Insulated combined electric appliance detection method and device, electronic equipment and readable storage medium

Abstract

The application provides an insulating combined electrical appliance detection method, comprising: receiving a first input, which is discharge data of an object to be identified; in response to the first input, extracting a characteristic quantity of the discharge data; receiving a second input, which is the characteristic quantity of the discharge data; in response to the second input, inputting the characteristic quantity of the discharge data into a multi-layer perception machine, and outputting a defect identification result of the object to be identified. The application solves the problem of inaccurate identification under different defect conditions, and avoids the situation of false detection and missed detection in the discharge detection process.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, electronic device, and readable storage medium for detecting insulated combined electrical appliances. Background Technology

[0002] With the continuous development of power grids, gas-insulated switchgear (GIS), which has advantages such as small footprint, good operational safety and stability, and low electromagnetic pollution, is being used more and more widely in power grids. However, during the installation, transportation, and operation of GIS, partial discharge (PD) phenomena with different characteristics may occur due to factors such as process, collision, and operating environment. Different types of partial discharge cause different degrees of damage to GIS. The traditional ultra-high frequency (UHF) method used for GIS partial discharge detection has problems such as poor resistance to intermittent power interference and easy false detection. Optical signal detection for GIS partial discharge has problems such as many detection blind spots and easy missed detection. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method for detecting insulated combined electrical appliances to solve the problem of numerous partial discharge detection errors in gas-insulated switchgear.

[0004] According to a first aspect of the present invention, a method for testing insulating combined electrical appliances is provided, comprising:

[0005] Receive a first input, which is the discharge data of the object to be identified;

[0006] In response to the first input, feature quantities of the discharge data are extracted;

[0007] Receive a second input, which is a characteristic quantity of the discharge data;

[0008] In response to the second input, the feature values ​​of the discharge data are input into the multilayer perceptron, and the defect identification result of the object to be identified is output.

[0009] Optionally, the first input is the first discharge data of the object to be identified measured by the first measurement method and the second discharge data of the object to be identified measured by the second measurement method;

[0010] The step of inputting the feature values ​​of the discharge data into a multilayer perceptron and outputting the defect identification result of the object to be identified includes:

[0011] The feature values ​​of the first discharge data and the feature values ​​of the second discharge data are respectively input into the multilayer sensor.

[0012] The multilayer perceptron outputs the first defect identification result of the object to be identified and the second defect identification result of the object to be identified.

[0013] Optionally, after the multilayer perceptron outputs the first defect identification result of the object to be identified and the second defect identification result of the object to be identified, it further includes:

[0014] Based on the defect types of the objects to be identified and the defect data corresponding to each defect type, a DS evidence theory model is constructed.

[0015] The first defect identification result and the second defect identification result of the object to be identified are input into the DS evidence theory model, and the third defect identification result after information fusion is output.

[0016] Optionally, the step of inputting the first defect identification result and the second defect identification result of the object to be identified into the DS evidence theory model, and outputting the third defect identification result after information fusion, includes:

[0017] The DS evidence theory model receives the first defect identification result of the object to be identified and the second defect identification result of the object to be identified.

[0018] Based on the fusion rules and basic probability assignment, the first defect identification result and the second defect identification result of the object to be identified are fused together, and the third defect identification result after information fusion is output.

[0019] According to a second aspect of the present invention, an insulation combination electrical appliance testing device is provided, comprising:

[0020] The first receiving module is used to receive a first input, wherein the first input is the discharge data of the object to be identified;

[0021] The extraction module, in response to the first input, extracts the feature quantities of the discharge data;

[0022] The second receiving module is used to receive a second input, which is a characteristic quantity of the discharge data;

[0023] The first identification module, in response to the second input, inputs the feature values ​​of the discharge data into the multilayer perceptron and outputs the defect identification result of the object to be identified.

[0024] Optionally, the first identification module includes:

[0025] The third receiving module is used to input the feature values ​​of the first discharge data and the feature values ​​of the second discharge data into the multilayer sensor, respectively.

[0026] The second identification module is used to identify the first defect identification result and the second defect identification result of the object to be identified output by the multilayer perceptron.

[0027] Optionally, the insulation switchboard testing device further includes:

[0028] The construction module is used to construct a DS evidence theory model based on the defect type of the object to be identified and the defect data corresponding to each defect type.

[0029] The fusion module is used to input the first defect identification result and the second defect identification result of the object to be identified into the DS evidence theory model, and output the third defect identification result after information fusion.

[0030] Optionally, the fusion module includes:

[0031] The fourth receiving module is used to receive the first defect identification result of the object to be identified and the second defect identification result of the object to be identified.

[0032] The fusion submodule is used to assign and fuse the first defect identification result of the object to be identified and the second defect identification result of the object to be identified according to the fusion rules and basic probabilities, and output the third defect identification result after information fusion.

[0033] According to a third aspect of the present invention, an electronic device is provided, comprising:

[0034] Processor; and

[0035] Stored program memory,

[0036] The program includes instructions that, when executed by the processor, cause the processor to perform the method according to any one of the first aspects of the invention.

[0037] According to a fourth aspect of the present invention, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are configured to cause a computer to perform the method according to any one of the first aspects of the present invention.

[0038] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:

[0039] By extracting feature quantities from the discharge data of the object to be identified, and using a multilayer perceptron to identify the discharge pattern of the feature quantities, the defect type of the object to be identified can be determined. This solves the problem of inaccurate identification under different defect conditions and avoids false detection and missed detection during the discharge detection process.

[0040] The above brief description is only an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more obvious and understandable, the following examples describe in detail the specific embodiments of the present invention. Attached Figure Description

[0041] Further details, features, and advantages of the invention are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0042] Figure 1 An internal defect diagram of a GIS according to an exemplary embodiment of the present invention is shown;

[0043] Figure 2 A flowchart of a method for testing insulated combined electrical appliances according to an exemplary embodiment of the present invention is shown;

[0044] Figure 3 A schematic diagram of a multilayer perceptron according to an exemplary embodiment of the present invention is shown;

[0045] Figure 4 A DS evidence theory fusion diagram based on a multilayer perceptron according to an exemplary embodiment of the present invention is shown;

[0046] Figure 5 A schematic block diagram of an insulation combination electrical appliance detection device according to an exemplary embodiment of the present invention is shown;

[0047] Figure 6 A structural block diagram of an exemplary electronic device that can be used to implement embodiments of the present invention is shown. Detailed Implementation

[0048] Embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While some embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the invention. It should be understood that the accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the invention.

[0049] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0050] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0051] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0052] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0053] The present invention will now be described with reference to the accompanying drawings. The technical solutions provided by the embodiments of this application will be explained in detail through specific examples and application scenarios.

[0054] like Figure 1 As shown, Figure 1 This application provides an internal defect map of a GIS (Gas Insulated Switchgear). Taking the object to be identified as a gas-insulated switchgear (GIS) as an example, when defects occur inside the GIS or impurities adhere to the surrounding area, the electric field strength of that part differs from the environment, and the electric field strength gradually increases. When it reaches the breakdown voltage, a fault occurs. At this time, partial discharge will occur in that part. Partial discharge is generally also accompanied by insulation defects, and the insulation at that location will deteriorate, causing uneven electric field strength. There are many causes of partial discharge; some common causes include... Figure 1 As shown.

[0055] like Figure 2 As shown, Figure 2 This is a flowchart illustrating a method for testing insulated combined electrical appliances according to an embodiment of this application. The method may include the following steps S201 to S204:

[0056] S201, Receive the first input, where the first input is the discharge data of the object to be identified.

[0057] In one optional embodiment, the object to be identified is a gas-insulated switchgear, and the discharge data of the object to be identified is the partial discharge phase distribution map (PRPD) of the gas-insulated switchgear.

[0058] S202, in response to the first input, extract the feature quantity of the discharge data.

[0059] In an optional embodiment, when the discharge data of the object to be identified is a partial discharge phase distribution map (PRPD) of a gas-insulated switchgear, the PRPD has a skewness S. k Steepness K u Number of local peaks P e Cross-correlation coefficient C c Asymmetry Parameters such as the Weibull distribution can be used to describe the shape differences and contour distinctions between positive and negative half-cycles in two-dimensional partial discharge patterns. Using statistical feature parameters of the PRPD pattern for pattern recognition can reduce the impact of the random distribution characteristics of partial discharge over discharge time, while also reflecting the repetitive characteristics of partial discharge within the power frequency cycle.

[0060] Among them, the skewness S is selected. k Steepness K u As a characteristic quantity of the discharge data, the skewness S k This describes the degree of asymmetry in the distribution of partial discharge from internal insulation defects within each power frequency cycle. A positive skewness indicates a positively skewed distribution, while a negative skewness indicates a negatively skewed distribution. Skewness S k The calculation is as follows:

[0061]

[0062] Where, x i denoted as σ, where μ is the amplitude of the partial discharge pulse, N is the number of partial discharge pulses, and σ is the standard deviation.

[0063] Steepness K u This describes the degree of bulging in the shape of the partial discharge pattern relative to a normal distribution. For a normal distribution, K... u The typical steepness is 3. (K) u The calculation is as follows:

[0064]

[0065] Where, x i denoted as σ, where μ is the amplitude of the partial discharge pulse, N is the number of partial discharge pulses, and σ is the standard deviation.

[0066] S203, Receive the second input, the second input being a characteristic quantity of the discharge data.

[0067] In this embodiment, after feature recognition is performed, a signal indicating successful feature recognition is received, and the feature quantity identified by feature recognition is used as the input for defect recognition.

[0068] S204, in response to the second input, the feature quantity of the discharge data is input into the multilayer perceptron, and the defect identification result of the object to be identified is output.

[0069] In this embodiment, directly determining the discharge type based solely on the partial discharge phase distribution map (PRPD) is prone to misjudgment. Therefore, further analysis and research of the map are necessary. Figure 3 As shown, a multilayer perceptron (MLP) is used for pattern recognition of partial discharges. The principle of the multilayer perceptron is as follows: Figure 3 As shown.

[0070] In this embodiment, the multilayer perceptron introduces one or more hidden layers on top of a single-layer neural network. Each layer is fully connected, and the output of each hidden layer is transformed by an activation function. After calculating the weighted sum of the hidden units in each hidden layer, an activation function is applied to the result to introduce nonlinearity into the neurons, allowing the neural network to arbitrarily approximate any nonlinear function. The tanh activation function transforms the values ​​of elements to between -1 and 1, approaching -1 when the input value is small and approaching 1 when the input value is large. Adding the tanh activation function to the neural network with hidden layers transforms the hidden variables using a nonlinear function, as shown below:

[0071]

[0072] Where w is the weight between the input layer x and the hidden layer h, k is the hidden unit label of the t-th layer, and h is the intermediate result of the calculation. The final output is obtained after weighted summation. As shown below:

[0073]

[0074] Where υ is the sum of the hidden layer h and the output. The weights w and v are learned from the data, and the number of nodes in the hidden layer can be freely set according to the complexity of the dataset.

[0075] In an optional embodiment, the skewness S k and steepness K u As the input parameter for multilayer perceptron recognition, K is used for the entire power frequency cycle. u and S kAs feature vectors l1 to l2, K of the negative half-cycle of the power frequency u and S k As feature vectors l3 to l4, K of the positive half-cycle of the power frequency u and S k As feature vectors l5 to l6;

[0076] In this embodiment, the first measurement method is ultra-high frequency (UHF) and the second measurement method is optical measurement. When using UHF, the feature vector of PRPD pattern recognition is A = [l1, l2, l3, l4, l5, l6], and when using optical measurement, the feature vector of PRPD pattern recognition is B = [l1, l2, l3, l4, l5, l6].

[0077] The multilayer perceptron provided in this embodiment is a two-layer perceptron model with 6 input nodes i = 6. The input nodes X1 to X6 correspond to the l1 to l6 feature vectors. The first hidden layer has 30 nodes, the second hidden layer has 20 nodes, and the output layer has 4 nodes, corresponding to the probabilities of the four types of insulation defects of the object to be identified.

[0078] In an optional embodiment, the first input is the first discharge data of the object to be identified measured by a first measurement method and the second discharge data of the object to be identified measured by a second measurement method;

[0079] The step of inputting the feature values ​​of the discharge data into a multilayer perceptron and outputting the defect identification result of the object to be identified includes:

[0080] S2041, input the feature values ​​of the first discharge data and the feature values ​​of the second discharge data into the multilayer sensor respectively;

[0081] S2042, the multilayer perceptron outputs the first defect identification result of the object to be identified and the second defect identification result of the object to be identified.

[0082] In this embodiment, the first measurement method is the ultra-high frequency (UHF) method, and the second measurement method is the optical method. The total number of samples for both the UHF and optical methods is N1 and N2, respectively, both 480. To maintain sample balance, each defect type has 120 sets of samples. 30 sets of samples are randomly selected from the discharge samples of each defect type as test samples, and the remaining 90 sets are used as training samples. The PRPD pattern recognition results for the UHF and optical methods are shown in Table 1.

[0083] Table 1 Partial discharge pattern recognition rates using UHF and optical methods

[0084]

[0085] It can be seen that both the UHF method and the optical method have relatively high recognition rates. The average recognition rate of the UHF method is 80.25%, with high recognition rates for air gap defects and surface discharges. The average recognition rate of the optical method is 90.75%, with high recognition rates for air gap defects, suspension defects, and tip defects.

[0086] Comparison revealed that the two methods have different recognition accuracies for different defects. This is because different defect models have different discharge modes, resulting in variations in the strength of the generated optical and high-frequency electromagnetic signals. The data shows that the UHF method and the optical measurement method are complementary in pattern recognition and can be jointly identified using information fusion theory.

[0087] In an optional embodiment, after the multilayer perceptron outputs the first defect identification result and the second defect identification result of the object to be identified, the method further includes:

[0088] S2043, Construct a DS evidence theory model based on the defect type of the object to be identified and the defect data corresponding to each defect type;

[0089] S2044, input the first defect identification result and the second defect identification result of the object to be identified into the DS evidence theory model, and output the third defect identification result after information fusion.

[0090] In this embodiment, the Dempster-Shafer (DS) evidence theory is a mathematical model for reasoning under uncertainty. Based on an extension of probability theory, it addresses situations involving incomplete, uncertain, and contradictory information, and is applicable to reasoning problems across various fields. The DS evidence theory model can effectively express not only stochastic uncertainty but also incomplete and subjectively uncertain information.

[0091] The DS evidence theory model can model the propositions and related evidence that require reasoning, based on a specific problem. At this stage, it is necessary to clarify the expression of the propositions, such as binary or multivariate propositions, and to consider the relationships between each proposition.

[0092] In the DS evidence theory, the set of all possible outcomes for judging a problem is called the identification frame. In this embodiment, the identification frame consists of four typical insulation defects: air gap discharge, floating discharge, surface discharge, and pinpoint discharge. For ease of description, the identification frame Φ is represented by A1, A2, A3, and A4, with an uncertainty of θ, as shown below:

[0093] Φ={A1,A2,A3,A4,θ} (5)

[0094] UHF electromagnetic and optical signal data are treated as two independent forms of evidence. All evidence is represented using a confidence function, which reflects the degree to which a proposition is believed to be true even with incomplete information. At this stage, the confidence level, uncertainty, and other information corresponding to each piece of evidence need to be converted into mathematical formulas.

[0095] The power set of the identification frame Φ constitutes the proposition set 2 Φ , If the function m:2 Φ →[0,1] satisfies the condition of formula (6), then the function m is called basic probability assignment (BPA), and m(A) is the basic probability number of proposition A, that is, the confidence of accurately assigning to A.

[0096]

[0097] The DS evidence theory model uses the output values ​​of partial discharge data measured by optical and UHF methods after pattern recognition by a multilayer perceptron as two independent pieces of evidence, and converts them into a BPA that satisfies the DS evidence theory. Finally, all the evidence is combined into a global trust function through a synthesis rule.

[0098] In an optional embodiment, the step of inputting the first defect identification result and the second defect identification result of the object to be identified into the DS evidence theory model, and outputting the third defect identification result after information fusion, includes:

[0099] S20441, the DS evidence theory model receives the first defect identification result of the object to be identified and the second defect identification result of the object to be identified;

[0100] S20442, according to the fusion rules and basic probability assignment, fuse the first defect identification result of the object to be identified and the defect identification result of the second object to be identified, and output the third defect identification result after information fusion.

[0101] In this embodiment, commonly used synthesis rules include the Dempster synthesis rule, the Dubois-Prade synthesis rule, and others. At this stage, the relationships and consistency between multiple pieces of evidence need to be considered to determine the global trust function.

[0102] The Dempster evidence theory provides the Dempster combination rule to achieve the fusion of multiple pieces of evidence, which is essentially an orthogonal sum of evidence. In this embodiment, let m1 and m2 be the basic probability assignments of optical signals and UHF signals, respectively, and let m represent m1 and m2. The new evidence after combination is then expressed by the Dempster combination rule as Equation (7).

[0103]

[0104] In this diagram, A1 to A4 represent various discharge defect types under the UHF method, and B1 to B4 represent discharge defect types under the optical method. k is called the conflict coefficient, used to measure the degree of conflict between pieces of evidence. The larger k is, the greater the conflict. When k = 1, it leads to the "Zadeh paradox," that is, when the pieces of evidence are highly conflicting, it will produce results that contradict common sense.

[0105] like Figure 4 The diagram shown is a DS evidence theory fusion diagram based on a multilayer perceptron. The first measurement method is the ultra-high frequency (UHF) method, and the second measurement method is the optical method. The UHF signal is acquired through the first measurement method, and the optical signal is acquired through the second measurement method. Then, the features of the UHF signal and the optical signal are extracted according to the DS evidence theory and input into the multilayer perceptron. The multilayer perceptron outputs BPA optical signal and BPA UHF signal and fuses them according to the evidence combination rules. The final result is obtained by making decisions and reasoning based on the synthesized trust function.

[0106] During the decision-making stage, it is necessary to design appropriate decision-making rules based on the specific problem, and combine domain knowledge and experience to interpret and analyze the results in order to arrive at accurate and reliable reasoning results.

[0107] After calculating the BPA of all possible outcomes under the identification framework Φ with all evidence, the relevant rules are used to determine the classification of insulation defects.

[0108] Rule 1:

[0109] Rule 2: m(A max1 )>m(θ)

[0110] Rule 3: m(A max1 )-m(A max2 )>ε

[0111] Rule 1 states that the proposition with the highest confidence level should be selected as the output; Rule 2 states that the BPA of the judgment result must be greater than the BPA of the uncertainty; and in Rule 3, m(A) max1 ) represents the maximum BPA of a certain judgment result type, m(A) max2 ) is the second largest BPA, and ε is the threshold value. That is, the difference between the judgment result BPA and the BPA of other propositions should be greater than the set threshold value. Only when the above three rules are met can the system judgment result be output.

[0112] In one optional embodiment, after training the insulation defect partial discharge data obtained by the UHF method and optical measurement method with a multilayer perceptron, according to... Figure 4Information fusion is performed using the DS evidence theory. For each set of partial discharge data samples, UHF electromagnetic signal and optical signal data are used as two independent evidence bodies. After training with a multilayer perceptron, the basic probability assignment function (BPA) for each defect type is obtained.

[0113] Taking a set of photoelectric joint detection data as an example, the fused BPA of the two signal outputs is calculated by equation (7), and the results are shown in Table 2. From the data in Table 2, it can be seen that when using the UHF partial discharge PRPD spectrum as the sole evidence for pattern recognition, the BPAs for suspended defects and tip defects are 0.38 and 0.36, respectively, with an uncertainty of 0.18. The identification result is a suspended defect, but the BPAs for suspended defects and tip defects are very similar, making misjudgment easy. When using the photoelectric partial discharge PRPD spectrum as the sole evidence for pattern recognition, the BPA for suspended defects is much higher than other types, but the uncertainty is still relatively high. After fusion using the Dempster combination rule, the BPA for suspended defects reaches 0.682, and the uncertainty decreases to 0.0305, confirming the identification result as a suspended defect.

[0114] Table 2 shows the basic probability assignment effect of a set of typical data.

[0115] air gap 0.04 0.02 0.0156 Floating 0.37 0.62 0.6510 Along the surface 0.03 0.02 0.0132 tip 0.37 0.24 0.2763 θ 0.18 0.11 0.0315

[0116] Based on the DS evidence theory, photoelectric fusion detection of partial discharge of typical insulation defects in GIS was performed on all samples obtained under the two measurement methods. The final identification results are shown in Table 3.

[0117] As shown in Table 3, the accuracy of partial discharge pattern recognition for typical insulation defects inside GIS was improved after information fusion using the DS evidence theory, with the recognition rate for each defect reaching over 87.8%. Surface discharge defects still had the lowest recognition rate. The DS evidence theory effectively combined the advantages of the UHF method and the optical measurement method, improving the imbalance in recognition rates when using a single measurement method for pattern recognition.

[0118] Table 3. Results of partial discharge pattern recognition based on DS evidence theory

[0119]

[0120] This embodiment is based on the fact that both the antenna and sensor have good sensitivity when measuring partial discharges of different defects in GIS, and the antenna and sensor have a certain degree of complementarity in measuring partial discharges of different defects. Therefore, the DS evidence theory model is used for information fusion. This embodiment proposes a GIS partial discharge photoelectric joint detection pattern recognition method based on multilayer perceptron and utilizing exploration and DS evidence theory. It performs information fusion and judgment decision on the statistical feature parameters of the partial discharge phase analysis spectrum obtained by the UHF method and the optical measurement method.

[0121] Compared with partial discharge pattern recognition using a single evidence body, the method provided in this application embodiment uses DS evidence theory information fusion for pattern recognition, achieving an average recognition rate of over 87.5% for various defects. The problem of unbalanced defect recognition results is improved, and the results are more stable and reliable.

[0122] This application embodiment extracts feature quantities from the discharge data of the object to be identified, and uses a multilayer perceptron to identify the discharge pattern of the feature quantities to determine the defect type of the object to be identified. This solves the problem of inaccurate identification under different defect conditions and avoids false detection and missed detection during the discharge detection process.

[0123] Corresponding to the above embodiments, see [link to relevant documentation]. Figure 5 This application embodiment also provides an insulation combined electrical appliance testing device 500, including:

[0124] The first receiving module 501 is used to receive a first input, wherein the first input is the discharge data of the object to be identified;

[0125] Extraction module 502, in response to the first input, extracts the feature quantities of the discharge data;

[0126] The second receiving module 503 is used to receive a second input, which is a characteristic quantity of the discharge data;

[0127] The first identification module 504, in response to the second input, inputs the feature quantity of the discharge data into the multilayer perceptron and outputs the defect identification result of the object to be identified.

[0128] Optionally, the first identification module 504 includes:

[0129] The third receiving module 5041 is used to input the feature values ​​of the first discharge data and the feature values ​​of the second discharge data into the multilayer sensor, respectively.

[0130] The second identification module 5042 is used to identify the first defect identification result and the second defect identification result of the object to be identified output by the multilayer perceptron.

[0131] Optionally, the insulation switchboard testing device 500 also includes:

[0132] Construction module 505 is used to construct a DS evidence theory model based on the defect type of the object to be identified and the defect data corresponding to each defect type;

[0133] The fusion module 506 is used to input the first defect identification result and the second defect identification result of the object to be identified into the DS evidence theory model, and output the third defect identification result after information fusion.

[0134] Optionally, the fusion module 506 includes:

[0135] The fourth receiving module 5061 is used to receive the first defect identification result and the second defect identification result of the object to be identified.

[0136] The fusion submodule 5062 is used to assign and fuse the first defect identification result of the object to be identified and the second defect identification result of the object to be identified according to the fusion rules and basic probabilities, and output the third defect identification result after information fusion.

[0137] This application embodiment extracts feature quantities from the discharge data of the object to be identified, and uses a multilayer perceptron to identify the discharge pattern of the feature quantities to determine the defect type of the object to be identified. This solves the problem of inaccurate identification under different defect conditions and avoids false detection and missed detection during the discharge detection process.

[0138] An exemplary embodiment of the present invention also provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores a computer program executable by the at least one processor, which, when executed by the at least one processor, causes the electronic device to perform a method according to an embodiment of the present invention.

[0139] An exemplary embodiment of the present invention also provides a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a computer's processor, is used to cause the computer to perform a method according to an embodiment of the present invention.

[0140] An exemplary embodiment of the present invention also provides a computer program product, including a computer program, wherein the computer program, when executed by a computer's processor, is used to cause the computer to perform a method according to an embodiment of the present invention.

[0141] refer to Figure 6The present invention will now be described in the form of a structural block diagram of an electronic device 600 that can serve as a server or client of the present invention, which is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0142] like Figure 6 As shown, the electronic device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. The RAM 603 may also store various programs and data required for the operation of the device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0143] Multiple components in electronic device 600 are connected to I / O interface 605, including: input unit 606, output unit 607, storage unit 608, and communication unit 609. Input unit 606 can be any type of device capable of inputting information to electronic device 600. Input unit 606 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of electronic device. Output unit 607 can be any type of device capable of presenting information and may include, but is not limited to, a display, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 608 may include, but is not limited to, disks and optical discs. Communication unit 609 allows electronic device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth™ devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.

[0144] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above. For example, in some embodiments, the traffic scheduling method can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 600 via ROM 602 and / or communication unit 609. In some embodiments, the computing unit 601 can be configured to perform the traffic scheduling method by any other suitable means (e.g., by means of firmware).

[0145] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0146] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0147] As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, device, and / or apparatus (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including machine-readable media that receive machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0148] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0149] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0150] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other.

Claims

1. A method for testing insulated combined electrical appliances, characterized in that, include: Receive a first input, the first input being the discharge data of the object to be identified, the first input being the first discharge data of the object to be identified measured by a first measurement method and the second discharge data of the object to be identified measured by a second measurement method, the first measurement method being the ultra-high frequency (UHF) method and the second measurement method being the optical method; In response to the first input, feature quantities of the discharge data are extracted, including skewness S. k and steepness K u The skewness is used to describe the degree of asymmetry in the distribution of partial discharge from internal defects in insulation within each power frequency cycle, and the steepness is used to describe the degree of convexity of the partial discharge pattern shape relative to the normal distribution. Receive a second input, which is a characteristic quantity of the discharge data; In response to the second input, the feature values ​​of the discharge data are input into a multilayer perceptron, and the defect identification result of the object to be identified is output. The multilayer perceptron has 6 input nodes i and X1 as the input node. X6 corresponds to l1 l6 characteristic quantities, where S is the value of the entire power frequency cycle. k and K u As characteristic quantities l1 and l2, S of the negative half-cycle of the power frequency k and K u As characteristic quantities l3 and l4, S of the positive half-cycle of the power frequency k and K u As feature quantities l5 and l6; The step of inputting the feature values ​​of the discharge data into a multilayer perceptron and outputting the defect identification result of the object to be identified includes: The feature values ​​of the first discharge data and the feature values ​​of the second discharge data are respectively input into the multilayer sensor. The multilayer perceptron outputs the first defect identification result of the object to be identified and the second defect identification result of the object to be identified. Based on the defect types of the objects to be identified and the defect data corresponding to each defect type, a DS evidence theory model is constructed. The first defect identification result and the second defect identification result of the object to be identified are input into the DS evidence theory model, and the third defect identification result after information fusion is output. The step of inputting the first defect identification result and the second defect identification result of the object to be identified into the DS evidence theory model, and outputting the third defect identification result after information fusion, includes: The DS evidence theory model receives the first defect identification result of the object to be identified and the second defect identification result of the object to be identified. Based on the fusion rules and basic probability assignment, the first defect identification result and the second defect identification result of the object to be identified are fused together, and the third defect identification result after information fusion is output.

2. A testing device for insulating combined electrical appliances, characterized in that, include: A first receiving module is configured to receive a first input, wherein the first input is discharge data of an object to be identified, wherein the first input is first discharge data of the object to be identified measured by a first measurement method and second discharge data of the object to be identified measured by a second measurement method, wherein the first measurement method is ultra-high frequency (UHF) method and the second measurement method is optical measurement method; The extraction module, in response to the first input, extracts the feature quantities of the discharge data, wherein the feature quantities of the discharge data include skewness S. k and steepness K u The skewness is used to describe the degree of asymmetry in the distribution of partial discharge from internal defects in insulation within each power frequency cycle, and the steepness is used to describe the degree of convexity of the partial discharge pattern shape relative to the normal distribution. The second receiving module is used to receive a second input, which is a characteristic quantity of the discharge data; The first identification module, in response to the second input, inputs the feature values ​​of the discharge data into a multilayer perceptron and outputs the defect identification result of the object to be identified. The multilayer perceptron has 6 input nodes i and X1 as the input node. X6 corresponds to l1 l6 characteristic quantities, where S is the value of the entire power frequency cycle. k and K u As characteristic quantities l1 and l2, S of the negative half-cycle of the power frequency k and K u As characteristic quantities l3 and l4, S of the positive half-cycle of the power frequency k and K u As feature quantities l5 and l6; The first identification module is further configured to: input the feature values ​​of the first discharge data and the feature values ​​of the second discharge data into the multilayer perceptron respectively; the multilayer perceptron outputs the first defect identification result of the object to be identified and the second defect identification result of the object to be identified; The construction module is used to construct a DS evidence theory model based on the defect type of the object to be identified and the defect data corresponding to each defect type. The fusion module is used to input the first defect identification result and the second defect identification result of the object to be identified into the DS evidence theory model, and output the third defect identification result after information fusion. The fusion module is further configured to receive the first defect identification result and the second defect identification result of the object to be identified by the DS evidence theory model; and to fuse the first defect identification result and the second defect identification result of the object to be identified according to the fusion rules and basic probability assignment, and output the third defect identification result after information fusion.

3. An electronic device, comprising: processor; as well as Stored program memory, The program includes instructions that, when executed by the processor, cause the processor to perform the method according to claim 1.

4. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to claim 1.

Images