Electric-thermal two-parameter two-dimensional explosion-proof cable joint monitoring method and system

Abstract

The invention discloses an electric-thermal two-parameter two-dimensional explosion-proof shell cable joint monitoring method and system, and belongs to the technical field of power cables. The method specifically comprises the following steps: measuring a temperature signal and an electric field signal in a space dimension and a time dimension by using a temperature and electric field integrated probe; performing data processing on the temperature signal and the electric field signal under the spatial dimension and the time dimension, comparing the processed data with a set threshold value, and outputting a judgment result; constructing a support vector machine fault model, inputting the temperature signal and the electric field signal under the spatial dimension and the time dimension into the support vector machine fault model, and outputting a supplementary judgment result; and early warning is carried out according to an abnormal judgment result and a fault judgment result, so that electric and thermal two-parameter and two-dimensional explosion-proof shell cable joint monitoring is realized. According to the invention, fault early warning under different dimensions is carried out through the collected temperature and space electric field characteristic information, and the time dimension and the space dimension complement each other, so that early fault detection is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of power cables, and more specifically, relates to a method and system for monitoring explosion-proof cable joints in two dimensions of electric and thermal parameters. Background Art

[0002] In order to adapt to the growing demand for electricity, the workload of cables has gradually increased. In addition, cables often operate in humid underground environments, which has led to an increase in the frequency of cable failures. Under my country's "dual carbon" goal, the access of large-scale clean energy has increased the uncertainty of the power grid, resulting in an increase in the randomness of electrical quantities when faults occur, which has exacerbated the difficulty of cable fault location and diagnosis. The development of cable fault location and diagnosis technology can be traced back to the last century. Early cable fault location technology mainly relied on manual inspection and testing methods, which were inefficient, susceptible to interference from human factors, and had poor accuracy. Later, various cable fault location technologies emerged, such as lightning wave method, inversion method, time domain reflection method, etc. Although these technologies have improved the accuracy and efficiency of cable fault location to a certain extent, they still have difficulties in locating complex faults. In many cases, cable health status assessment today is still carried out in the form of regular maintenance. This method cannot detect faults in real time and eliminate hidden dangers in a timely manner, and the workload is huge.

[0003] Prior art document 1 (CN118501356B) discloses an algorithmic gas detection system for underground cable rooms in substations based on multiple multi-dimensional preset data models. The disadvantage is that the arrangement of the spatial array is only to provide data at more points, and the horizontal comparison of spatial dimension data is not used as a criterion.

[0004] Prior art document 2 (CN118795263B) discloses a high-voltage power equipment detection method and system, which has the disadvantage that it relies on current and the magnetic field it generates for detection, and can only detect faults caused by sudden current changes. Summary of the invention

[0005] In order to solve the deficiencies in the prior art, the present invention provides a method and system for monitoring explosion-proof cable joints in two dimensions: electric and thermal. On the one hand, it can collect the temperature and spatial electric field characteristics of the cable joints in the explosion-proof shell in real time, including the temperature T in different time and space dimensions. nt and the electric field E nt Data can also be used to carry out fault warning in different dimensions through the collected temperature and spatial electric field characteristic information. The time and space dimensions complement each other, so that the temperature and spatial electric field characteristics can be used as the basis for judging the fault warning of the explosion-proof shell built-in cable connector. This method not only improves the efficiency of online monitoring of explosion-proof shell cable connectors, but also reduces the warning judgment error caused by subjective human judgment.

[0006] The present invention adopts the following technical solution.

[0007] The first aspect of the present invention provides an electric and thermal two-parameter dual-dimensional explosion-proof housing cable joint monitoring method, comprising:

[0008] Measure the temperature signal and electric field signal in the spatial dimension and time dimension of the cable joint;

[0009] Perform data processing on the temperature signal and electric field signal in the spatial dimension and the time dimension, compare the processed data with the set threshold value, and output the judgment result;

[0010] Construct a support vector machine fault model, input the temperature signal and electric field signal in the spatial dimension and the time dimension into the support vector machine fault model, and output the result of supplementary judgment;

[0011] Early warning is issued based on the results of the judgment and the results of the supplementary judgment, realizing the dual-dimensional monitoring of explosion-proof shell cable joints with two parameters of electricity and heat.

[0012] Preferably, the spatial dimension data includes simultaneously obtaining temperature signals and electric field signals of different cable joints, with the cable joints being used as monitoring points;

[0013] The time dimension data includes the temperature signal and electric field signal continuously collected at the same cable joint in different time series.

[0014] Preferably, data processing is performed on the temperature signal and the electric field signal in the spatial dimension and the time dimension, and the processed data is compared with a set threshold value, specifically including:

[0015] The temperature signal and the electric field signal in the time dimension are smoothed by using the time series analysis method, and the temperature signal and the electric field signal at the time t after the smoothing process are output;

[0016] Perform difference calculation on the temperature signal and electric field signal in the spatial dimension, and output the temperature difference and electric field difference between the monitoring points;

[0017] The temperature signal and electric field signal at time t after smoothing are compared with the set threshold to obtain the time dimension judgment result, and the temperature difference and electric field difference between the output monitoring points are compared with the set threshold to obtain the space dimension judgment result, and the time dimension and space dimension judgment results are output.

[0018] Preferably, the temperature signal and electric field signal in the time dimension are smoothed by using the time series analysis method, and the simple moving average SMA at time t is calculated. t As the temperature signal and electric field signal at time t after smoothing, the simple moving average SMA at time t t , expressed as follows:

[0019]

[0020] In the formula,

[0021] m is the window size of the moving average, m≤n, n represents the nth monitoring point,

[0022] t represents the time, t≥m-1,

[0023] x i Represents the electric field signal and temperature signal in the time dimension at the i-th moment.

[0024] Preferably, a difference operation is performed on the temperature signal and the electric field signal in the spatial dimension, and the temperature difference between the monitoring points is expressed by the following formula:

[0025] ΔT a-bt =|T at -T bt | (2)

[0026] In the formula,

[0027] ΔT a-bt is the temperature difference between monitoring point a and monitoring point b at time t;

[0028] T at is the temperature signal of monitoring point a at time t;

[0029] T bt is the temperature signal of monitoring point b at time t;

[0030] The electric field difference between the monitoring points is expressed as follows:

[0031] ΔE a-bt =|E at -E bt | (4)

[0032] In the formula,

[0033] ΔE a-bt is the electric field difference between monitoring point a and monitoring point b at time t;

[0034] E at is the electric field signal of monitoring point a at time t;

[0035] E bt is the electric field signal of monitoring point b at time t.

[0036] Preferably, the temperature signal and the electric field signal at time t after smoothing are compared with a set threshold to obtain a time dimension determination result, which specifically includes:

[0037] The electric field amplitude threshold in the time dimension is set to 0.1 kV / m. When the electric field amplitude of the electric field signal at time t after smoothing is greater than 0.1 kV / m, the corresponding monitoring point is judged to be abnormal, and the judgment result is output;

[0038] The temperature amplitude threshold in the time dimension is set to 70°C. When the temperature amplitude of the temperature signal at time t after smoothing is greater than 70°C, the corresponding monitoring point is judged to be abnormal and the judgment result is output.

[0039] Preferably, in the spatial dimension, the temperature difference including the same monitoring point A is always greater than the set temperature threshold, and the temperature difference not including A is always less than the set temperature threshold, and the monitoring point A is judged to be abnormal, and the judgment result is output;

[0040] The electric field difference including the same monitoring point A is always greater than the set electric field threshold, and the electric field difference excluding A is always less than the set electric field threshold. Monitoring point A is judged to be abnormal and the judgment result is output.

[0041] Preferably, N groups of historical cable temperature and electric field signals are collected as data sets during operation, including normal state data, pre-fault and fault state data, and each group of data includes temperature characteristics, electric field characteristics and a normal state mark or fault state mark label;

[0042] The data set is divided into a training set and a test set according to a set ratio, and the support vector machine fault model is imported. The support vector machine fault model is trained with the training set, and the temperature features, electric field features and corresponding labels are used to make the support vector machine fault model learn to distinguish between normal and fault states;

[0043] The model is evaluated using the test set, the kernel parameters and penalty parameters of the radial basis function are initialized, and the parameters are adjusted through cross-validation. After the optimal parameters are found, the model is retrained using the optimal parameters to obtain the support vector machine fault model.

[0044] The second aspect of the present invention provides an electric and thermal two-parameter dual-dimensional explosion-proof shell cable joint monitoring system, and the operation of the electric and thermal two-parameter dual-dimensional explosion-proof shell cable joint monitoring method comprises:

[0045] Data acquisition module: used to measure the temperature signal and electric field signal in the spatial dimension and time dimension of the cable joint;

[0046] Abnormal judgment module: used to process the temperature signal and electric field signal in the spatial dimension and time dimension, compare the processed data with the set threshold, and output the judgment result;

[0047] Support vector machine fault judgment module: used to build a support vector machine fault model, input the temperature signal and electric field signal in the spatial dimension and the time dimension into the support vector machine fault model, and output the result of supplementary judgment;

[0048] Early warning module: used to issue early warning based on the judgment results and supplementary judgment results, and realize dual-dimensional monitoring of explosion-proof cable joints with two parameters of electricity and heat.

[0049] Preferably, temperature and electric field integrated probes are arranged at different cable joints of the cable. The temperature and electric field integrated probes are integrated probes of micro temperature measurement sensors and space electric field sensing, and are used to obtain temperature and electric field signals at the same position of the cable joint in the explosion-proof shell to obtain temperature signals and electric field signals.

[0050] Compared with the prior art, the beneficial effects of the present invention include at least: the present invention provides a method and system for monitoring explosion-proof cable joints in two dimensions of electric and thermal parameters, which can collect the temperature and spatial electric field characteristics of the cable joints in the explosion-proof shell in real time, including temperature and electric field data in different time and space dimensions, and can also perform fault warnings in different dimensions through the collected temperature and spatial electric field characteristic information. The time and space dimensions complement each other, and abnormal cables are found at the same time point and under the same operating conditions as a separate criterion to make up for the deficiency of the time dimension; when the warning in the time dimension fails, it can still be guaranteed to rely on the space dimension to warn the faulty cable, so that the temperature and space electric field characteristics are used as the judgment basis for the fault warning of the explosion-proof shell built-in cable joint, and the electric field parameters are proposed, which are more suitable for monitoring insulation faults of cable joints and better monitoring of early insulation faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a schematic diagram of an overall process provided according to an embodiment of the present invention;

[0052] Figure 2 It is a schematic diagram of a system interface provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The described embodiments are only embodiments of a part of the present invention, not all embodiments. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention.

[0054] like Figure 1 As shown, embodiment 1 of the present invention provides a method for monitoring explosion-proof cable joints in two dimensions of electric and thermal parameters, and the specific steps include:

[0055] Step 1: Use the temperature and electric field integrated probe to measure the temperature signal and electric field signal in the spatial dimension and the time dimension;

[0056] The two-parameter monitoring of the cable joint temperature and the space electric field in the explosion-proof shell is achieved through a micro-temperature sensor and a space electric field sensing integrated probe, which can obtain the temperature and electric field signals at the same position of the cable joint in the explosion-proof shell in real time to obtain the temperature signal and the electric field signal.

[0057] The spatial dimension data indicates that the temperature signals and electric field signals of different cable joints are obtained simultaneously, and the cable joints are used as monitoring points; the time dimension data indicates the continuous collection of temperature signals and electric field signals at the same cable joint in different time series.

[0058] The lateral dimension data (spatial dimension) of different monitoring points includes arranging multiple temperature and electric field integrated probes at different cable joints of the same type of cable. The temperature and electric field integrated probes are integrated probes of micro temperature measurement sensors and spatial electric field sensing. The temperature signals and electric field signals of different cable joints are obtained and recorded at the same time. For the temperature value, the temperature amplitude T of the temperature signal at the nth monitoring point is recorded. n ; For the electric field value, record the effective value E of the electric field signal at the nth monitoring point n ;

[0059] The longitudinal dimension data processing (time dimension) of the same monitoring point includes continuously collecting temperature signals and electric field signals at different time series at the same cable joint and recording them;

[0060] Step 2: Process the temperature signal and electric field signal in the spatial dimension and time dimension of step 1, compare the processed data with the set threshold, and output the judgment result;

[0061] Step 2.1: Use the time series analysis method to smooth the data in the time dimension, and output the electric field signal and temperature signal of the smoothed data in the time dimension at time t;

[0062] The data of the time dimension is preprocessed by time series analysis to eliminate the short-term normal fluctuations in the data. Here, the simple moving average method is selected, which is a basic method for smoothing time series data. It eliminates short-term fluctuations in the data by calculating the average value of the data in a certain time window, thereby showing the long-term trend of the data more clearly. Suppose the data of the time dimension are x1, x2, x3..., xn, and the window size of the moving average is m, m≤n, and the simple moving average SMA of the data of the time dimension at time t (t≥m-1) is calculated. tAs the electric field signal and temperature signal of the smoothed time dimension data at time t, the calculation formula is:

[0063]

[0064] In the formula,

[0065] m is the window size of the moving average, m≤n, n represents the nth monitoring point,

[0066] t represents the time,

[0067] x i Represents the electric field signal and temperature signal in the time dimension at the i-th moment.

[0068] Step 2.2: Perform difference calculation on the data in the spatial dimension and output the temperature difference and electric field difference between the monitoring points;

[0069] In the spatial dimension, the difference between the values ​​at different monitoring points is used for discrimination, that is, the difference between the values ​​monitored by any two cable joints is calculated. The temperature difference between the monitoring points is expressed by the following formula:

[0070] ΔT a-bt =|T at -T bt | (2)

[0071] In the formula,

[0072] ΔT a-bt is the temperature difference between monitoring point a and monitoring point b at time t;

[0073] T at is the temperature signal of monitoring point a at time t;

[0074] T bt is the temperature signal of monitoring point b at time t.

[0075] Assuming that the monitoring point is n=3, the temperature differences between monitoring points 1 and 2, 2 and 3, and 1 and 3 at time t are:

[0076]

[0077] In the formula,

[0078] ΔT 1-2t , ΔT 2-3t , ΔT 1-3t are the temperature differences between monitoring point 1 and monitoring point 2, monitoring point 2 and monitoring point 3, and monitoring point 1 and monitoring point 3 at time t;

[0079] T 1t 、T 2t 、T3t are the temperatures of monitoring point 1, monitoring point 2 and monitoring point 3 at time t respectively.

[0080] The electric field difference between the monitoring points is expressed as follows:

[0081] ΔE a-bt =|E at -E bt | (4)

[0082] In the formula,

[0083] ΔE a-bt is the electric field difference between monitoring point a and monitoring point b at time t;

[0084] E at is the electric field signal of monitoring point a at time t;

[0085] E bt is the electric field signal of monitoring point b at time t.

[0086] The electric field differences between monitoring points 1 and 2, 2 and 3, and 1 and 3 at time t are:

[0087]

[0088] In the formula,

[0089] ΔE 1-2t , ΔE 2-3t , ΔE 1-3t are the electric field differences between monitoring point 1 and monitoring point 2, monitoring point 2 and monitoring point 3, and monitoring point 1 and monitoring point 3 at time t;

[0090] E 1t 、E 2t 、E 3t They are the electric fields at monitoring point 1, monitoring point 2 and monitoring point 3 at time t respectively.

[0091] If for multiple moments, t = 1, 2, 3, ... j, taking the electric field data as an example, ΔE 1-2t and ΔE 1-3t is always greater than a set threshold, but ΔE 2-3t If the value is always less than the set threshold, then point 1 may be abnormal. The threshold can be determined based on the temperature and electric field difference range when the cable is working normally.

[0092] Step 2.3: Compare the electric field signal and temperature signal of the smoothed time dimension data at time t with the set threshold to obtain the time dimension judgment result, compare the temperature difference and electric field difference between the output monitoring points with the set threshold to obtain the space dimension judgment result, and output the time dimension and space dimension judgment results.

[0093] The electric field signal and temperature signal of the smoothed time dimension data at time t are compared with the set threshold to obtain the time dimension determination result, which specifically includes:

[0094] The electric field amplitude threshold in the time dimension is set to 0.1 kV / m. When the electric field amplitude of the electric field signal at time t of the smoothed time dimension data is greater than 0.1 kV / m, the corresponding monitoring point is judged to be abnormal, and the result of the abnormal judgment is output;

[0095] In the spatial dimension, if the temperature difference including the same monitoring point A is always greater than the set temperature threshold, and the temperature difference not including A is always less than the set temperature threshold, the monitoring point A is judged to be abnormal, and the abnormal judgment result is output;

[0096] The electric field difference including the same monitoring point A is always greater than the set electric field threshold, and the electric field difference excluding A is always less than the set electric field threshold. Monitoring point A is judged to be abnormal, and the result of abnormal judgment is output.

[0097] After preprocessing using the time series analysis method, the temperature signals monitored at different monitoring points at different times are obtained. The temperature signal of the nth monitoring point at time t is T nt ; The electric field signal of the nth monitoring point at time t is E nt .

[0098] The system interface, such as Figure 2 As shown, it includes a terminal device status display unit (online status, geographic location), an electric heat monitoring result display unit, an ambient temperature monitoring unit, and an alarm unit;

[0099] Step 3: Construct a support vector machine fault model, input the temperature signal and electric field signal in the spatial dimension and time dimension obtained in step 1 into the support vector machine fault model, obtain the monitoring point fault, and output the supplementary judgment result;

[0100] Because it is easy to miss small-scale insulation defects in current monitoring, supplementary monitoring by SVM (Support Vector Machine) is used to solve the problem of missed alarms.

[0101] The specific implementation steps are as follows:

[0102] During operation, N groups of historical cable temperature and electric field strength data are collected as data sets, including normal state data, pre-fault and fault state data. Each group of data contains temperature characteristics, electric field characteristics and a normal state mark or fault state mark label;

[0103] The data set is divided into a training set and a test set according to a set ratio, and the support vector machine fault model is imported. The support vector machine fault model is trained with the training set, and the temperature features, electric field features and corresponding labels are used to make the support vector machine fault model learn to distinguish between normal and fault states;

[0104] The model is evaluated using the test set, the kernel parameters and penalty parameters of the radial basis function are initialized, and the parameters are adjusted through cross-validation. After the optimal parameters are found, the model is retrained using the optimal parameters to obtain the support vector machine fault model.

[0105] First, taking 1000 sets of data as an example, 1000 sets of cable temperature and electric field strength data are collected during operation, of which 600 sets are normal state data and 400 sets are pre-fault or fault state data. These data are organized into a data set, each of which contains two features (temperature and electric field strength) and a label (y, normal state is marked as -1, fault state is marked as 1).

[0106] Use a radial basis function RBF kernel selection toolkit, such as the Scikit-learn library in Python. Divide the dataset into a training set and a test set, for example, 80% of the data is used for training and 20% of the data is used for testing. Then, import the support vector machine fault model and train it. During the training process, you need to select the kernel parameter and penalty parameter of the radial basis function. Parameters initialize the kernel parameter of the radial basis function to 0.1 and the penalty parameter to 1.

[0107] Then, perform model evaluation and fault warning, and use the test set to evaluate the performance of the model. You can calculate indicators such as accuracy and recall. Use cross-validation to adjust parameters and find better parameter combinations. For example, use grid search cross-validation to traverse different kernel parameters and penalty parameter values. After finding the best parameters, retrain the model with the best parameters, and then perform fault warning, which can improve the accuracy and reliability of the model.

[0108] For example, space refers to different monitoring points, that is, different cable heads; each cable head will collect a temperature data and an electric field data at each time, and correspond to a state, normal or abnormal. Assuming the temperature is 70, the electric field is 0.2, and the state is normal, then a matrix [70, 0.2, -1] is formed; then there are multiple matrices to form a large number of databases, and the 70 and 0.2 are the input data, and -1 is the output data. Output -1 represents normal, and output 1 represents abnormal.

[0109] Currently, there are three ways to judge the abnormality of monitoring points. These three judgment methods require that the temperature and electric field in the time dimension meet the threshold, and the temperature and electric field in the space dimension meet the threshold. The support vector machine fault model judges that the monitoring point is the threshold. If any of the three is met, it can be considered that there is an abnormality in this monitoring point and an early warning will be issued.

[0110] Step 4: Combine the judgment result of step 2 with the supplementary judgment result of step 3 to issue an early warning, so as to realize the dual-dimensional monitoring of explosion-proof cable joints with two parameters of electricity and heat.

[0111] Based on the results of the judgment and the supplementary judgment, early warning is issued to realize the dual-dimensional monitoring of explosion-proof shell cable joints with two parameters of electricity and heat.

[0112] Compared with the prior art, the beneficial effects of the present invention include at least: the present invention provides a method and system for monitoring explosion-proof cable joints in two dimensions of electric and thermal parameters, which can collect the temperature and spatial electric field characteristics of the cable joints in the explosion-proof shell in real time, including temperature and electric field data in different time and space dimensions, and can also perform fault warnings in different dimensions through the collected temperature and spatial electric field characteristic information. The time and space dimensions complement each other, and abnormal cables are found at the same time point and under the same operating conditions as a separate criterion to make up for the deficiency of the time dimension; when the warning in the time dimension fails, it can still be guaranteed to rely on the space dimension to warn the faulty cable, so that the temperature and space electric field characteristics are used as the judgment basis for the fault warning of the explosion-proof shell built-in cable joint, and the electric field parameters are proposed, which are more suitable for monitoring insulation faults of cable joints and better monitoring of early insulation faults.

[0113] like Figure 2 As shown, embodiment 2 of the present invention provides an electric and thermal two-parameter dual-dimensional explosion-proof shell cable joint monitoring system, and operation embodiment 1 provides an electric and thermal two-parameter dual-dimensional explosion-proof shell cable joint monitoring method, including:

[0114] Data acquisition module: used to measure the temperature signal and electric field signal in the spatial dimension and time dimension of the cable joint;

[0115] Abnormal judgment module: used to process the temperature signal and electric field signal in the spatial dimension and time dimension, compare the processed data with the set threshold, and output the judgment result;

[0116] Support vector machine fault judgment module: used to build a support vector machine fault model, input the temperature signal and electric field signal in the spatial dimension and the time dimension into the support vector machine fault model, and output the result of supplementary judgment;

[0117] Early warning module: used to issue early warning based on the judgment results and supplementary judgment results, and realize dual-dimensional monitoring of explosion-proof cable joints with two parameters of electricity and heat.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A dual-dimensional electric and thermal two-parameter explosion-proof cable joint monitoring method, characterized in that: Measure the temperature signal and electric field signal in the spatial dimension and time dimension of the cable joint; Perform data processing on the temperature signal and electric field signal in the spatial dimension and the time dimension, compare the processed data with the set threshold value, and output the judgment result; Construct a support vector machine fault model, input the temperature signal and electric field signal in the spatial dimension and the time dimension into the support vector machine fault model, and output the result of supplementary judgment; Based on the results of the judgment and the supplementary judgment, early warning is issued to realize the dual-dimensional monitoring of explosion-proof shell cable joints with two parameters of electricity and heat.

2. The method for monitoring explosion-proof cable joints of an electric and thermal two-parameter dual-dimensional explosion-proof housing according to claim 1 is characterized by: The spatial dimension data includes obtaining temperature signals and electric field signals of different cable joints simultaneously, taking the cable joints as monitoring points; The time dimension data includes the temperature signal and electric field signal continuously collected at the same cable joint in different time series.

3. The method for monitoring explosion-proof cable joints of an electric and thermal two-parameter dual-dimensional explosion-proof housing according to claim 1 is characterized in that: Data processing is performed on the temperature signal and electric field signal in the spatial dimension and the time dimension, and the processed data is compared with the set threshold value, specifically including: The temperature signal and the electric field signal in the time dimension are smoothed by using the time series analysis method, and the temperature signal and the electric field signal at the time t after the smoothing process are output; Perform difference calculation on the temperature signal and electric field signal in the spatial dimension, and output the temperature difference and electric field difference between the monitoring points; The temperature signal and electric field signal at time t after smoothing are compared with the set threshold to obtain the time dimension judgment result, and the temperature difference and electric field difference between the output monitoring points are compared with the set threshold to obtain the space dimension judgment result, and the time dimension and space dimension judgment results are output.

4. The method for monitoring explosion-proof cable joints of an electric and thermal two-parameter dual-dimensional explosion-proof housing according to claim 3 is characterized by: The time series analysis method is used to smooth the temperature signal and electric field signal in the time dimension, and the simple moving average SMA at time t is t As the temperature signal and electric field signal at time t after smoothing, the simple moving average SMA at time t t , expressed as follows: In the formula, m is the window size of the moving average, m≤n, n represents the nth monitoring point, t represents the time, t≥m-1, x i Represents the electric field signal and temperature signal in the time dimension at the i-th moment.

5. The method for monitoring explosion-proof cable joints of an electric and thermal two-parameter dual-dimensional explosion-proof housing according to claim 3 is characterized by: The temperature signal and electric field signal in the spatial dimension are subjected to difference calculation, and the temperature difference between the monitoring points is expressed by the following formula: ΔT a-bt =|T at -T bt | (2) In the formula, ΔT a-bt is the temperature difference between monitoring point a and monitoring point b at time t; T at is the temperature signal of monitoring point a at time t; T bt is the temperature signal of monitoring point b at time t; The electric field difference between the monitoring points is expressed as follows: ΔE a-bt =|E at -E bt | (4) In the formula, ΔE a-bt is the electric field difference between monitoring point a and monitoring point b at time t; E at is the electric field signal of monitoring point a at time t; E bt is the electric field signal of monitoring point b at time t.

6. The method for monitoring explosion-proof cable joints of an electric and thermal two-parameter dual-dimensional explosion-proof housing according to claim 3 is characterized by: The temperature signal and electric field signal at time t after smoothing are compared with the set threshold to obtain the time dimension judgment result, which specifically includes: The electric field amplitude threshold in the time dimension is set to 0.1 kV / m. When the electric field amplitude of the electric field signal at time t after smoothing is greater than 0.1 kV / m, the corresponding monitoring point is judged to be abnormal, and the judgment result is output; The temperature amplitude threshold in the time dimension is set to 70°C. When the temperature amplitude of the temperature signal at time t after smoothing is greater than 70°C, the corresponding monitoring point is judged to be abnormal and the judgment result is output.

7. The method for monitoring explosion-proof cable joints in an electric and thermal dual-parameter dual-dimensional manner according to claim 3 is characterized by: In the spatial dimension, if the temperature difference including the same monitoring point A is always greater than the set temperature threshold, and the temperature difference not including A is always less than the set temperature threshold, the monitoring point A is judged to be abnormal, and the judgment result is output; The electric field difference including the same monitoring point A is always greater than the set electric field threshold, and the electric field difference excluding A is always less than the set electric field threshold. Monitoring point A is judged to be abnormal and the judgment result is output.

8. The method for monitoring explosion-proof cable joints in an electric and thermal dual-parameter dual-dimensional manner according to claim 1 is characterized in that: During operation, N groups of historical cable temperature and electric field signals are collected as data sets, including normal state data, pre-fault and fault state data. Each group of data contains temperature characteristics, electric field characteristics and a normal state mark or fault state mark label; The data set is divided into a training set and a test set according to a set ratio, and the support vector machine fault model is imported. The support vector machine fault model is trained with the training set, and the temperature features, electric field features and corresponding labels are used to make the support vector machine fault model learn to distinguish between normal and fault states; The model is evaluated using the test set, the kernel parameters and penalty parameters of the radial basis function are initialized, the parameters are adjusted through cross-validation, and after the optimal parameters are found, the model is retrained using the optimal parameters to obtain the support vector machine fault model.

9. An electric heating two-parameter dual-dimensional explosion-proof shell cable joint monitoring system, running an electric heating two-parameter dual-dimensional explosion-proof shell cable joint monitoring method according to any one of claims 1 to 8, characterized in that: Data acquisition module: used to measure the temperature signal and electric field signal in the spatial dimension and time dimension of the cable joint; Abnormal judgment module: used to process the temperature signal and electric field signal in the spatial dimension and time dimension, compare the processed data with the set threshold, and output the judgment result; Support vector machine fault judgment module: used to build a support vector machine fault model, input the temperature signal and electric field signal in the spatial dimension and the time dimension into the support vector machine fault model, and output the result of supplementary judgment; Early warning module: used to issue early warning based on the judgment results and supplementary judgment results, and realize dual-dimensional monitoring of explosion-proof cable joints with two parameters of electricity and heat.

10. The electric and thermal two-parameter dual-dimensional explosion-proof shell cable joint monitoring system according to claim 9, characterized in that: Temperature and electric field integrated probes are arranged at different cable joints of the cable. The temperature and electric field integrated probes are integrated probes of micro temperature measurement sensors and space electric field sensing, and are used to obtain temperature and electric field signals at the same position of the cable joint in the explosion-proof shell to obtain temperature signals and electric field signals.

Images