Power cable skin temperature abnormity monitoring and early warning method

Abstract

The invention relates to the technical field of intelligent sensing systems, in particular to a power cable skin temperature anomaly monitoring and early warning method, which comprises the following steps of: acquiring temperature data of each monitoring point of a power cable skin and a current signal at a joint in real time, analyzing a current effective value, and evaluating the similarity degree of temperature and current distribution of each monitoring point in a preset time so as to determine an associated feature value and an affected feature value, classifying the monitoring points based on the feature values, identifying the severity degree of the influence of the external heat source, and performing comprehensive analysis in combination with feature indexes. And finally, taking the comprehensive characteristic index as a smoothing factor, and predicting the temperature of the monitoring point by using an exponential smoothing algorithm so as to improve the accuracy and early warning capability of temperature monitoring.

Description

Technical Field

[0001] This application relates to the field of intelligent sensing system technology, specifically to a method for monitoring and early warning of abnormal temperature on the outer sheath of power cables. Background Technology

[0002] Power cables are critical equipment for urban power grids and clean energy access, and their safe and stable operation directly affects the entire power system. As a key connection for transmitting electrical energy, the safe operation of power cables is of paramount importance. In the safety monitoring of power cables, continuous temperature monitoring of the cable sheath is a core early warning method for preventing faults and ensuring power supply reliability. The temperature of the cable insulation layer rises during operation, and continuous overheating is a direct cause of insulation aging and eventual breakdown. Many cable faults are often preceded by abnormally high local temperatures at the potential fault locations. Therefore, by monitoring the sheath temperature, real-time assessment of the cable's operating status and overload warnings can usually be achieved.

[0003] With advancements in technology, high-precision temperature sensors are gradually replacing traditional manual temperature measurement methods. Through high-precision thermocouple sensors in intelligent sensing systems, the temperature of power cable joints and other areas prone to high temperatures can be monitored in real time, and real-time temperature predictions can be made. This technology effectively enables the monitoring and early warning of abnormal surface temperatures in power cables.

[0004] In monitoring the surface temperature of underground power cables, fixed smoothing factors cannot adapt to the dynamic temperature changes caused by load variations and external heat sources (such as industrial heat sources, geothermal heat, and sunlight). This results in a slow response to sudden temperature changes. Furthermore, the axial heat conduction of the cable causes temperature data from different monitoring points to be correlated. Conventional smoothing methods are unable to effectively capture this spatial correlation, thus ignoring important temperature gradient information. Consequently, this leads to serious deviations in the monitoring and early warning of abnormal cable surface temperatures. Summary of the Invention

[0005] In view of the above, it is necessary to provide a method for monitoring and early warning of abnormal temperature on the surface of power cables to solve the above problems.

[0006] One embodiment of this application provides a method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable, the method comprising: Collect temperature data at each monitoring point on the surface of the target power cable at various times, and acquire current signals at the power cable joints at various times. Analyze the current signal within a preset time period at each temperature sampling moment to obtain the effective value of the current; based on each temperature sampling moment and the preset time length before it, analyze the similarity between the distribution of temperature data and the effective value of current at each monitoring point, determine the associated characteristic value of each monitoring point at each temperature sampling moment, and combine the dispersion of the associated characteristic value to determine the affected characteristic value of each monitoring point at each temperature sampling moment. A preset adjustment interval is used to classify the monitoring points based on the affected feature values. According to the severity of the impact of external heat sources on the monitoring points in each category after classification, combined with the distribution characteristics of the affected feature values ​​and the temperature distribution characteristics, the comprehensive feature index of each monitoring point at each adjustment time is determined. The obtained comprehensive feature index is used as a smoothing factor, and the temperature of the corresponding monitoring point is predicted by an exponential smoothing algorithm to obtain the temperature monitoring results.

[0007] Specifically, determining the associated feature value of each monitoring point at each temperature sampling time involves determining the negative correlation mapping result between the distance features of the temperature data and the effective current value of each monitoring point at each temperature sampling time and for a preset time period prior.

[0008] Specifically, determining the affected characteristic value of each monitoring point at each temperature sampling time involves: Obtain the degree of dispersion of the associated feature values ​​of all temperature data at each monitoring point at each temperature sampling time and the preset time length prior to that time. The negative correlation mapping result of the associated feature value obtained by each monitoring point at each temperature sampling time is positively fused with the obtained dispersion to obtain the affected feature value of each monitoring point at each temperature sampling time.

[0009] Specifically, determining the comprehensive characteristic index of each monitoring point at each adjustment time involves: Based on the distribution characteristics of the affected characteristic values ​​of each category after classification, monitoring points that are not affected by external heat sources are screened out, and their comprehensive characteristic indicators are determined. The severity of the impact of external heat sources on the remaining monitoring points is analyzed. Based on the similarity of the temperature distribution of the remaining monitoring points with the monitoring points on both sides within each adjustment time and the preset time before each adjustment time, combined with the affected characteristic values, the comprehensive characteristic index of each remaining monitoring point at each adjustment time is determined.

[0010] Specifically, the screening of monitoring points unaffected by external heat sources and the determination of their comprehensive characteristic indicators are as follows: For each adjustment time, if the range of cluster centers of all categories after classification is less than the preset threshold, it is determined that all monitoring points are not affected by external heat sources, and the affected feature value after normalization by the preset multiple is used as the comprehensive feature index of the corresponding monitoring point. Otherwise, the monitoring points in the category with the smallest mean of affected characteristic values ​​are judged as monitoring points not affected by external heat sources, and the normalized affected characteristic values ​​are used as the comprehensive characteristic indicators of the corresponding monitoring points.

[0011] Specifically, determining the comprehensive characteristic index of each of the remaining monitoring points at each adjustment time involves: Based on the location distribution characteristics among the remaining monitoring points, and combined with the similarity of temperature distribution between each monitoring point and the monitoring points on both sides, the difference characteristic value of each remaining monitoring point at each adjustment time is determined; Calculate the sum of the natural number 1 and the difference feature value, and use the product of the sum and the affected feature value as the comprehensive feature index of each remaining monitoring point at each adjustment time.

[0012] Specifically, the process of determining the difference characteristic value of each of the remaining monitoring points at each adjustment time is as follows: The remaining monitoring points are those affected by external heat sources; For each monitoring point affected by an external heat source, if at most only one adjacent monitoring point is affected by the external heat source, then 0 is taken as the difference characteristic value of each monitoring point. Otherwise, threshold segmentation is performed on the associated feature values ​​of all monitoring points affected by external heat sources at each adjustment time and the preset time length before that time to obtain the optimal threshold; curve fitting is performed on the temperature data corresponding to the associated feature values ​​of each monitoring point affected by external heat sources at each adjustment time and the preset time length before that time that are less than the optimal threshold. Based on the difference between the similarity of the fitted curves of each monitoring point affected by external heat sources and all its neighboring monitoring points, the difference feature value of each monitoring point affected by external heat sources at each adjustment time is determined.

[0013] Specifically, the difference feature values ​​are: Obtain the distance feature between the fitted curve of each monitoring point affected by the external heat source and each of its adjacent monitoring points; use the negative correlation mapping result of the difference between the two distance features obtained for each monitoring point affected by the external heat source as the difference feature value.

[0014] Specifically, obtaining the temperature monitoring results means: when the predicted temperature exceeds a preset monitoring threshold, a temperature anomaly warning is issued; otherwise, monitoring continues.

[0015] The monitoring threshold is set according to the material of the outer sheath of the power cable.

[0016] This application has at least the following beneficial effects: This application provides fundamental data support by real-time acquisition of temperature data and current signals at joints at each monitoring point on the sheath of the target power cable. This lays a solid foundation for subsequent analysis, enabling the monitoring system to reflect the cable's operating status in a timely manner. Analyzing the current signals within a preset time period at each temperature sampling moment and obtaining their effective values ​​helps identify the cable's electrical characteristics under different load conditions. Changes in the effective current value are usually related to temperature changes, thus providing important electrical indications of temperature anomalies. Analyzing the similarity between the distribution of temperature data and effective current values ​​at each monitoring point within a preset time period reveals the relationship between temperature and current changes. This correlation helps to more accurately assess the cable's operating status under different conditions, thereby improving the reliability of predictions. By combining the dispersion of the correlated characteristic values, the accuracy of each... The affected characteristic values ​​of monitoring points at each adjustment time provide quantitative indicators for further analysis of cable temperature changes under the influence of external heat sources. Classifying monitoring points based on these affected characteristic values, and combining this with the severity of the impact from external heat sources, effectively identifies the most vulnerable monitoring points, allowing for priority monitoring. This classification method improves the targeting and efficiency of monitoring. Furthermore, analyzing the interference from external heat sources and axial heat conduction interference from adjacent monitoring points further enhances the reliability of temperature forecasting using intelligent sensors. Utilizing comprehensive characteristic indicators as smoothing factors at each adjustment time, and employing an exponential smoothing algorithm for temperature prediction, the prediction results can more flexibly adapt to temperature fluctuations. This method not only improves prediction accuracy but also enhances the system's responsiveness, facilitating timely early warnings. Attached Figure Description

[0017] Figure 1 A flowchart of a method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable provided in this application; Figure 2 A flowchart illustrating the process of obtaining the comprehensive feature indicators provided in this application. Detailed Implementation

[0018] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0020] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged, and some steps can also be deleted.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0022] This application proposes a method for monitoring and early warning of abnormal temperature in the sheath of power cables, applied in the field of intelligent sensing systems. (See attached document.) Figure 1 The method includes the following steps: S1: Collect temperature data at each monitoring point on the surface of the target power cable at each time, and obtain the current signal at the power cable joint at each time.

[0023] In monitoring abnormal temperatures on the outer sheath of power cables, a section of cable between two joints in an underground cable is first selected for study. Starting from one joint of the target power cable, a thermocouple wireless temperature sensor is installed at intervals of 2-5 meters. The specific interval can be adjusted according to requirements; in this embodiment, the interval is 3 meters, but the implementer can adjust it according to actual conditions. The thermocouple sampling frequency is 1Hz, which is used to effectively capture temperature data in situations where the cable may experience rapid temperature changes such as frequent load changes, short circuits, and the initial stage of overload. Simultaneously, a high-precision current sensor is connected to the joint of the power cable to acquire current data in real time during the cable's power transmission process. The sampling rate of the current sensor is set to 10kHz for current data acquisition.

[0024] Considering the severe electromagnetic interference in cable usage scenarios, wavelet threshold denoising is used to denoise the acquired temperature and current signals separately. This algorithm decomposes the signal into different frequency bands, and reconstructs the signal after thresholding the noisy high-frequency coefficients in the signal, which can effectively remove the noise interference caused by electromagnetic interference. Wavelet threshold denoising is a well-known technique, and will not be elaborated further.

[0025] S2: Analyze the current signal within a preset time period at each temperature sampling moment to obtain the effective value of the current; based on each temperature sampling moment and the preset time length before it, analyze the similarity between the distribution of temperature data and the effective value of current at each monitoring point, determine the associated characteristic value of each monitoring point at each temperature sampling moment, and combine the dispersion of the associated characteristic value to determine the affected characteristic value of each monitoring point at each temperature sampling moment.

[0026] After real-time acquisition and preprocessing of temperature data from different monitoring points and current signals at joints, considering that temperature data usually exhibits a certain trend, a quadratic exponential smoothing method is used to predict and provide feedback on the temperature conditions at different monitoring points when monitoring and issuing early warnings for abnormal temperatures on the power cable sheath. The specific steps are as follows: Taking the temperature data at each moment as the center, the effective value of the current signal within a preset time period before and after it is used as the current data corresponding to the temperature data at each moment. In this embodiment, the preset time period is 0.5s, and the implementer can adjust it according to the actual situation. The calculation of the effective value of the current signal is a well-known existing technology, and this application will not elaborate on it.

[0027] Furthermore, each monitoring point is analyzed. Under normal circumstances, the cable's overall temperature gradually changes with thermal inertia as the load current dynamically changes. When a local area is disturbed by an external heat source, the overall temperature change trend of the cable is disrupted, resulting in sudden temperature changes in local areas. Therefore, to identify the temperature change characteristics within the local area of ​​each monitoring point at different temperature sampling times, this application adjusts the smoothing factor every 60 temperature sampling times and continues to use this value after each adjustment until the next adjustment time. For each monitoring point, before each adjustment time, a sequence of temperature data from the previous 30 temperature sampling times is compiled in chronological order; this sequence is called the reference temperature sequence for each monitoring point at each adjustment time.

[0028] Taking monitoring point a as an example, the reference temperature sequence of monitoring point a at time t is obtained, and the effective current value corresponding to each element in the reference temperature sequence of monitoring point a at time t is also obtained. The sequence of all corresponding effective current values ​​arranged in chronological order is taken as the effective current sequence at time t. Further, the elements in the reference temperature sequence and the effective current sequence of monitoring point a at time t are normalized using Min-Max standardization. Specifically, the maximum and minimum temperatures and current values ​​obtained within the historical 24 hours at time t are used as preset normalization maximum and minimum values, and the temperature and current data are normalized to the range [0,1]. The distance feature between the normalized reference temperature sequence and the effective current sequence of monitoring point a at time t is calculated. In this embodiment, the distance feature between the two sequences uses DTW distance to address the problem of the cable's temperature response lagging behind the current due to thermal inertia. The reciprocal of the obtained distance feature is taken as the correlation feature value of monitoring point a at time t. It should be noted that when calculating the correlation characteristic value, a very small positive number needs to be added to the denominator to ensure that the denominator is not zero. In this embodiment, the very small positive number is 0.1. When the correlation characteristic value is larger, it indicates that the correlation between temperature and current is higher, and the temperature is mainly affected by the fluctuation caused by changes in load current. When the correlation characteristic value is smaller, it indicates that the correlation between temperature and current is smaller, and the temperature is more severely affected by external interference.

[0029] Considering that the cable sheath is locally affected by external heat sources, only local monitoring points within the cable experience a temperature surge due to the heat source. The remaining areas, less affected or unaffected by external heat sources, still maintain a high correlation with the load current. Therefore, taking monitoring point a as an example, this study obtains the dispersion of the correlation characteristic values ​​of all temperature data at time t within the reference temperature sequence for monitoring point a. In this embodiment, the dispersion of the correlated feature values ​​is calculated using the standard deviation. In other embodiments, the dispersion can also be calculated using the range, coefficient of variation, etc. The influence of the heat source on monitoring point a at time t is analyzed to obtain the affected feature values. Specifically, the negative correlation mapping result of the correlated feature values ​​obtained at monitoring point a at time t is positively fused with the obtained dispersion to obtain the affected feature value of monitoring point a at time t, denoted as... In this embodiment, forward fusion employs a multiplication calculation method.

[0030] The larger the affected characteristic value, the more significant the influence of external heat sources on the monitoring point, causing the correlation between its temperature and current to change with the influence of external heat sources. At the current moment, it is being affected by external heat source interference. Based on thermal inertia, the temperature is mainly affected by external heat source interference. The smaller the affected characteristic value, the more stable the correlation between temperature and current is, the less affected or unaffected by external heat sources. At the current moment, the temperature is mainly affected by load changes.

[0031] S3: Set an adjustment interval, classify the monitoring points based on the affected feature values, and determine the comprehensive feature index of each monitoring point at each adjustment time according to the severity of the impact of external heat sources on the monitoring points in each category after classification, combined with the distribution characteristics of the affected feature values ​​and the temperature distribution characteristics; use the obtained comprehensive feature index as a smoothing factor, and use the exponential smoothing algorithm to predict the temperature of the corresponding monitoring point to obtain the temperature monitoring result.

[0032] Clustering is performed on the affected feature values ​​of all monitoring points on the power cable at time t. In this embodiment, K-means clustering is used to divide the associated feature values ​​into clusters, with k=3. The three clusters are output after clustering. K-means clustering is a well-known technique, and its details will not be elaborated further. If the range of the cluster centers of all clusters is less than a preset threshold, it is determined that all monitoring points are not affected by external heat sources. In this embodiment, the preset threshold is set to 0.05 times the threshold value. value, This represents the maximum affected feature value of all monitoring points at time t. The affected feature value obtained for each monitoring point is normalized using Min-Max standardization to the range [0.1,1]. The affected feature value after normalization by 0.1 times is used as the comprehensive feature index of each monitoring point. Otherwise, the monitoring points in the cluster with the smallest element mean are obtained to represent the monitoring points whose elements in the reference temperature sequence are not affected by external heat sources. The normalized affected feature value is used as the comprehensive feature index of the corresponding monitoring point at time t.

[0033] Analyzing the monitoring points in the other two clusters obtained at time t, considering that underground cables are usually in a closed environment, the heat generated by the load in the cable is mainly dispersed through axial heat conduction, that is, heat is dispersed from the monitoring point with higher temperature to the areas on both sides. Therefore, when the target monitoring point is interfered with by an external heat source, the monitoring points on both sides of the affected monitoring point usually exhibit similar temperature change characteristics with a time delay under the influence of axial heat conduction. Furthermore, the closer the monitoring point is to the heat source, the more drastic the temperature change; the farther away, the less interference there is.

[0034] At time t, considering that each monitoring point is affected by an external heat source, the main manifestation is that the heat source is located on one side of the monitoring point or close to the monitoring point. Therefore, when the heat source is located on one side of the monitoring point, the heat source will also have a significant impact on the adjacent monitoring points on the same side of the heat source. However, the adjacent monitoring points on the other side may be less affected by the heat source due to the greater distance and poorer axial heat conduction. Thus, there may be one or two monitoring points that are severely affected by the heat source on the left and right sides of the target monitoring point. When the external heat source interference approaches the target monitoring point, the heat source has a severe impact on the target monitoring point, and the monitoring points on both sides of the target monitoring point may also be affected by the heat source. At this time, the axial heat conduction effect of the target monitoring point to both sides is better, resulting in both sides being affected by the external heat source interference.

[0035] If, among the remaining monitoring points obtained after clustering at time t, monitoring point a has at most one adjacent monitoring point on one side, it indicates that the monitoring point on the other side is greatly affected by the external heat source and has a temperature effect on the monitoring point through axial heat conduction. Therefore, 0 is directly used as the difference characteristic value of monitoring point a at time t.

[0036] Furthermore, the correlation feature values ​​corresponding to all elements in the reference temperature sequence at time t obtained after clustering at time t are used as input, and the optimal threshold is output using the Otsu thresholding method. The Otsu thresholding method is a well-known technique and will not be elaborated further. Taking monitoring point a as an example, in the reference temperature sequence of monitoring point a at time t, temperature data at temperature sampling times where the correlation feature value is less than the optimal threshold are obtained. The least squares method is used to perform curve fitting on the obtained temperature data. The distance feature between the curve fitted by monitoring point a and the curves fitted by the two adjacent monitoring points is calculated. In this embodiment, the Fraser distance is used for calculation. The Fraser distance is a well-known technique and will not be elaborated further. The reciprocal of the absolute value of the difference between the two distance features obtained at monitoring point a is calculated and denoted as the difference feature value of monitoring point a at time t. At this point, the denominator needs to be increased. , It is a very small positive number used to ensure that the denominator is not 0, and its value is 0.1.

[0037] When the heat source is located between monitoring point a and its adjacent monitoring point on one side, the two monitoring points are more closely affected by the external heat source, and the fitted curves overlap more. However, among the two adjacent monitoring points on the left and right sides of monitoring point a, the adjacent monitoring point on the other side is mainly affected by the axial heat conduction of monitoring point a, and the curve overlap is relatively low. When the external heat source is closer to one side of the monitoring point, the monitoring points on both sides of monitoring point a are mainly affected by temperature through axial heat conduction, and the curves on both sides are similar to the central curve. The larger the difference characteristic value, the more likely that monitoring point a is located in the central region affected by the external heat source. The temperature changes of the monitoring points on both sides are affected by the axial heat conduction of the middle monitoring point and the external heat source, resulting in a similar effect of external heat influence. The smaller the difference characteristic value, the more likely that the temperature changes of monitoring point a and its adjacent monitoring point on one side are similar due to the influence of the external heat source. That is, monitoring point a is located on one side of the external heat source, while the adjacent monitoring point on the other side is mainly affected by the axial heat conduction effect of monitoring point a. Due to its distance from the heat source, the actual effect of the external heat source is poor. The temperature region of the adjacent monitoring point on this side has a poor correlation with the target monitoring point.

[0038] For the two clusters with higher means in the clustering results at time t, a comprehensive feature index is constructed by combining the obtained affected feature values ​​and difference feature values. The specific formula is as follows: ,right Increment by 1 to ensure The value is consistently greater than 1. ;in, This represents the comprehensive characteristic index of monitoring point a at time t; This represents the difference characteristic value of monitoring point a at time t. This represents the affected characteristic value of monitoring point a at time t. When the acquired... The larger the value, the more significant the impact of external heat sources on monitoring point a. Simultaneously, it is affected by axial heat conduction within the high-heat monitoring area, indicating that the current temperature is severely affected by external interference. The comprehensive characteristic indicators corresponding to all monitoring points acquired at the current moment are normalized. In this embodiment, the global maximum P-value occurring within the past 24 hours is used as the denominator to normalize the maximum value of the comprehensive characteristic indicators for all monitoring points at the current moment. The normalized comprehensive characteristic indicators are then used as a smoothing factor for the corresponding monitoring point at the current moment, for subsequent temperature warning predictions for that monitoring point.

[0039] The flowchart for obtaining comprehensive feature indicators is as follows: Figure 2 As shown.

[0040] Based on the above steps, after obtaining the optimal smoothing factor for each temperature monitoring point at each adjustment time, the reference temperature sequence for each monitoring point at the corresponding time is used as input. Using the obtained optimal smoothing factor, temperature prediction is performed through quadratic exponential smoothing to obtain the predicted temperature data for the corresponding monitoring point. Quadratic exponential smoothing is a well-known technique, and its specific steps will not be elaborated further. When the comprehensive characteristic index P-value increases (indicating severe interference at the monitoring point and its central location), the smoothing factor increases, the weight of the current data in the quadratic exponential smoothing model increases, and its dependence on historical trends decreases. This allows the model to more sensitively follow sudden temperature changes, overcoming the lag of traditional algorithms, and thus triggering timely warnings in the early stages of fires or external high-temperature intrusion.

[0041] Furthermore, a standard temperature threshold is set, with the monitoring threshold range set to 80–90 degrees Celsius. Specific settings are made based on the outer sheath material of the power cable; for example, the temperature threshold for commonly used recoverable temperature-sensing cables is typically 85°C. The real-time temperature data acquired from each monitoring point is fed back along with the predicted temperature data. When the predicted temperature exceeds the preset monitoring threshold, a temperature anomaly warning is issued, along with the estimated time of the anomaly, thus completing the monitoring and warning of abnormal power cable sheath temperature.

[0042] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0043] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for monitoring and early warning of abnormal temperature on the outer sheath of power cables, characterized in that, The method includes the following steps: Collect temperature data at each monitoring point on the surface of the target power cable at various times, and acquire current signals at the power cable joints at various times. Analyze the current signal within a preset time period at each temperature sampling moment to obtain the effective value of the current; based on each temperature sampling moment and the preset time length before it, analyze the similarity between the distribution of temperature data and the effective value of current at each monitoring point, determine the associated characteristic value of each monitoring point at each temperature sampling moment, and combine the dispersion of the associated characteristic value to determine the affected characteristic value of each monitoring point at each temperature sampling moment. A preset adjustment interval is used to classify the monitoring points based on the affected feature values. According to the severity of the impact of external heat sources on the monitoring points in each category after classification, combined with the distribution characteristics of the affected feature values ​​and the temperature distribution characteristics, the comprehensive feature index of each monitoring point at each adjustment time is determined. The obtained comprehensive feature index is used as a smoothing factor, and the temperature of the corresponding monitoring point is predicted by an exponential smoothing algorithm to obtain the temperature monitoring results.

2. The method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable as described in claim 1, characterized in that, The determination of the correlation feature value of each monitoring point at each temperature sampling time is specifically the negative correlation mapping result of the distance feature between the temperature data and the effective current value of each monitoring point at each temperature sampling time and the preset time length prior.

3. The method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable as described in claim 1, characterized in that, The determination of the affected characteristic value of each monitoring point at each temperature sampling time is specifically as follows: Obtain the degree of dispersion of the associated feature values ​​of all temperature data at each monitoring point at each temperature sampling time and the preset time length prior to that time. The negative correlation mapping result of the associated feature value obtained by each monitoring point at each temperature sampling time is positively fused with the obtained dispersion to obtain the affected feature value of each monitoring point at each temperature sampling time.

4. The method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable as described in claim 1, characterized in that, The determination of the comprehensive characteristic index of each monitoring point at each adjustment time is specifically as follows: Based on the distribution characteristics of the affected characteristic values ​​of each category after classification, monitoring points that are not affected by external heat sources are screened out, and their comprehensive characteristic indicators are determined. The severity of the impact of external heat sources on the remaining monitoring points is analyzed. Based on the similarity of the temperature distribution of the remaining monitoring points with the monitoring points on both sides within each adjustment time and the preset time before each adjustment time, combined with the affected characteristic values, the comprehensive characteristic index of each remaining monitoring point at each adjustment time is determined.

5. The method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable as described in claim 4, characterized in that, The monitoring points that are not affected by external heat sources are selected, and their comprehensive characteristic indicators are determined, specifically as follows: For each adjustment time, if the range of cluster centers of all categories after classification is less than the preset threshold, it is determined that all monitoring points are not affected by external heat sources, and the affected feature value after normalization by the preset multiple is used as the comprehensive feature index of the corresponding monitoring point. Otherwise, the monitoring points in the category with the smallest mean of affected characteristic values ​​are judged as monitoring points not affected by external heat sources, and the normalized affected characteristic values ​​are used as the comprehensive characteristic indicators of the corresponding monitoring points.

6. The method for monitoring and early warning of abnormal temperature on the surface of a power cable as described in claim 4, characterized in that, The determination of the comprehensive characteristic index of each of the remaining monitoring points at each adjustment time is specifically as follows: Based on the location distribution characteristics among the remaining monitoring points, and combined with the similarity of temperature distribution between each monitoring point and the monitoring points on both sides, the difference characteristic value of each remaining monitoring point at each adjustment time is determined; Calculate the sum of the natural number 1 and the difference feature value, and use the product of the sum and the affected feature value as the comprehensive feature index of each remaining monitoring point at each adjustment time.

7. The method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable as described in claim 6, characterized in that, The process of determining the difference characteristic value of each of the remaining monitoring points at each adjustment time is as follows: The remaining monitoring points are those affected by external heat sources; For each monitoring point affected by an external heat source, if at most only one adjacent monitoring point is affected by the external heat source, then 0 is taken as the difference characteristic value of each monitoring point. Otherwise, threshold segmentation is performed on the associated feature values ​​of all monitoring points affected by external heat sources at each adjustment time and the preset time length before that time to obtain the optimal threshold; curve fitting is performed on the temperature data corresponding to the associated feature values ​​of each monitoring point affected by external heat sources at each adjustment time and the preset time length before that time that are less than the optimal threshold. Based on the difference between the similarity of the fitted curves of each monitoring point affected by external heat sources and all its neighboring monitoring points, the difference feature value of each monitoring point affected by external heat sources at each adjustment time is determined.

8. The method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable as described in claim 7, characterized in that, The specific difference feature values ​​are: Obtain the distance feature between the fitted curve of each monitoring point affected by the external heat source and each of its adjacent monitoring points; use the negative correlation mapping result of the difference between the two distance features obtained for each monitoring point affected by the external heat source as the difference feature value.

9. The method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable as described in claim 1, characterized in that, Specifically, the obtained temperature monitoring results are as follows: when the predicted temperature exceeds the preset monitoring threshold, a temperature anomaly warning is issued; otherwise, monitoring continues.

10. The method for monitoring and early warning of abnormal temperature on the outer sheath of a power cable as described in claim 9, characterized in that, The monitoring threshold is set according to the outer sheath material of the power cable.

Images