Insulation test method and system for wiring cable of intelligent box-type substation

Abstract

The invention relates to the technical field of cable testing, in particular to an insulation testing method and system for a wiring cable of an intelligent box-type substation, and the method comprises the steps: connecting a to-be-tested cable with an insulation testing system, and collecting the detection excitation current and detection excitation voltage of the to-be-tested cable at each collection moment through the insulation testing system in a testing process, calculating a cable loss factor of the detection excitation current at each acquisition moment; analyzing the dielectric loss index of the detection excitation current at each acquisition moment; determining a current calibration index of the detection excitation current at each acquisition moment; and obtaining the calibrated detection excitation current based on the current calibration index, and carrying out an insulation test on the cable in combination with the detection excitation voltage. According to the invention, errors of a measurement result caused by dielectric loss can be avoided, and the accuracy and reliability of a cable insulation test are improved.

Description

Technical Field

[0001] The present application relates to the technical field of cable testing, and particularly to an insulation testing method and system for wiring cables of intelligent box-type substations. Background Art

[0002] A box-type substation is mainly composed of high-voltage switchgear, power transformers, and low-voltage distribution devices combined in different wiring modes. Box-type substations have developed rapidly in the process of the development of power supply and distribution towards miniaturization and intelligence due to advantages such as small volume and flexible installation and layout. Along with the application of new technologies and new processes, the intelligence level of box-type substations is getting higher and higher. Intelligent box-type substations meet the actual needs of unattended operation and dispatching automation, have stronger independent working capabilities for intelligent control units, and can intelligently overhaul the states of primary and secondary equipment of the entire power system, reducing the workload of equipment inspection and maintenance costs.

[0003] To ensure the stable operation of the power system, the clear and error-free transmission of signals in the communication system, etc., and to ensure that the cables in the intelligent box-type substation can maintain good performance during long-term operation, it is necessary to regularly test the insulation performance of the cables in the intelligent box-type substation to prevent potential local problems from escalating into systematic risks.

[0004] Traditional methods for testing the insulation performance of cables, such as the high-frequency current method, are non-destructive testing methods that do not reduce the insulation life of the cables and can detect tiny insulation defects. However, when the insulation layer of the cable is placed in an alternating electric field, a dielectric loss phenomenon will occur, and the dielectric loss phenomenon will cause a certain error in the measured current, resulting in low reliability. Therefore, when directly using the detected current to test and analyze the insulation performance of the cable, errors will occur, leading to a decrease in the accuracy of insulation measurement. Summary of the Invention

[0005] To solve the above technical problems, an insulation testing method and system for wiring cables of intelligent box-type substations are provided to solve the existing problems.

[0006] The solution of the present application to solve the technical problems is to provide an insulation testing method and system for wiring cables of intelligent box-type substations, including the following steps:

[0007] In a first aspect, an embodiment of the present application provides an insulation testing method for wiring cables of intelligent box-type substations, the method including the following steps:

[0008] Connect the cable to be tested to the insulation testing system, and collect the detected excitation current and detected excitation voltage at each acquisition moment during the testing process of the cable under test through the insulation testing system;

[0009] Calculate the cable loss factor of the detection excitation current at each acquisition moment according to the local difference situation of the detection excitation current and the change intensity of the detection excitation current at each acquisition moment;

[0010] Calculate the dielectric loss index of the detection excitation current at each acquisition moment according to the change characteristics of the local loss degree of the cable loss factor of the detection excitation current at each acquisition moment;

[0011] Calculate the current calibration index of the detection excitation current at each acquisition moment according to the local fluctuation degree and trend change situation of the dielectric loss index of the detection excitation current at each acquisition moment;

[0012] Obtain the calibrated detection excitation current based on the current calibration index, and perform an insulation test on the cable in combination with the detection excitation voltage.

[0013] Preferably, calculating the cable loss factor of the detection excitation current at each acquisition moment includes:

[0014] Construct a local current sequence, a current difference sequence, and a current sign sequence of the detection excitation current at each acquisition moment respectively;

[0015] For any symbol term in the current sign sequence, count the number of symbol terms that are the same as each symbol term among all the symbol terms before each symbol term, and use the ratio of the number of symbol terms to the total number of symbol terms in the current sign sequence as the first weight of each symbol term;

[0016] Calculate the Sigmoid function value of each element in the current difference sequence, and use the reciprocal of the function value as the first ratio of each element;

[0017] Based on the first ratio of each element in the current difference sequence and the first weight of the symbol term of each element, determine the cable loss coefficient of each element and the cable loss factor of the detection excitation current at each acquisition moment in turn.

[0018] Preferably, the method for determining the cable loss coefficient of each element and the cable loss factor of the detection excitation current at each acquisition moment is:

[0019] The cable loss coefficient of each element in the current difference sequence consists of the first ratio of each element and the first weight of the symbol term of each element; wherein, the cable loss coefficient is positively correlated with the first ratio and the first weight respectively;

[0020] The cable loss factor of the detection excitation current at each acquisition moment consists of the cable loss coefficients of all elements in the current difference sequence; wherein, the cable loss factor is positively correlated with the cable loss coefficient.

[0021] Preferably, the construction methods for the local current sequence, current difference sequence, and current sign sequence of the detection excitation current at each acquisition moment are as follows:

[0022] Construct the local current sequence of the detection excitation current at each acquisition moment according to the detection excitation currents at the previous K acquisition moments, where K is a preset number;

[0023] Denote the first-order difference sequence of the local current sequence of the detection excitation current at each acquisition moment as the current difference sequence of the detection excitation current at each acquisition moment;

[0024] Mark the sign terms of all elements greater than 0, less than 0, and equal to 0 in the current difference sequence as 1, -1, and 0 respectively;

[0025] Denote the sequence composed of the sign terms of all elements in the current difference sequence as the current sign sequence of the detection excitation current at each acquisition moment.

[0026] Preferably, calculating the dielectric loss index of the detection excitation current at each acquisition moment includes:

[0027] Use the sequence of cable loss factors of the detection excitation current at a preset number of acquisition moments before each acquisition moment arranged in chronological order as the first characteristic sequence of the detection excitation current at each acquisition moment;

[0028] Determine the upward trend value of the first characteristic sequence based on the fitting result of the cable loss factors within the first characteristic sequence;

[0029] The dielectric loss index of the detection excitation current at each acquisition moment consists of two parts: the Hurst exponent of the first characteristic sequence and the upward trend value. Among them, the dielectric loss index is positively correlated with the Hurst exponent and the upward trend value respectively.

[0030] Preferably, the method for determining the upward trend value is:

[0031] Perform data fitting on the cable loss factors within the first characteristic sequence with time as the abscissa and the cable loss factor as the ordinate to obtain a fitting curve, and take the average value of the slopes of all the cable loss factors on the fitting curve as the upward trend value of the first characteristic sequence.

[0032] Preferably, calculating the current calibration index of the detection excitation current at each acquisition moment includes:

[0033] Arrange the dielectric loss indices of the detection excitation current at a preset number of acquisition moments before each acquisition moment in chronological order to obtain the second characteristic sequence of the detection excitation current at each acquisition moment;

[0034] Perform a stationarity test on the second feature sequence, and record the reciprocal value of the obtained test parameter as the test factor;

[0035] Based on the degree of element fluctuation in the second feature sequence, weight the dielectric loss index in the second feature sequence to obtain the fluctuation factor of the detected excitation current at each acquisition moment;

[0036] The current calibration index of the detected excitation current at each acquisition moment consists of two parts: the fluctuation factor and the test factor. Among them, the current calibration index is positively correlated with the fluctuation factor and the test factor respectively.

[0037] Preferably, the method for obtaining the fluctuation factor of the detected excitation current at each acquisition moment is as follows:

[0038] Obtain the first-order difference sequence of the second feature sequence, use the absolute value of each element in the first-order difference sequence as the weight to linearly weight the dielectric loss indices in the second feature sequence except the first one, and use the sum of the weighted results as the fluctuation factor of the detected excitation current at each acquisition moment.

[0039] Preferably, the method for obtaining the calibrated detected excitation current and combining it with the detected excitation voltage to complete the insulation test of the cable is as follows:

[0040] For the detected excitation current at each acquisition moment and the current calibration index of the detected excitation current at the corresponding moment, use the trained neural network model to obtain the calibrated detected excitation current at each acquisition moment;

[0041] Calculate the resistance at each acquisition moment by the ratio of the detected excitation voltage at each acquisition moment to the calibrated detected excitation current at the corresponding acquisition moment, and perform an insulation test on the cable.

[0042] In a second aspect, an embodiment of the present application further provides an insulation test system for the wiring cable of an intelligent box-type substation, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the insulation test method for the wiring cable of the intelligent box-type substation described in any one of the above are implemented.

[0043] The present application has at least the following beneficial effects:

[0044] This application constructs a cable loss factor by analyzing the dielectric loss characteristics of a cable. The beneficial effect is that it can reflect the degree of dielectric loss in the cable, facilitating more accurate identification of the cable's state characteristics in subsequent steps; based on the cable loss factor, it analyzes the growth characteristics of dielectric loss and constructs a dielectric loss index. The beneficial effect is that it can reflect the growth degree of dielectric loss in the cable, taking into account the changing trend of the test results of the cable insulation performance, and then evaluating the validity of the measurement data; based on the dielectric loss index, it analyzes the fluctuation degree of dielectric loss to construct a current calibration index. The beneficial effect is that it can reflect the degree of correction of the detection excitation current, taking into account the state stability of the cable during insulation detection, and then evaluating the error degree of the measurement result for calibration; based on the current calibration index, it calibrates the detection excitation current, and uses the calibrated detection excitation current to test the insulation performance of the cable. The beneficial effect is to avoid errors in the measurement results caused by dielectric loss and improve the accuracy and reliability of cable insulation testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The following further elaborates in detail on the insulation testing method for the wiring cable of an intelligent box-type substation according to this application with reference to the accompanying drawings.

[0046] Figure 1 It is a flowchart of the steps of the insulation testing method for the wiring cable of an intelligent box-type substation provided by this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] In order to make the objectives, technical solutions, and advantages of this application clearer, the following further elaborates in detail on the insulation testing method and system for the wiring cable of an intelligent box-type substation proposed by this application with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain this application and are not intended to limit this application.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs.

[0049] Please refer to Figure 1 , which shows a flowchart of the steps of the insulation testing method for the wiring cable of an intelligent box-type substation provided by an embodiment of this application. The method includes the following steps:

[0050] Step S001, collect the detection excitation current and detection excitation voltage at each acquisition moment during the test of the cable under test through an insulation testing system.

[0051] The main function of the intelligent box-type substation in the power system is to achieve the transmission and distribution of electric energy, providing stable electric energy for multiple power consumption nodes. As the medium for electric energy transmission, the wiring cable can transmit and distribute the electric energy output by the intelligent box-type substation to each power consumption node. Therefore, the performance of the wiring cable directly affects the reasonable distribution and use of electric energy. Testing the insulation performance of the wiring cable is a necessary test to ensure that there are no safety hazards such as leakage and short circuit during the process of transmitting electric energy.

[0052] In this application, a high-frequency current sensor (HFCT) and a high-frequency voltage sensor are clamped around the grounding wire of the cable used to connect the distribution equipment in the intelligent box-type substation to ensure good contact, so as to capture the current information generated by the partial discharge of the cable metal core. Furthermore, during the process of testing the insulation performance of the cable, the current and voltage data are respectively collected through the high-frequency current sensor and the high-frequency voltage sensor. The collected current and voltage data are recorded as the high-frequency detection excitation current and the high-frequency detection excitation voltage, where the frequencies of the high-frequency detection excitation voltage and the high-frequency detection excitation current are the same, and the acquisition duration is T, and the acquisition time interval is t. During the process of collecting data, due to factors such as environmental interference, there may be abnormal situations such as missing values in the collected high-frequency detection excitation current. Therefore, the missing value filling method is used to fill and process the missing data.

[0053] Preferably, as an embodiment of this application, the high-frequency detection excitation current and the high-frequency detection excitation voltage with a collection frequency of 5 kHz are collected, the collection duration is 60 s, and the collection time interval is 0.5 s; the missing values in the collected high-frequency detection excitation current and the high-frequency detection excitation voltage are filled and processed by the regression filling method. Among them, the regression filling method is a well-known technology, and the specific process will not be elaborated in this application.

[0054] It should be understood that in an embodiment of this application, this embodiment only provides a missing value filling method, that is, the regression filling method. As other implementation manners, the implementer can adopt other missing value filling methods in the prior art, such as the nearest neighbor filling, random forest filling, etc. This application does not make special restrictions on this.

[0055] Thus, the detection excitation current and the detection excitation voltage at each acquisition moment are obtained.

[0056] Step S002, calculate the cable loss factor of the detection excitation current at each acquisition moment according to the local difference situation of the detection excitation current at each acquisition moment and the change intensity of the detection excitation current.

[0057] Due to the existence of a certain dielectric constant in the insulation layer of the cable, the insulation layer of the cable can be regarded as a capacitor. When a high-frequency detection excitation current is applied to the metal core in the cable, the interaction between the polarization process of the insulating material and the alternating electric field will cause dielectric loss. That is, in an alternating electric field, the electric dipoles inside the cable will attempt to rearrange following the change of the electric field. This process is not completely elastic and will generate a certain amount of heat, resulting in energy loss. It is mainly manifested as heat loss and conductance loss, which further causes the detection excitation current to become smaller.

[0058] Based on the above analysis, a cable loss factor is constructed here to reflect the degree of dielectric loss of the cable. The specific method is as follows:

[0059] According to the detection excitation currents at the K acquisition moments before each acquisition moment, a local current sequence of the detection excitation current at each acquisition moment is constructed. Among them, if the number of all detection excitation currents before a certain acquisition moment is less than K, the missing value filling method is used to fill it completely so that the length of the local current sequence is K, where K is a preset number;

[0060] Preferably, as an embodiment of the present application, the detection excitation currents at 50 acquisition moments before each acquisition moment are used to form a local current sequence of the detection excitation current at each acquisition moment; for those with a length of the local current sequence less than K, the mean filling method is used to fill it completely.

[0061] Furthermore, the first-order difference sequence of the local current sequence of the detection excitation current at each acquisition moment is denoted as the current difference sequence of the detection excitation current at each acquisition moment;

[0062] And the sign terms of all elements greater than 0, less than 0, and equal to 0 in the current difference sequence are respectively marked as 1, -1, and 0. The sequence composed of the sign terms of all elements in the current difference sequence of the detection excitation current at each acquisition moment is denoted as the current sign sequence of the detection excitation current at each acquisition moment;

[0063] For any sign term in the current sign sequence, taking the u-th sign term as an example, the number of elements in the first u - 1 sign terms that are the same as the u-th sign term is counted, and the ratio of the number of elements to the total number of elements in the current sign sequence is used as the first weight of the u-th sign term;

[0064] Secondly, the u-th element in the current difference sequence is input into the Sigmoid function, and the reciprocal of the corresponding function output result is used as the first ratio of the u-th element;

[0065] Here, the cable loss coefficient of each element is determined by using the first ratio of each element in the current difference sequence and the first weight of the sign value of each element. Then, the cable loss factor corresponding to the detected excitation current of the current difference sequence is determined by the cable loss coefficients of all elements.

[0066] Among them, the cable loss coefficients are positively correlated with the first ratio and the first weight respectively; the cable loss factor is positively correlated with the cable loss coefficient.

[0067] It should be understood that the positive correlation means that the dependent variable increases as the independent variable increases and decreases as the independent variable decreases. The specific relationship can be a multiplicative relationship, an additive relationship, the power of an exponential function, etc., which is determined by the actual application and is not specifically limited in this application.

[0068] Preferably, as an embodiment of this application, the product of the first ratio of each element in the current difference sequence and the first weight of each element's sign term is used as the cable loss coefficient of each element; the sum of the cable loss coefficients of all elements in the current difference sequence is used as the cable loss factor of the detected excitation current.

[0069] It should be noted that the more obvious the downward trend of the high-frequency detection excitation current passing through the cable and the greater the degree of decrease, the greater the cable loss coefficient, indicating that the dielectric loss degree of the cable is greater, so the obtained cable loss factor is greater. If the dielectric loss degree of the cable is relatively large, when the cable is used in a distribution device, the thermal effect of the detection excitation current is more obvious, the energy loss is greater, and the impact on the accuracy of cable insulation testing is greater.

[0070] Thus, the cable loss factors of the detected excitation current at each acquisition moment are obtained.

[0071] Step S003, calculate the dielectric loss index of the detected excitation current at each acquisition moment according to the change characteristics of the local loss degree of the cable loss factor of the detected excitation current at each acquisition moment.

[0072] During the operation of the intelligent box-type substation, with the long-term use of the cable, due to factors such as the thermal effect, oxidation, and mechanical stress of the detection excitation current, the cable may age, which may further cause changes in the microscopic structure of the cable, reduce the insulation effect, increase the capacitance of the insulating layer in the cable, and further increase the dielectric loss. When the dielectric loss degree in the cable is relatively large, the impedance effect on the detection excitation current in the cable will become stronger, resulting in an increase in the thermal effect of the detection excitation current. At the same time, combined with the law of conservation of energy, it can be obtained that the attenuation degree of the actually transmitted detection excitation current is greater, and the cable loss factor has a stronger growth trend.

[0073] Based on the above analysis, the dielectric loss index is calculated to reflect the growth degree of dielectric loss in the cable. The specific method is as follows:

[0074] A1: For any acquisition moment, obtain the cable loss factors of the detection excitation current at the K acquisition moments before each acquisition moment. Arrange the K cable loss factors in chronological order to obtain the first characteristic sequence of the detection excitation current at each acquisition moment; and calculate the Hurst index of the first characteristic sequence to characterize the stability of the long-term trend of the first characteristic sequence in the time dimension. Among them, the Hurst index is a commonly used technology in the field of time series processing, and the specific process will not be elaborated;

[0075] A2: Take the time as the abscissa and the cable loss factor as the ordinate for the cable loss factors within the first characteristic sequence for data fitting to obtain a fitting curve. Take the mean value of the slopes of all the cable loss factors on the fitting curve as the upward trend value of the first characteristic sequence;

[0076] A3: Use the Hurst index and the upward trend value of the first characteristic sequence to form the dielectric loss index of the first characteristic sequence. The dielectric loss index is positively correlated with the Hurst index and the upward trend value respectively. Preferably, as an embodiment, the product of the Hurst index and the upward trend value is used as the dielectric loss index.

[0077] It should be noted that data fitting is a commonly used technology in the field of data processing, and the specific fitting process will not be elaborated. Commonly used data fitting methods include but are not limited to least squares fitting, polynomial fitting, and function fitting. Preferably, as an embodiment, a polynomial fitting is used to obtain the fitting curve.

[0078] It should be noted that in the first characteristic sequence, the larger the value of the cable loss factor and the more obvious the upward trend of the cable loss factor, it indicates that the degree of dielectric loss in the cable is greater and has a gradually increasing trend. At this time, the hindering effect of the cable on the detection excitation current is stronger, which may cause certain fluctuations in the test results of the cable insulation performance, that is, the accuracy and reliability of the cable insulation test results are relatively low. Therefore, the larger the dielectric loss index.

[0079] Thus, the dielectric loss index of the detection excitation current at each acquisition moment is obtained.

[0080] Step S004: Calculate the current calibration index of the detection excitation current at each acquisition moment according to the local fluctuation degree and trend change of the dielectric loss index of the detection excitation current at each acquisition moment.

[0081] The dielectric loss index obtained through the above steps reflects the growth degree of the dielectric loss in the cable at each acquisition moment. If the fluctuation degree of the dielectric loss is large, it indicates that the cable state may undergo significant changes at this time. The specific significant changes are divided into environmental changes (such as humidity, temperature, electromagnetic interference) and significant changes in the internal structure of the cable. To obtain more accurate cable insulation test results, it is necessary to calibrate the detection excitation current to a greater extent. Based on the above analysis, calculate the current calibration index to reflect the calibration degree of the detection excitation current at each acquisition moment. The specific method is as follows:

[0082] First, calculate the dielectric loss index of the detection excitation current at all acquisition moments respectively, obtain the dielectric loss index of the detection excitation current at the previous K acquisition moments before each acquisition moment, and arrange the K dielectric loss indexes in chronological order to obtain the second characteristic sequence of the detection excitation current at each acquisition moment;

[0083] Secondly, perform a stationarity test on the second characteristic sequence, and record the reciprocal of the obtained test parameter p as the test factor. It should be noted that when the p value is 0, let the test factor be 1;

[0084] After that, obtain the first-order difference sequence of the second characteristic sequence. The element value size in the first-order difference sequence reflects the data fluctuation size occurring at each dielectric loss index in the second characteristic sequence. Use the absolute value of each element in the first-order difference sequence as the weight to linearly weight the dielectric loss indexes other than the first dielectric loss index in the second characteristic sequence, and take the sum of the weighted results as the fluctuation factor of the detection excitation current at each acquisition moment;

[0085] Subsequently, use the fluctuation factor of the detection excitation current at each acquisition moment and the test factor to determine the current calibration index of the detection excitation current at each acquisition moment. Among them, the current calibration indexes of the detection excitation currents at each acquisition moment are positively correlated with the calibration factor and the fluctuation factor respectively.

[0086] Preferably, as an embodiment of the present application, the product between the calibration factor and the fluctuation factor is used as the current calibration index of the detection excitation current at each acquisition moment.

[0087] It should be noted that the larger the current calibration index, the greater the fluctuation degree of the dielectric loss index in the second characteristic sequence, the more unstable the cable state during cable insulation detection, the more likely the measurement result is to have a large error, and the more calibration is required to improve the accuracy and reliability of the insulation test result.

[0088] So far, the current calibration indexes of the detection excitation currents at each acquisition moment are obtained.

[0089] Step S005: Obtain the calibrated detection excitation current based on the current calibration index of the detection excitation current at each acquisition moment, and combine the detection excitation voltage to complete the insulation test of the cable.

[0090] Based on the current calibration index of the detection excitation current at each acquisition moment obtained in the above steps, calibrate the cable insulation test data. The specific method is as follows:

[0091] Take 70% of the detection excitation current as the training sample, take the detection excitation current of the defect-free cable under the same test conditions as the calibrated ideal value, take the calibrated ideal value as the label set of the model training set, input the detection excitation current of the training sample and the label set, and train the neural network model.

[0092] Take the detection excitation current at each acquisition moment and the current calibration index of the detection excitation current at the corresponding moment as the input of the trained neural network model, and output the calibrated detection excitation current at each acquisition moment to improve the reliability of the detection excitation current.

[0093] Preferably, as an embodiment of the present application, train the BP neural network model, use Adam as the optimization algorithm, use the mean square error function MSE as the loss function, and obtain the calibrated detection excitation current.

[0094] It should be understood that in this embodiment, the BP neural network model is trained to obtain the calibrated detection excitation current. As other implementation manners, the implementer can use other neural network model methods in the prior art to obtain the calibrated detection excitation current, and the present application does not make special restrictions on this.

[0095] Further, according to Ohm's law, calculate the resistance at each acquisition moment by the ratio of the high-frequency detection excitation voltage at each acquisition moment to the calibrated detection excitation current at the corresponding acquisition moment, and take the magnitude of the resistance as the result of the cable insulation test to reflect the insulation degree of the cable and complete the insulation test of the cable.

[0096] It should be noted that the greater the resistance, the higher the insulation degree of the cable. In the power system, when transmitting the detection excitation current between distribution equipment through this cable, the safety degree is higher; the smaller the resistance, the lower the insulation degree of the cable, and it is more necessary to arrange relevant personnel to carry out maintenance and replacement in time to avoid safety problems when carrying out the detection excitation current transmission task between distribution equipment.

[0097] Based on the same inventive concept as the above method, an embodiment of the present invention further provides an insulation test system for the wiring cable of an intelligent box-type substation, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above methods for the insulation test of the wiring cable of the intelligent box-type substation.

[0098] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0099] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. The key points of each embodiment are the differences from other embodiments.

Claims

1. An insulation test method for a wiring cable of an intelligent box-type substation, characterized in that: The method comprises the following steps: Connect the cable to be tested to the insulation test system, and collect the detection excitation current and detection excitation voltage at each collection moment during the test of the cable to be tested through the insulation test system; Calculate the cable loss factor of the detection excitation current at each acquisition moment according to the local difference of the detection excitation current at each acquisition moment and the change intensity of the detection excitation current; Calculate the dielectric loss index of the detection excitation current at each acquisition moment according to the variation characteristics of the local loss degree of the cable loss factor of the detection excitation current at each acquisition moment; Calculate the current calibration index of the detection excitation current at each acquisition moment according to the local fluctuation degree and trend change of the dielectric loss index of the detection excitation current at each acquisition moment; A calibrated detection excitation current is obtained based on the current calibration index, and an insulation test is performed on the cable in combination with the detection excitation voltage.

2. The insulation testing method for the intelligent box-type substation wiring cable according to claim 1, characterized in that: The calculation of the cable loss factor of the detection excitation current at each acquisition moment includes: The local current sequence, the current difference sequence and the current sign sequence of the detection excitation current at each acquisition moment are respectively constructed; For any symbol item in the current symbol sequence, count the number of symbol items that are the same as each symbol item in all symbol items before each symbol item, and use the ratio of the number of symbol items to the total number of symbol items in the current symbol sequence as the first weight of each symbol item; Calculating the Sigmoid function value of each element in the current difference sequence, and taking the reciprocal of the function value as the first ratio of each element; Based on the first ratio of each element in the current difference sequence, the first weight of the sign term of each element sequentially determines the cable loss coefficient of each element and the cable loss factor of the detection excitation current at each acquisition moment.

3. The insulation testing method for the intelligent box-type substation wiring cable according to claim 2, characterized in that: The method for determining the cable loss coefficient of each element and the cable loss factor of the detection excitation current at each acquisition moment is: The cable loss coefficient of each element in the current differential sequence is composed of two parts: a first ratio of each element and a first weight of a sign term of each element; wherein the cable loss coefficient is positively correlated with the first ratio and the first weight respectively; The cable loss factor of the detected excitation current at each acquisition moment is composed of the cable loss coefficients of all elements in the current differential sequence; wherein the cable loss factor is positively correlated with the cable loss coefficient.

4. The insulation testing method for the intelligent box-type substation wiring cable according to claim 2, characterized in that: The method for constructing the local current sequence, current differential sequence and current symbol sequence of the detection excitation current at each acquisition moment is: Constructing a local current sequence of the detection excitation current at each acquisition moment according to the detection excitation current at K acquisition moments before each acquisition moment, wherein K is a preset number; Recording the first-order difference sequence of the local current sequence of the detection excitation current at each acquisition moment as the current difference sequence of the detection excitation current at each acquisition moment; Mark the sign items of all elements in the current differential sequence that are greater than 0, less than 0, and equal to 0 as 1, -1, and 0, respectively; A sequence composed of the sign items of all elements in the current difference sequence is recorded as a current sign sequence of the detection excitation current at each acquisition moment.

5. The insulation testing method for the intelligent box-type substation wiring cable according to claim 1, characterized in that: The step of calculating the dielectric loss index of the detection excitation current at each acquisition moment includes: A sequence of cable loss factors of the excitation current detected at a preset number of acquisition moments before each acquisition moment, which is composed in chronological order, is used as a first characteristic sequence of the excitation current detected at each acquisition moment; Determining an upward trend value of the first characteristic sequence based on a fitting result of a cable loss factor in the first characteristic sequence; The dielectric loss index of the excitation current detected at each acquisition moment consists of two parts: the Hurst index of the first characteristic sequence and the rising trend value, wherein the dielectric loss index is positively correlated with the Hurst index and the rising trend value respectively.

6. The insulation testing method for the intelligent box-type substation wiring cable according to claim 5, characterized in that: The method for determining the rising trend value is: The cable loss factors in the first characteristic sequence are fitted with time as the horizontal axis and the cable loss factor as the vertical axis to obtain a fitting curve, and the average of the slopes of all the cable loss factors on the fitting curve is used as the rising trend value of the first characteristic sequence.

7. The insulation testing method for the intelligent box-type substation wiring cable according to claim 1, characterized in that: The step of calculating the current calibration index of the detection excitation current at each acquisition moment includes: Arrange the dielectric loss indexes of the excitation current detected at a preset number of acquisition moments before each acquisition moment in chronological order to obtain a second characteristic sequence of the excitation current detected at each acquisition moment; Performing a stationarity test on the second characteristic sequence, and recording the inverse value of the obtained test parameter as a test factor; Weighting the dielectric loss index in the second characteristic sequence based on the fluctuation degree of the elements in the second characteristic sequence to obtain a fluctuation factor of the detection excitation current at each acquisition moment; The current calibration index for detecting the excitation current at each acquisition moment is composed of two parts: the fluctuation factor and the verification factor, wherein the current calibration index is positively correlated with the fluctuation factor and the verification factor respectively.

8. The insulation testing method for the intelligent box-type substation wiring cable according to claim 7, characterized in that: The method for acquiring the fluctuation factor of the excitation current detected at each acquisition moment is: Obtain a first-order difference sequence of the second characteristic sequence, use the absolute value of each element in the first-order difference sequence as a weight to linearly weight the remaining dielectric loss indices in the second characteristic sequence except the first dielectric loss index, and use the cumulative sum of the weighted results as the fluctuation factor of the detection excitation current at each acquisition moment.

9. The insulation testing method for the intelligent box-type substation wiring cable according to claim 1, characterized in that: The calibrated detection excitation current is obtained, and combined with the detection excitation voltage, the insulation test of the cable is completed. The specific method is: According to the detection excitation current at each acquisition moment and the current calibration index of the detection excitation current at the corresponding moment, the trained neural network model is used to obtain the calibrated detection excitation current at each acquisition moment; The resistance at each acquisition moment is calculated by the ratio of the detection excitation voltage at each acquisition moment to the detection excitation current after calibration at the corresponding acquisition moment, and the insulation test of the cable is performed.

10. An insulation test system for a wiring cable of an intelligent box-type substation, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the insulation testing method for the connection cable of the intelligent box-type substation are implemented as described in any one of claims 1-9.

Images