Cable insulation electric field monitoring and early warning method based on digital twinning

Abstract

The invention discloses a cable insulation electric field monitoring and early warning method based on digital twinning, and relates to the technical field of cable detection. The method comprises the following steps: constructing a cable mirror image digital model according to a topological structure of a to-be-monitored cable section, obtaining influence factor historical data # imgabs0 # at an ith monitoring node and capacitance value actual measurement historical data # imgabs1 # measured by an FEF sensor in the topological structure of the to-be-monitored cable section, calculating a correction coefficient # imgabs4 # of a dielectric constant according to the obtained data # imgabs2 # and # imgabs3 #, and calculating the correction coefficient # imgabs4 # of the dielectric constant according to the correction coefficient # imgabs4 # of the dielectric constant. And synchronizing influence factor real-time data # imgabs5 # at the ith monitoring node to the cable mirror image digital model, calculating a corrected capacitance correction value # imgabs7 # according to the obtained correction coefficient # imgabs6 #, and judging the aging condition of a cable insulation layer at the ith monitoring node in the topological structure of the to-be-monitored cable section according to the obtained capacitance correction value # imgabs8 #, and the deviation degree between the settlement result of the capacitance value data and the actual condition is reduced, so that the judgment accuracy of the aging condition of the cable insulation layer is improved.

Description

Technical Field

[0001] The present invention relates to the field of cable detection technology, and in particular to a cable insulation electric field monitoring and early warning method based on digital twins. Background Art

[0002] Cross-linked polyethylene (XLPE) insulated power cables have become the preferred product for power cables required for urban power grid transformation and construction due to their reasonable process structure, strong corrosion resistance, and simple installation and laying. With the increasing application of cross-linked polyethylene power cables in urban power grid transformation, the stability of their operation is directly related to the safe operation of the entire power system. Studies have shown that the formation of water trees in the cable insulation layer is an important factor that shortens the insulation life of cross-linked polyethylene cables. Therefore, it is of great significance to detect water trees in the insulation layer.

[0003] The fringe electric field dielectric method is a widely used non-destructive measurement method. It is an effective means to detect water trees in cable insulation layers. The detection principle is to link the dielectric properties of the material being tested with the corresponding physical properties and study the basic theoretical relationship between the two. It has the characteristics of non-destructive measurement. The FEF sensor is a typical application of the fringe electric field method, which is connected to the electric field sensor chip. When the insulation layer ages and produces water trees, the dielectric constant of the insulation layer material will change significantly because the dielectric constant of water is much greater than that of cross-linked polyethylene, causing the capacitance value between the FEF sensor electrodes to change. The chip converts the capacitance value into a voltage value and transmits it to the controller. The controller can accurately read the change in capacitance value, thereby quickly judging the aging of the insulation layer.

[0004] When using the edge electric field dielectric method to detect the water tree phenomenon of the cable insulation layer, when calculating the capacitance value data of the FEF sensor, it is necessary to obtain the dielectric constant at the detection point. However, when the environment around the cable at the detection point, such as temperature and humidity, changes, the dielectric constant of the air will be affected, causing a deviation between the calculated capacitance value data and the actual situation, resulting in inaccurate judgment of the aging of the cable insulation layer. To this end, we propose a cable insulation electric field monitoring and early warning method based on digital twins. Summary of the invention

[0005] The main purpose of the present invention is to provide a cable insulation electric field monitoring and early warning method based on digital twins, which can effectively solve the problems in the background technology.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is: The cable insulation electric field monitoring and early warning method based on digital twin includes: According to the topological structure of the cable segment to be monitored, a cable mirror digital model is constructed; Obtain the historical data of the influencing factors at the i-th monitoring node in the topological structure of the cable segment to be monitored And the measured historical data of capacitance value measured by FEF sensor ,in, It is represented as the kth measured value of the jth influencing factor at the i-th monitoring node; It is represented as the kth measured value of the capacitance value at the i-th monitoring node, wherein the influencing factors include at least one of the cable environment temperature, the cable environment humidity, the electric field strength, and the cable thickness; According to the acquired data and Calculating the correction factor for the dielectric constant ; Synchronize the real-time data of the influencing factors at the i-th monitoring node To the cable mirror digital model, according to the correction coefficient obtained Calculate the corrected capacitance value ; According to the obtained capacitance correction value The aging condition of the cable insulation layer at the i-th monitoring node in the topological structure of the monitored cable segment is judged.

[0007] Corrected capacitance real-time value The calculation formula is: ; In the formula, is the electrode fork index of the FEF sensor; L is the electrode length; a and b are the interval between electrodes and the width of a single electrode plate, respectively; is the dielectric constant.

[0008] Correction factor The acquisition process includes the following steps: Take the kth measured value of the influencing factor at any monitoring node q As input, the kth measured value of the corresponding capacitance value For output, a first neural network model is constructed, and the first neural network model is trained. The number of hidden layers of the model is adjusted according to the error of the training result and the actual alarm result until the accuracy is not less than the first expected value. The k-th predicted value of the capacitance value at the q-th monitoring node is output through the constructed first neural network model. , where k=1,2,...,K; K is the total amount of data sampling at the qth monitoring node; Based on the predicted value obtained Calculate the first correction factor , the calculation formula is: ; Take the kth measured value of the influencing factors at all monitoring nodes As input, the kth measured value of the corresponding capacitance value For output, a second neural network model is constructed, and the second neural network model is trained. According to the training results and the error of the actual alarm results, the number of hidden layers of the model is adjusted until the accuracy is not lower than the second expected value. The kth predicted value of the capacitance value at the i-th monitoring node is output through the constructed second neural network model. , where i=1,2,...,N; N is the total number of monitoring nodes; Based on the predicted value obtained Calculate the second correction factor , the calculation formula is: ; The first correction factor obtained and the second correction factor Perform weighted averaging to obtain the correction factor of the dielectric constant , the calculation formula is: ,in, All ∈ (0,1), and .

[0009] The calculation formulas of the first expected value and the second expected value are: ;in, represents the rth expected value; Represents the number of input samples of the rth neural network; represents the output function of the rth neural network; represents the mth output sample of the rth neural network, , where, when r=1, =K; when r=2, =N.

[0010] The first neural network model and the second neural network model are both BP neural network models. r The input layer of neurons, q r The input layer h and s of neurons r The hidden layer has three layers of neurons, among which, ; Expressed as a pair The calculation result is rounded up; h is an integer from 1 to 9.

[0011] The process of judging the aging of the cable insulation layer includes the following steps: Obtain the capacitance correction value of the i-th monitoring node in the topological structure of the cable segment to be monitored in the initial state and the aging failure state respectively and ; Obtain the capacitance correction value of the cable segment to be monitored during use ,when ≥ When it is determined to be in an aging failure state; when > ≥ Calculate the capacitance correction value when Corrected capacitance value under aging failure condition The distance between , the calculation formula is: ; Use distance The sample set is created by the numerical value of , and the mean and standard deviation in the sample set are obtained. The mean and standard deviation are used to standardize the data. The standardization formula is: , where z is the standard parameter, σ is the variance of the sample data, and μ is the mean of the sample data; Use the function to convert the standard parameter interval into Adjust to the range of [0,1] and use Function value of capacitance correction value Classification is performed, and the classification mechanism is: when When the capacitance correction value Classification is level one; when When the capacitance correction value Classification is level 2; when When the capacitance correction value The classification is divided into three levels; among them, Function The minimum and maximum values ​​of Corrected value based on capacitance The classification results determine the degree of cable insulation aging, among which, when the capacitance correction value When classified as level 1, the cable insulation layer is slightly aged; when the capacitance correction value is When classified as Level 2, the cable insulation layer is moderately aged; when the capacitance correction value is When classified as level three, the degree of aging of the cable insulation layer is severe.

[0012] The method further comprises the following steps: According to the judgment result of the aging of the cable insulation layer, the early warning strategy at the i-th monitoring node in the topological structure of the cable segment to be monitored is formulated, specifically: When the aging degree of the cable insulation layer at the i-th monitoring node is slightly aged, a first-level cable insulation warning is issued; When the aging degree of the cable insulation layer at the i-th monitoring node is moderate, a second-level cable insulation warning is issued; When the aging degree of the cable insulation layer at the i-th monitoring node is severe, a third-level cable insulation warning is issued.

[0013] The present invention has the following beneficial effects: Compared with the existing technology, the cable mirror digital model is constructed according to the topological structure of the cable segment to be monitored, and the historical data of the influencing factors at the i-th monitoring node in the topological structure of the cable segment to be monitored is obtained. And the measured historical data of capacitance value measured by FEF sensor , based on the data obtained and Calculating the correction factor for the dielectric constant , synchronize the real-time data of the influencing factors at the i-th monitoring node To the cable mirror digital model, according to the correction coefficient obtained Calculate the corrected capacitance value , based on the obtained capacitance correction value The aging condition of the cable insulation layer at the i-th monitoring node in the topological structure of the monitored cable segment is judged, and the corrected capacitance value is obtained by analyzing the dielectric constant, thereby reducing the deviation between the settlement result of the capacitance value data and the actual situation, thereby improving the accuracy of judging the aging condition of the cable insulation layer. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of the cable insulation electric field monitoring and early warning method based on digital twins of the present invention; Figure 2 A schematic diagram of the structure of a neural network model constructed in the solution of the present invention; Figure 3 This is a detection principle diagram of an interdigital FEF sensor in an embodiment of the present invention; Figure 4 Schematic diagram of a simplified model of a sensor capacitor in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The present invention will be further described below in conjunction with specific implementation methods, wherein the accompanying drawings are only used for exemplary descriptions and represent only schematic diagrams rather than actual drawings, and should not be understood as limiting the present invention. In order to better illustrate the specific implementation methods of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product.

[0016] The specific implementation process of the technical solution of the present invention includes the following steps: Step 1: Construct a cable mirror digital model based on the topological structure of the cable segment to be monitored; Step 2: Obtain the historical data of the influencing factors at the i-th monitoring node in the topological structure of the cable segment to be monitored And the measured historical data of capacitance value measured by FEF sensor In this embodiment, the FEF sensor is an interdigital FEF sensor, and the detection principle of the interdigital FEF sensor is as follows: Figure 3 As shown, the simplified model of the sensor capacitance is as follows Figure 4 As shown, It is represented as the kth measured value of the jth influencing factor at the i-th monitoring node; It is represented as the kth measured value of the capacitance value at the i-th monitoring node, wherein the influencing factors include at least one of the cable environment temperature, the cable environment humidity, the electric field strength, and the cable thickness. In this embodiment, all the above influencing factors are taken as examples for description; Step 3: Construct data sequences A1, A2, B1, and B2 based on the acquired data, where the specific expressions of each sequence are as follows: ; ; ; ; Step 4: Based on the acquired data and Calculating the correction factor for the dielectric constant The acquisition process includes the following steps: Step 41: Take the kth measured value of the influencing factor at any monitoring node q As input, the kth measured value of the corresponding capacitance value As output, as shown in the data sequences A1 and B1 in step 3, the first column of data in sequence A1 is used as input, and the first column of data in sequence B1 is used as output, respectively, to construct a first neural network model, and by training the established first neural network model, the number of hidden layers of the model is adjusted according to the error of the training result and the actual alarm result until the accuracy is not less than the first expected value, and the k-th predicted value of the capacitance value at the q-th monitoring node is output through the constructed first neural network model , where k=1,2,...,K; K is the total amount of data sampling at the qth monitoring node; Step 42: Based on the obtained prediction value Calculate the first correction factor , the calculation formula is: ; Step 43: Take the kth measured value of the influencing factor at all monitoring nodes As input, the kth measured value of the corresponding capacitance value As output, as shown in the data sequences A2 and B2 in step 3, the first column of data in sequence A2 is used as input, and the first column of data in sequence B2 is used as output, respectively, to construct a second neural network model, and by training the established second neural network model, the number of hidden layers of the model is adjusted according to the error of the training result and the actual alarm result until the accuracy is not less than the second expected value, and the kth predicted value of the capacitance value at the i-th monitoring node is output through the constructed second neural network model , where i=1,2,...,N; N is the total number of monitoring nodes; Step 44: Based on the obtained prediction value Calculate the second correction factor , the calculation formula is: ; Step 45: Obtain the first correction coefficient and the second correction factor Perform weighted averaging to obtain the correction factor of the dielectric constant , the calculation formula is: ,in, All ∈ (0,1), and ; It should be noted that the calculation formulas for the first expected value and the second expected value are: ; in, represents the rth expected value; Represents the number of input samples of the rth neural network; represents the output function of the rth neural network; represents the mth output sample of the rth neural network, , where, when r=1, =K; when r=2, =N; The first neural network model and the second neural network model are both BP neural network models, and the network model has p r The input layer of neurons, q r The input layer of neurons h and s r The hidden layer has three layers of neurons, among which, ; Expressed as a pair The calculation result is rounded up; h is an integer from 1 to 9. In this embodiment, the number of neurons in the input layer of the first neural network model and the second neural network model is 4, and the number of neurons in the output layer is 1. When h=3, the structures of the first neural network model and the second neural network model are both three-layer structures with an input layer of 4 neurons, an input layer of 1 neuron, and a hidden layer of 6 neurons. The specific structural form is as follows Figure 2 As shown; Step 5: Synchronize the real-time data of influencing factors at the i-th monitoring node To the cable mirror digital model, according to the correction coefficient obtained Calculate the corrected capacitance value , the calculation formula is: ; In the formula, is the electrode fork index of the FEF sensor; L is the electrode length; a and b are the interval between electrodes and the width of a single electrode plate, respectively; is the dielectric constant; Step 6: Correct the capacitance value based on the obtained value The aging condition of the cable insulation layer at the i-th monitoring node in the topological structure of the monitored cable segment is judged, and the judgment process includes the following steps: Step 61: Obtain the capacitance correction value of the i-th monitoring node in the topological structure of the cable segment to be monitored in the initial state and the aging failure state respectively. and ; Step 62: Obtain the capacitance correction value of the cable segment to be monitored during use ,when ≥ When it is determined to be in an aging failure state; Step 63: When > ≥ Calculate the capacitance correction value when Corrected capacitance value under aging failure condition The distance between , the calculation formula is: ; Step 64: Using distance The sample set is created by the numerical value of , and the mean and standard deviation in the sample set are obtained. The mean and standard deviation are used to standardize the data. The standardization formula is: , where z is the standard parameter, σ is the variance of the sample data, and μ is the mean of the sample data; Step 65: Use the function to set the standard parameter interval Adjust to the range of [0,1] and use Function value of capacitance correction value Classification is performed, and the classification mechanism is: when When the capacitance correction value Classification is level one; when When the capacitance correction value Classification is level 2; when When the capacitance correction value The classification is divided into three levels; among them, Function The minimum and maximum values ​​of Step 66: Correct the value according to the capacitance The classification results determine the degree of cable insulation aging, among which, when the capacitance correction value When classified as level 1, the cable insulation layer is slightly aged; when the capacitance correction value is When classified as Level 2, the cable insulation layer is moderately aged; when the capacitance correction value is When classified as level three, the degree of aging of the cable insulation layer is severe.

[0017] Step 7: According to the judgment result of the aging of the cable insulation layer, formulate the early warning strategy at the i-th monitoring node in the topological structure of the cable segment to be monitored, specifically: When the aging degree of the cable insulation layer at the i-th monitoring node is slightly aged, a first-level cable insulation warning is issued; When the aging degree of the cable insulation layer at the i-th monitoring node is moderate, a second-level cable insulation warning is issued; When the aging degree of the cable insulation layer at the i-th monitoring node is severe, a third-level cable insulation warning is issued.

[0018] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A cable insulation electric field monitoring and early warning method based on digital twins, characterized in that: include: According to the topological structure of the cable segment to be monitored, a cable mirror digital model is constructed; Obtain the historical data of the influencing factors at the i-th monitoring node in the topological structure of the cable segment to be monitored And the measured historical data of capacitance value measured by FEF sensor ,in, It is represented as the kth measured value of the jth influencing factor at the i-th monitoring node; It is represented as the kth measured value of the capacitance at the i-th monitoring node; According to the acquired data and Calculating the correction factor for the dielectric constant ; Synchronize the real-time data of the influencing factors at the i-th monitoring node To the cable mirror digital model, according to the correction coefficient obtained Calculate the corrected capacitance value ; According to the obtained capacitance correction value The aging condition of the cable insulation layer at the i-th monitoring node in the topological structure of the monitored cable segment is judged.

2. The cable insulation electric field monitoring and early warning method based on digital twin according to claim 1 is characterized in that: Corrected capacitance real-time value The calculation formula is: ； In the formula, is the electrode fork index of the FEF sensor; L is the electrode length; a and b are the interval between electrodes and the width of a single electrode plate, respectively; is the dielectric constant.

3. The cable insulation electric field monitoring and early warning method based on digital twin according to claim 1 is characterized in that: Correction factor The acquisition process includes the following steps: Take the kth measured value of the influencing factor at any monitoring node q As input, the kth measured value of the corresponding capacitance value For output, a first neural network model is constructed, and the first neural network model is trained. The number of hidden layers of the model is adjusted according to the error of the training result and the actual alarm result until the accuracy is not less than the first expected value. The k-th predicted value of the capacitance value at the q-th monitoring node is output through the constructed first neural network model. , where k=1,2,...,K; K is the total amount of data sampling at the qth monitoring node; Based on the predicted value obtained Calculate the first correction factor , the calculation formula is: ; Take the kth measured value of the influencing factors at all monitoring nodes As input, the kth measured value of the corresponding capacitance value For output, a second neural network model is constructed, and the second neural network model is trained. According to the training results and the error of the actual alarm results, the number of hidden layers of the model is adjusted until the accuracy is not lower than the second expected value. The kth predicted value of the capacitance value at the i-th monitoring node is output through the constructed second neural network model. , where i=1,2,...,N; N is the total number of monitoring nodes; Based on the predicted value obtained Calculate the second correction factor , the calculation formula is: ; The first correction factor obtained and the second correction factor Perform weighted averaging to obtain the correction factor of the dielectric constant , the calculation formula is: ,in, All ∈ (0,1), and .

4. The cable insulation electric field monitoring and early warning method based on digital twin according to claim 3 is characterized in that: The calculation formulas of the first expected value and the second expected value are: ;in, represents the rth expected value; Represents the number of input samples of the rth neural network; represents the output function of the rth neural network; represents the mth output sample of the rth neural network, , where, when r=1, =K; when r=2, =N.

5. The cable insulation electric field monitoring and early warning method based on digital twin according to claim 3 is characterized in that: The first neural network model and the second neural network model are both BP neural network models. r The input layer of neurons, q r The input layer of neurons h and s r The hidden layer has three layers of neurons, among which, ; Expressed as a pair The calculation result is rounded up; h is an integer from 1 to 9.

6. The cable insulation electric field monitoring and early warning method based on digital twin according to claim 1 is characterized in that: The process of judging the aging of the cable insulation layer includes the following steps: Obtain the capacitance correction value of the i-th monitoring node in the topological structure of the cable segment to be monitored in the initial state and the aging failure state respectively and ; Obtain the capacitance correction value of the cable segment to be monitored during use ,when ≥ When it is determined to be in an aging failure state; when > ≥ Calculate the capacitance correction value when Corrected capacitance value under aging failure condition The distance between , the calculation formula is: ; Use distance The sample set is created by the numerical value of , and the mean and standard deviation in the sample set are obtained. The mean and standard deviation are used to standardize the data. The standardization formula is: , where z is the standard parameter, σ is the variance of the sample data, and μ is the mean of the sample data; Use the function to convert the standard parameter interval into Adjust to the range of [0,1] and use Function value of capacitance correction value Classification is performed, and the classification mechanism is: when When the capacitance correction value Classification is level one; when When the capacitance correction value Classification is level 2; when When the capacitance correction value The classification is divided into three levels; among them, Function The minimum and maximum values ​​of Corrected value based on capacitance The classification results determine the degree of cable insulation aging, among which, when the capacitance correction value When classified as level 1, the cable insulation layer is slightly aged; when the capacitance correction value is When classified as Level 2, the cable insulation layer is moderately aged; when the capacitance correction value is When classified as level three, the degree of aging of the cable insulation layer is severe.

7. The cable insulation electric field monitoring and early warning method based on digital twin according to claim 6 is characterized in that: The method further comprises the following steps: According to the judgment result of the aging of the cable insulation layer, the early warning strategy at the i-th monitoring node in the topological structure of the cable segment to be monitored is formulated, specifically: When the aging degree of the cable insulation layer at the i-th monitoring node is slightly aged, a first-level cable insulation warning is issued; When the aging degree of the cable insulation layer at the i-th monitoring node is moderate, a second-level cable insulation warning is issued; When the aging degree of the cable insulation layer at the i-th monitoring node is severe, a third-level cable insulation warning is issued.

8. The cable insulation electric field monitoring and early warning method based on digital twin according to claim 1 is characterized in that: The influencing factors include at least one of cable environment temperature, cable environment humidity, electric field strength, and cable thickness.

Images