based on the temperature change rate and local high temperature characteristics, so that the temperature evaluation error is less than 1°C, and obtains stable high temperature evaluation data, includes: Step S81, obtaining initial acquisition frequency data through the sensor array, adjusting the acquisition frequency according to the measurement point interval, and obtaining a preliminary frequency distribution; Step S82, extracting temperature data from the preliminary frequency distribution, calculating the temperature change rate using a differential method, extracting a change trend from the temperature change rate, and determining a basis for dynamic adjustment; Step S83: If the change trend exceeds the change threshold, the algorithm parameters are adjusted by the gradient descent method to obtain an optimized parameter set; Step S84, using the optimized parameter set to process the local high temperature data, extracting the high temperature distribution from the local high temperature features, and obtaining feature enhanced data; Step S85, updating the acquisition frequency through feature enhancement data, recalculating the change value based on the temperature change rate, and determining whether the evaluation error is reduced; Step S86: If the evaluation error does not meet the standard, the high temperature feature is processed by the random forest algorithm to obtain stable high temperature data; Step S87: adjusting the operating frequency of the sensor array according to the stable high temperature data, and generating final high temperature evaluation data from the adjusted frequency.

Images