Detection method, device, medium and system for partial discharge capability of switchgear

Abstract

The present application provides a method, device, medium and system for detecting the partial discharge capability of a switch cabinet. The present application relates to the technical field of power system maintenance. The method uses a wavelet transform method to perform denoising on real-time collected data to obtain final collected data; uses a K-means clustering algorithm to classify historical collected data to obtain classification results, and uses the classification results as a training set, uses the training set to train a BP neural network model to obtain the current BP neural network model; uses the current BP neural network model to process the final collected data to obtain the current partial discharge level of the switch cabinet. This solves the problem that most existing solutions are single detection methods, and because only two detection methods are used, the frequency band coverage is insufficient, which easily leads to low accuracy of the detection results.

Description

Technical Field

[0001] The present application relates to the technical field of power system maintenance, and in particular to a method, device, medium and system for detecting partial discharge capability of a switch cabinet. Background Art

[0002] At present, switchgear has been widely used in power systems. As a power distribution equipment directly facing users, the safe and stable operation of switchgear is of vital importance to both the power grid and users. Partial discharge is the main manifestation of insulation defects in switchgear. Therefore, through the monitoring and analysis of partial discharge data, the insulation defects of switchgear can be effectively identified, which makes it easier for operation and maintenance personnel to understand the equipment status in time and carry out targeted maintenance and elimination of equipment defects.

[0003] Commonly used detection methods for partial discharge of switch cabinets include pulse current method, ultrasonic method, transient voltage to ground detection method, ultra-high frequency method, etc. These methods can all realize non-invasive detection of partial discharge, but the detection principles are different, so the scope of application is different. Among them, the pulse current method uses the characteristics of current pulses to determine the degree of partial discharge. However, the frequency band range that this method can detect is small, and it is easily affected by other signals in the field environment; the ultrasonic method has high sensitivity in the low frequency band, and can effectively identify and locate defective parts, but the sensitivity is low in the high frequency band; the transient voltage to ground detection method has strong anti-interference ability, but the cost is high and it cannot accurately locate the fault; the ultra-high frequency method has a wide frequency bandwidth range, which can locate the fault, but the cost is high and it is easily affected by electromagnetic interference during use.

[0004] At present, partial discharge detection of distribution network switch cabinets mostly relies on operation and maintenance personnel to go to the site with handheld devices, and the level of automation and intelligence of distribution operation and maintenance is low. At the same time, due to the large number of distribution rooms that need to be maintained, only periodic detection at intervals can be carried out, and daily monitoring of partial discharge cannot be achieved. In addition, when the operation and maintenance personnel conduct detection, they are close to the distribution equipment, which poses a safety hazard.

[0005] Most of the existing methods are single detection methods, and when using joint detection methods, most of them only use two detection methods, which leads to insufficient frequency band coverage and easily leads to inaccurate detection results.

[0006] The existing switch cabinet partial discharge detection system requires the installation of an additional power supply system, which is costly and increases the complexity of the system structure.

[0007] That is, most of the existing solutions are single detection methods. Since only two detection methods are used, the frequency band coverage is insufficient, which easily leads to low accuracy of the detection results. Summary of the invention

[0008] The main purpose of the present application is to provide a method for detecting the partial discharge capability of a switch cabinet, a device for detecting the partial discharge capability of a switch cabinet, a computer-readable storage medium, a processor and an electronic device, so as to at least solve the problem that most of the existing solutions are single detection methods, and since only two detection methods are used, the frequency band coverage is insufficient, which easily leads to the problem of low accuracy of the detection results.

[0009] In order to achieve the above object, according to one aspect of the present application, a method for detecting partial discharge capability of a switch cabinet is provided, the method comprising:

[0010] Acquire the data collected by the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor on the switch cabinet to obtain real-time collected data, wherein the ultra-high frequency sensor is used to collect the ultra-high frequency data collected by the pulse current detector, the ultrasonic sensor and the transient ground voltage sensor;

[0011] Using wavelet transform method to perform denoising on the real-time collected data to obtain final collected data;

[0012] Using K-means clustering algorithm to classify historical collected data to obtain classification results, and using the classification results as training sets, using the training sets to train the BP neural network model to obtain the current BP neural network model;

[0013] The final collected data is processed using the current BP neural network model to obtain a current partial discharge level of the switch cabinet, wherein the current partial discharge level represents a real-time partial discharge capability of the switch cabinet.

[0014] Optionally, the current BP neural network model is used to process the final collected data to obtain the current partial discharge level of the switch cabinet, including: using the current BP neural network model to process the collected data of the pulse current detector in the final collected data to obtain a first partial discharge score; using the current BP neural network model to process the collected data of the ultrasonic sensor in the final collected data to obtain a second partial discharge score; using the current BP neural network model to process the collected data of the transient ground voltage sensor in the final collected data to obtain a third partial discharge score; using the current BP neural network model to process the collected data of the ultra-high frequency sensor in the final collected data to obtain a fourth partial discharge score; performing weighted sum processing on the first partial discharge score, the second partial discharge score, the third partial discharge score and the fourth partial discharge score to obtain a final partial discharge score, and determining the current partial discharge level according to the range of the current partial discharge level.

[0015] Optionally, the current BP neural network model is used to process the collected data of the pulse current detector in the final collected data to obtain a first local discharge score, including: using the discharge amount, number of discharges and phase in the final collected data as inputs related to the current BP neural network model and the collected data of the pulse current detector; and using the output related to the current BP neural network model and the collected data of the pulse current detector as the first local discharge score.

[0016] Optionally, the current BP neural network model is used to process the collected data of the ultrasonic sensor in the final collected data to obtain a second partial discharge score, including: using the effective value, maximum value, first correlation and second correlation of the collected data of the ultrasonic sensor in the final collected data as inputs of the current BP neural network model related to the collected data of the ultrasonic sensor, the first correlation characterizing the correlation between the collected signal of the ultrasonic sensor and a first signal frequency threshold, the second correlation characterizing the correlation between the collected signal of the ultrasonic sensor and a second signal frequency threshold, and the second signal frequency threshold is greater than the first signal frequency threshold; using the output of the current BP neural network model related to the collected data of the ultrasonic sensor as the second partial discharge score.

[0017] Optionally, the current BP neural network model is used to process the collected data of the transient ground voltage sensor in the final collected data to obtain a third partial discharge score: the transient ground voltage return value of the collected data of the transient ground voltage sensor in the final collected data is used as the input of the current BP neural network model related to the collected data of the transient ground voltage sensor; the output of the current BP neural network model related to the collected data of the transient ground voltage sensor is used as the third partial discharge score.

[0018] Optionally, the current BP neural network model is used to process the collected data of the ultra-high frequency sensor in the final collected data to obtain a fourth partial discharge score: the effective value, maximum value and frequency of the collected data of the ultra-high frequency sensor in the final collected data are used as inputs of the current BP neural network model related to the collected data of the ultra-high frequency sensor; and the output of the current BP neural network model related to the collected data of the ultra-high frequency sensor is used as the fourth partial discharge score.

[0019] Optionally, the current partial discharge level is one of the following: a first partial discharge level, a second partial discharge level and a third partial discharge level, which characterize the real-time partial discharge capability of the switch cabinet in descending order. After the final collected data is processed by the current BP neural network model to obtain the current partial discharge level of the switch cabinet, the method further includes: when the current partial discharge level is the first partial discharge level, determining that the real-time partial discharge capability of the switch cabinet is qualified; when the current partial discharge level is the second partial discharge level, generating a first prompt message to prompt that the switch cabinet needs to be inspected within a preset time; when the current partial discharge level is the third partial discharge level, generating a second prompt message to prompt that the switch cabinet needs to be repaired.

[0020] According to another aspect of the present application, a device for detecting partial discharge capability of a switch cabinet is provided, the device comprising:

[0021] An acquisition unit is used to acquire the data collected by the pulse current detector, the ultrasonic sensor, the transient ground voltage sensor and the ultra-high frequency sensor on the switch cabinet to obtain real-time collected data, wherein the ultra-high frequency sensor is used to collect the ultra-high frequency data collected by the pulse current detector, the ultrasonic sensor and the transient ground voltage sensor;

[0022] A first processing unit is used to perform denoising on the real-time collected data by using a wavelet transform method to obtain final collected data;

[0023] The second processing unit is used to classify the historical collected data using a K-means clustering algorithm to obtain a classification result, and use the classification result as a training set to train the BP neural network model to obtain a current BP neural network model;

[0024] The third processing unit is used to process the final collected data using the current BP neural network model to obtain the current partial discharge level of the switch cabinet, where the current partial discharge level represents the real-time partial discharge capability of the switch cabinet.

[0025] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the methods described.

[0026] According to another aspect of the present application, a system for detecting the partial discharge capability of a switch cabinet is provided, comprising: a pulse current detector, an ultrasonic sensor, a transient ground voltage sensor, a UHF sensor and a controller, wherein the controller communicates with the pulse current detector, the ultrasonic sensor, the transient ground voltage sensor and the UHF sensor respectively, and the controller is used to execute any one of the methods described.

[0027] By applying the technical solution of the present application, the real-time collected data is obtained by acquiring the collected data of the switch cabinet by the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor; the real-time collected data is denoised by wavelet transform method to obtain the final collected data; the historical collected data is classified by K-means clustering algorithm to obtain the classification result, and the classification result is used as the training set, and the BP neural network model is trained by the training set to obtain the current BP neural network model; the final collected data is processed by the current BP neural network model to obtain the current partial discharge level of the switch cabinet. The joint detection of partial discharge is realized, that is, the problem that most of the existing solutions are single detection methods, and because only two detection methods are used, the frequency band coverage is insufficient, which easily leads to the problem of low accuracy of the detection result. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0029] Figure 1 A schematic flow chart of a method for detecting partial discharge capability of a switch cabinet provided in accordance with an embodiment of the present application is shown;

[0030] Figure 2 A schematic flow chart of another method for detecting partial discharge capability of a switch cabinet provided according to an embodiment of the present application is shown;

[0031] Figure 3 A structural block diagram of a device for detecting partial discharge capability of a switch cabinet provided according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0032] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0033] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0035] For the convenience of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0036] Pulse current method: By applying a short pulse current inside the switch cabinet and observing the spectrum characteristics of the partial discharge signal, the possible partial discharge defects can be judged. This method is suitable for detecting partial discharge in high-voltage equipment and can quickly and accurately locate faults.

[0037] Ultrasonic method: Use ultrasonic detectors to scan the switch cabinet and determine whether there is a partial discharge defect by measuring the sound wave propagation speed and sound wave reflection characteristics. This method has a high sensitivity for detecting tiny cracks and air gaps in insulating materials.

[0038] Transient voltage-to-ground detection method: By measuring the transient voltage change between the switch cabinet and the ground, it is possible to determine whether there is a partial discharge phenomenon. This method is suitable for detecting partial discharge defects inside the equipment and can quickly locate the fault point.

[0039] UHF method: Use UHF signal sensors to monitor the switch cabinet and analyze the signal's spectrum characteristics and amplitude changes to determine whether there is a partial discharge fault. This method is very effective in detecting partial discharge defects in high-voltage equipment and can provide accurate fault location information.

[0040] K-means algorithm is a commonly used clustering algorithm, which divides the data set into K clusters so that the data points in the same cluster are closest to each other, while the data points in different clusters are farthest away.

[0041] The BP neural network model is a commonly used artificial neural network model. It consists of an input layer, a hidden layer, and an output layer. The back propagation algorithm is used to continuously adjust the connection weights between neurons to achieve the learning and training process. It can be used to solve classification, regression, and other problems.

[0042] As introduced in the background technology, switch cabinets are currently widely used in power systems. As power distribution equipment directly facing users, the safe and stable operation of switch cabinets is of vital importance to both power grids and users. Partial discharge is the main manifestation of insulation defects in switch cabinets. Therefore, through the monitoring and analysis of partial discharge data, effective identification of insulation defects in switch cabinets can be achieved, which is convenient for operation and maintenance personnel to understand the equipment status in a timely manner and to carry out targeted maintenance and elimination of equipment defects. Commonly used detection methods for partial discharge of switch cabinets mainly include pulse current method, ultrasonic method, transient voltage to ground detection method, ultra-high frequency method, etc. These methods can all achieve non-invasive detection of partial discharge, but the detection principles are different, so the scope of application is different. Among them, the pulse current method uses the characteristics of current pulses to determine the degree of partial discharge. However, the frequency band range that this method can detect is small, and it is easily affected by other signals in the field environment. The ultrasonic method has high sensitivity in the low frequency band and can effectively identify and locate the defective parts, but it has low sensitivity in the high frequency band. The transient voltage to ground detection method has strong anti-interference ability, but the cost is high and it cannot accurately locate the fault. The frequency bandwidth range of the ultra-high frequency method is wide, and the fault can be located, but the cost is high and it is easily affected by electromagnetic interference during use. At present, the partial discharge detection of distribution network switch cabinets mostly relies on the operation and maintenance personnel to carry out handheld devices on site, and the level of automation and intelligence of distribution operation and maintenance is low. At the same time, due to the large number of distribution rooms that need to be maintained, only periodic detection can be carried out at intervals, and daily monitoring of partial discharge cannot be achieved. In addition, when the operation and maintenance personnel conduct detection, they are close to the distribution equipment, which poses a safety hazard. Most of the existing methods are single detection methods, and when using the joint detection method, most of them only use two detection methods, and the frequency band coverage is insufficient, which easily leads to inaccurate detection results. The existing switch cabinet partial discharge detection system needs to install an additional power supply system, which is costly and increases the complexity of the system structure. That is, most of the existing solutions are single detection methods. Since only two detection methods are used, the frequency band coverage is insufficient, which easily leads to low accuracy of the detection results. In order to solve the problem that most of the existing solutions are single detection methods, since only two detection methods are used, the frequency band coverage is insufficient, which easily leads to low accuracy of the detection results, the embodiments of the present application provide a detection method, device, medium and system for the partial discharge capability of a switch cabinet.

[0043] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0044] In this embodiment, a method for detecting the partial discharge capability of a switch cabinet is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0045] Figure 1 FIG. 1 is a flow chart of a method for detecting partial discharge capability of a switch cabinet according to an embodiment of the present application. Figure 1 As shown, the method comprises the following steps:

[0046] Step S101, acquiring data collected by a pulse current detector, an ultrasonic sensor, a transient ground voltage sensor and a UHF sensor on a switch cabinet to obtain real-time collected data, wherein the UHF sensor is used to collect UHF data collected by the pulse current detector, the ultrasonic sensor and the transient ground voltage sensor;

[0047] Specifically, the pulse current detector is used to detect current pulse signals; the ultrasonic sensor is used to collect ultrasonic signals; and the transient ground voltage sensor is used to collect ground wave signals. The pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor signal acquisition module work simultaneously to detect the partial discharge of the switch cabinet. The present application performs fusion detection on the partial discharge signal of the switch cabinet based on the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor, overcoming the problem of insufficient detection accuracy caused by traditional single or two sensors. The discharge position can be accurately located by the fusion detection method. Combined with the signals sent back by the ultrasonic sensor and the ultra-high frequency sensor, the fault position can be comprehensively judged based on information such as the signal reception time and the signal attenuation degree. By combining multiple detection methods, the joint detection of partial discharge of the switch cabinet is realized, including the pulse current method, the transient ground voltage method, the ultrasonic method and the ultra-high frequency method. The problem of insufficient accuracy of the traditional single or two-unit detection method is solved.

[0048] Step S102, using wavelet transform method to perform denoising on the above real-time collected data to obtain final collected data;

[0049] Specifically, the collected signal is denoised using a finite-length attenuated wavelet basis using wavelet transform:

[0050] ;

[0051] Where f(t) is the original signal, It is the basis function of wavelet transform. WT (α, τ) is the signal of the final collected data when the scale factor is α and the displacement is τ. α and τ are the scale factor and displacement respectively. The scalable wavelet basis function is multiplied with the original signal to obtain the frequency components contained in the signal at the current scale. Compared with Fourier transform, wavelet transform has more scale α and displacement τ, which correspond to frequency and time respectively. It can extract information from the signal more effectively and obtain the time-frequency spectrum of partial discharge.

[0052] Step S103, using K-means clustering algorithm to classify the historical collected data to obtain classification results, and using the above classification results as a training set, using the above training set to train the BP neural network model to obtain the current BP neural network model;

[0053] First, the K-means algorithm is used to cluster the training set and divide the data set into different categories. The classification results of the K-means algorithm are used as the input features of the BP neural network, that is, the category of each sample is added to the original data set as a new feature. The data set with the K-means classification results is used as the training set of the BP neural network, and the category of each sample is used as the output label. The BP neural network model is trained to learn the relationship between the input features and the output labels. The trained BP neural network model is used to predict new samples to obtain the output results. Finally, the classification type of the sample is determined according to the classification score value of the output result, and the category to which the sample belongs can be determined according to the size of the classification score value.

[0054] Through this method, the clustering results of the K-means algorithm can be combined with the BP neural network model, so that the BP neural network can output the corresponding classification score value, thereby determining the classification type of the sample. This method can improve the accuracy and reliability of classification.

[0055] Specifically, the classification and feature extraction of partial discharge signals are realized through the K-means clustering algorithm and the BP neural network algorithm to determine whether the insulation state of the switch cabinet is at a safe level. The processed signal is input into the BP neural network algorithm trained with historical data, and the algorithm classification number is obtained according to the K-means clustering algorithm. Finally, the classification results and feature information of the partial discharge signal are obtained, and compared with the historical data, the insulation state of the switch cabinet and the development trend of partial discharge can be determined.

[0056] Step S104, using the current BP neural network model to process the final collected data to obtain the current partial discharge level of the switch cabinet, wherein the current partial discharge level represents the real-time partial discharge capability of the switch cabinet.

[0057] In the above steps, the real-time collected data is obtained by acquiring the collected data of the switch cabinet by the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor; the real-time collected data is denoised by wavelet transform method to obtain the final collected data; the historical collected data is classified by K-means clustering algorithm to obtain the classification result, and the classification result is used as the training set, and the BP neural network model is trained by the training set to obtain the current BP neural network model; the final collected data is processed by the current BP neural network model to obtain the current partial discharge level of the switch cabinet. The joint detection of partial discharge is realized, that is, the problem that most of the existing solutions are single detection methods, and the frequency band coverage is insufficient due to the use of only two detection methods, which easily leads to low accuracy of the detection results.

[0058] Specifically, the BP neural network algorithm is used as a classifier for the pulse current detector, ultrasonic sensor, transient ground voltage sensor and UHF sensor to collect signals, and has a three-layer topological structure of input layer, hidden layer and output layer, which are connected by neurons to form a neural network structure. The output layer neurons correspond to the defect types obtained based on historical data. The number of hidden layers is generally set according to experience, and is set to 6 in this system. Four classifiers based on BP neural network are established to calculate the four detectors respectively. The input layer transmits the input feature vector to the hidden layer, and the optimal weight is obtained after training, thereby establishing a classification model.

[0059] In one embodiment of the present application, the step S104 of using the current BP neural network model to process the final collected data to obtain the current partial discharge level of the switch cabinet includes: using the current BP neural network model to process the collected data of the pulse current detector in the final collected data to obtain a first partial discharge score; using the current BP neural network model to process the collected data of the ultrasonic sensor in the final collected data to obtain a second partial discharge score; using the current BP neural network model to process the collected data of the transient ground voltage sensor in the final collected data to obtain a third partial discharge score; using the current BP neural network model to process the collected data of the ultra-high frequency sensor in the final collected data to obtain a fourth partial discharge score; performing weighted sum processing on the first partial discharge score, the second partial discharge score, the third to the fourth partial discharge score, and obtaining a final partial discharge score, and determining the current partial discharge level according to the range of the current partial discharge level.

[0060] Specifically, the wavelet filtering algorithm and the BP neural network algorithm are combined to preprocess, classify and extract features of each sensor signal. Based on the analysis of historical data, effective judgment and classification of the insulation status of the switch cabinet can be achieved.

[0061] In the specific implementation process, the current BP neural network model is used to process the collected data of the pulse current detector in the final collected data to obtain a first local discharge score, including: using the discharge amount, number of discharges and phase in the final collected data as inputs related to the current BP neural network model and the collected data of the pulse current detector; using the output related to the current BP neural network model and the collected data of the pulse current detector as the first local discharge score.

[0062] Specifically, the BP neural network is used to identify the pulse current signal, and the input layer is three layers, corresponding to the characteristic quantities of the pulse current: discharge amount, discharge number and phase.

[0063] In one embodiment of the present application, the current BP neural network model is used to process the collected data of the ultrasonic sensor in the final collected data to obtain a second partial discharge score, including: using the effective value, maximum value, first correlation and second correlation of the collected data of the ultrasonic sensor in the final collected data as inputs of the current BP neural network model related to the collected data of the ultrasonic sensor, the first correlation characterizing the correlation between the collected signal of the ultrasonic sensor and the first signal frequency threshold, the second correlation characterizing the correlation between the collected signal of the ultrasonic sensor and the second signal frequency threshold, and the second signal frequency threshold is greater than the first signal frequency threshold; using the output of the current BP neural network model related to the collected data of the ultrasonic sensor as the second partial discharge score.

[0064] Specifically, for the recognition of ultrasonic signals, the neural network input layer is 4 layers, corresponding to its characteristic quantities: effective value, maximum value, first correlation and second correlation; the above-mentioned first signal frequency threshold can be 50Hz, and the above-mentioned second signal frequency threshold can be 100Hz.

[0065] In the specific implementation process, the current BP neural network model is used to process the collected data of the transient ground voltage sensor in the final collected data to obtain a third partial discharge score: the transient ground voltage return value of the collected data of the transient ground voltage sensor in the final collected data is used as the input of the current BP neural network model related to the collected data of the transient ground voltage sensor; the output of the current BP neural network model related to the collected data of the transient ground voltage sensor is used as the third partial discharge score.

[0066] Specifically, for the transient ground voltage signal, the input layer is set to the obtained transient ground voltage return value.

[0067] In one embodiment of the present application, the current BP neural network model is used to process the collected data of the above-mentioned UHF sensor in the above-mentioned final collected data to obtain a fourth local discharge score: the effective value, maximum value and frequency of the collected data of the above-mentioned UHF sensor in the above-mentioned final collected data are used as inputs related to the above-mentioned current BP neural network model and the collected data of the above-mentioned UHF sensor; the output related to the above-mentioned current BP neural network model and the collected data of the above-mentioned UHF sensor is used as the above-mentioned fourth local discharge score.

[0068] Specifically, for UHF signals, the input layer is set to 3 layers, corresponding to effective value, maximum value and frequency respectively.

[0069] In a specific implementation process, the current partial discharge level is one of the following: a first partial discharge level, a second partial discharge level and a third partial discharge level, which characterize the real-time partial discharge capability of the switch cabinet in descending order. After the final collected data is processed by the current BP neural network model to obtain the current partial discharge level of the switch cabinet, the method further includes: when the current partial discharge level is the first partial discharge level, determining that the real-time partial discharge capability of the switch cabinet is qualified; when the current partial discharge level is the second partial discharge level, generating a first prompt message to prompt that the switch cabinet needs to be inspected within a preset time; when the current partial discharge level is the third partial discharge level, generating a second prompt message to prompt that the switch cabinet needs to be repaired.

[0070] Specifically, the discharge spectrum of historical data corresponds to the degree of insulation damage, and the K-Means clustering algorithm is used to obtain the number of categories for dividing the severity of discharge. The elbow coefficient method and the silhouette coefficient method are combined to obtain the optimal number of categories.

[0071] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the method for detecting the partial discharge capability of the switch cabinet of the present application will be described in detail below in combination with specific embodiments.

[0072] This embodiment relates to a specific method for detecting the partial discharge capability of a switch cabinet, such as Figure 2 As shown, including:

[0073] Acquire the data collected by the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor on the switch cabinet to obtain real-time collected data. The ultra-high frequency sensor is used to collect ultra-high frequency data collected by the pulse current detector, ultrasonic sensor and transient ground voltage sensor;

[0074] The wavelet transform method is used to denoise the real-time collected data to obtain the final collected data;

[0075] The K-means clustering algorithm is used to classify the historical collected data to obtain the classification results, and the classification results are used as training sets. The training sets are used to train the BP neural network model to obtain the current BP neural network model;

[0076] The final collected data is processed using the current BP neural network model to obtain the current partial discharge level of the switchgear, which represents the real-time partial discharge capability of the switchgear;

[0077] When the current partial discharge level is the first partial discharge level, it is determined that the real-time partial discharge capability of the switch cabinet is qualified; when the current partial discharge level is the second partial discharge level, a first prompt message is generated to prompt that the switch cabinet needs to be inspected within a preset time; when the current partial discharge level is the third partial discharge level, a second prompt message is generated to prompt that the switch cabinet needs to be repaired.

[0078] The embodiment of the present application also provides a device for detecting the partial discharge capability of a switch cabinet. It should be noted that the device for detecting the partial discharge capability of a switch cabinet in the embodiment of the present application can be used to execute the detection method for the partial discharge capability of a switch cabinet provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0079] The following is an introduction to the detection device for partial discharge capability of a switch cabinet provided in an embodiment of the present application.

[0080] Figure 3 1 is a structural block diagram of a device for detecting partial discharge capability of a switch cabinet provided according to an embodiment of the present application. Figure 3 As shown, the device comprises:

[0081] The acquisition unit 31 is used to acquire the data collected by the pulse current detector, the ultrasonic sensor, the transient ground voltage sensor and the ultra-high frequency sensor on the switch cabinet to obtain real-time collected data. The ultra-high frequency sensor is used to collect the ultra-high frequency data collected by the pulse current detector, the ultrasonic sensor and the transient ground voltage sensor;

[0082] The first processing unit 32 is used to perform denoising on the real-time collected data using a wavelet transform method to obtain final collected data;

[0083] The second processing unit 33 is used to classify the historical collected data using a K-means clustering algorithm to obtain a classification result, and use the classification result as a training set to train the BP neural network model to obtain a current BP neural network model;

[0084] The third processing unit 34 is used to process the final collected data using the current BP neural network model to obtain the current partial discharge level of the switch cabinet, where the current partial discharge level represents the real-time partial discharge capability of the switch cabinet.

[0085] In the above device, the real-time collected data is obtained by acquiring the collected data of the switch cabinet by the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor; the real-time collected data is denoised by wavelet transform method to obtain the final collected data; the historical collected data is classified by K-means clustering algorithm to obtain the classification result, and the classification result is used as the training set, and the BP neural network model is trained by the training set to obtain the current BP neural network model; the final collected data is processed by the current BP neural network model to obtain the current partial discharge level of the switch cabinet. The joint detection of partial discharge is realized, that is, the problem that most of the existing solutions are single detection methods, and because only two detection methods are used, the frequency band coverage is insufficient, which easily leads to low accuracy of the detection results.

[0086] In one embodiment of the present application, the third processing unit includes a first processing module, a second processing module, a third processing module, a fourth processing module and a fifth processing module, the first processing module is used to use the current BP neural network model to process the collected data of the pulse current detector in the final collected data to obtain a first partial discharge score; the second processing module is used to use the current BP neural network model to process the collected data of the ultrasonic sensor in the final collected data to obtain a second partial discharge score; the third processing module is used to use the current BP neural network model to process the collected data of the transient ground voltage sensor in the final collected data to obtain a third partial discharge score; the fourth processing module is used to use the current BP neural network model to process the collected data of the ultra-high frequency sensor in the final collected data to obtain a fourth partial discharge score; the fifth processing module is used to perform weighted sum processing on the first partial discharge score, the second partial discharge score, the third partial discharge score and the fourth partial discharge score to obtain a final partial discharge score, and determine the current partial discharge level according to the range of the current partial discharge level.

[0087] In one embodiment of the present application, the first processing module includes a first processing sub-module and a second processing sub-module, the first processing sub-module is used to use the discharge amount, number of discharges and phase in the above-mentioned final collected data as the input related to the above-mentioned current BP neural network model and the collected data of the above-mentioned pulse current detector; the second processing sub-module is used to use the output related to the above-mentioned current BP neural network model and the collected data of the above-mentioned pulse current detector as the above-mentioned first local discharge score.

[0088] In one embodiment of the present application, the second processing module includes a third processing sub-module and a fourth processing sub-module, the third processing sub-module is used to use the effective value, maximum value, first correlation and second correlation of the collected data of the above-mentioned ultrasonic sensor in the above-mentioned final collected data as the input related to the above-mentioned current BP neural network model and the collected data of the above-mentioned ultrasonic sensor, the above-mentioned first correlation characterizes the correlation between the collected signal of the above-mentioned ultrasonic sensor and the first signal frequency threshold, the above-mentioned second correlation characterizes the correlation between the collected signal of the above-mentioned ultrasonic sensor and the second signal frequency threshold, and the above-mentioned second signal frequency threshold is greater than the above-mentioned first signal frequency threshold; the fourth processing sub-module is used to use the output related to the above-mentioned current BP neural network model and the collected data of the above-mentioned ultrasonic sensor as the above-mentioned second partial discharge score.

[0089] In one embodiment of the present application, the third processing module includes a fifth processing submodule and a sixth processing submodule, the fifth processing submodule is used to use the transient ground voltage return value of the collection data of the above-mentioned transient ground voltage sensor in the above-mentioned final collection data as the input related to the above-mentioned current BP neural network model and the collection data of the above-mentioned transient ground voltage sensor; the sixth processing submodule is used to use the output related to the above-mentioned current BP neural network model and the collection data of the above-mentioned transient ground voltage sensor as the above-mentioned third partial discharge score.

[0090] In one embodiment of the present application, the fourth processing module includes a seventh processing sub-module and an eighth processing sub-module, the seventh processing sub-module is used to use the effective value, maximum value and frequency of the collection data of the above-mentioned ultra-high frequency sensor in the above-mentioned final collection data as the input related to the above-mentioned current BP neural network model and the collection data of the above-mentioned ultra-high frequency sensor; the eighth processing sub-module is used to use the output related to the above-mentioned current BP neural network model and the collection data of the above-mentioned ultra-high frequency sensor as the above-mentioned fourth partial discharge score.

[0091] In one embodiment of the present application, the above-mentioned device also includes a determination unit, a first generation unit and a second generation unit. The determination unit is used to determine that the above-mentioned real-time partial discharge capability of the above-mentioned switch cabinet is qualified when the above-mentioned current partial discharge level is the above-mentioned first partial discharge level; the first generation unit is used to generate a first prompt information when the above-mentioned current partial discharge level is the above-mentioned second partial discharge level, so as to prompt that the above-mentioned switch cabinet needs to be inspected within a preset time; the second generation unit is used to generate a second prompt information when the above-mentioned current partial discharge level is the above-mentioned third partial discharge level, so as to prompt that the above-mentioned switch cabinet needs to be repaired.

[0092] The above-mentioned switch cabinet partial discharge capability detection device comprises a processor and a memory, wherein the above-mentioned acquisition unit, the first processing unit and the second processing unit are stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions. The above-mentioned modules are all located in the same processor; or, the above-mentioned modules are located in different processors in any combination.

[0093] The processor contains a kernel, which calls the corresponding program unit from the memory. One or more kernels can be set, and the kernel parameters can be adjusted to solve the problem that most existing solutions use a single detection method. Since only two detection methods are used, the frequency band coverage is insufficient, which easily leads to low accuracy of the detection results.

[0094] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0095] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for detecting the partial discharge capability of the switch cabinet.

[0096] Specifically, the detection method of the partial discharge capability of the switchgear includes:

[0097] Step S101, acquiring data collected by a pulse current detector, an ultrasonic sensor, a transient ground voltage sensor and a UHF sensor on a switch cabinet to obtain real-time collected data, wherein the UHF sensor is used to collect UHF data collected by the pulse current detector, the ultrasonic sensor and the transient ground voltage sensor;

[0098] Specifically, the pulse current detector is used to detect current pulse signals; the ultrasonic sensor is used to collect ultrasonic signals; and the transient ground voltage sensor is used to collect ground wave signals. The pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor signal acquisition module work simultaneously to detect the partial discharge of the switch cabinet. The present application performs fusion detection on the partial discharge signal of the switch cabinet based on the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor, overcoming the problem of insufficient detection accuracy caused by traditional single or two sensors. The discharge position can be accurately located by the fusion detection method. Combined with the signals sent back by the ultrasonic sensor and the ultra-high frequency sensor, the fault position can be comprehensively judged based on information such as the signal reception time and the signal attenuation degree. By combining multiple detection methods, the joint detection of partial discharge of the switch cabinet is realized, including the pulse current method, the transient ground voltage method, the ultrasonic method and the ultra-high frequency method. The problem of insufficient accuracy of the traditional single or two-unit detection method is solved.

[0099] Step S102, using wavelet transform method to perform denoising on the above real-time collected data to obtain final collected data;

[0100] Specifically, the collected signal is denoised using a finite-length attenuated wavelet basis using wavelet transform, and its expression is shown in formula (1):

[0101] (1)

[0102] Where f(t) is the original signal, It is the basis function of wavelet transform, α and τ are scale factor and displacement respectively. The scalable wavelet basis function is multiplied with the original signal to obtain the frequency components contained in the signal at the current scale. Compared with Fourier transform, wavelet transform has more scale α and displacement τ, which correspond to frequency and time respectively, and can more effectively extract information from the signal to obtain the time-frequency spectrum of partial discharge.

[0103] Step S103, using K-means clustering algorithm to classify the historical collected data to obtain classification results, and using the above classification results as a training set, using the above training set to train the BP neural network model to obtain the current BP neural network model;

[0104] Specifically, the classification and feature extraction of partial discharge signals are realized through the K-means clustering algorithm and the BP neural network algorithm to determine whether the insulation state of the switch cabinet is at a safe level. The processed signal is input into the BP neural network algorithm trained with historical data, and the algorithm classification number is obtained according to the K-means clustering algorithm. Finally, the classification results and feature information of the partial discharge signal are obtained, and compared with the historical data, the insulation state of the switch cabinet and the development trend of partial discharge can be determined.

[0105] Step S104, using the current BP neural network model to process the final collected data to obtain the current partial discharge level of the switch cabinet, wherein the current partial discharge level represents the real-time partial discharge capability of the switch cabinet.

[0106] Specifically, the BP neural network algorithm is used as a classifier for the pulse current detector, ultrasonic sensor, transient ground voltage sensor and UHF sensor to collect signals, and has a three-layer topological structure of input layer, hidden layer and output layer, which are connected by neurons to form a neural network structure. The output layer neurons correspond to the defect types obtained based on historical data. The number of hidden layers is generally set according to experience, and is set to 6 in this system. Four classifiers based on BP neural network are established to calculate the four detectors respectively. The input layer transmits the input feature vector to the hidden layer, and the optimal weight is obtained after training, thereby establishing a classification model.

[0107] Optionally, the current BP neural network model is used to process the final collected data to obtain the current partial discharge level of the switch cabinet, including: using the current BP neural network model to process the collected data of the pulse current detector in the final collected data to obtain a first partial discharge score; using the current BP neural network model to process the collected data of the ultrasonic sensor in the final collected data to obtain a second partial discharge score; using the current BP neural network model to process the collected data of the transient ground voltage sensor in the final collected data to obtain a third partial discharge score; using the current BP neural network model to process the collected data of the ultra-high frequency sensor in the final collected data to obtain a fourth partial discharge score; performing weighted sum processing on the first partial discharge score, the second partial discharge score, the third to fourth partial discharge scores and the fourth partial discharge score to obtain a final partial discharge score, and determining the current partial discharge level according to the range of the current partial discharge level.

[0108] Optionally, the current BP neural network model is used to process the collected data of the pulse current detector in the final collected data to obtain a first local discharge score, including: using the discharge amount, number of discharges and phase in the final collected data as inputs related to the current BP neural network model and the collected data of the pulse current detector; and using the output related to the current BP neural network model and the collected data of the pulse current detector as the first local discharge score.

[0109] Optionally, the current BP neural network model is used to process the collected data of the ultrasonic sensor in the final collected data to obtain a second local discharge score, including: using the effective value, maximum value, first correlation and second correlation of the collected data of the ultrasonic sensor in the final collected data as inputs of the current BP neural network model related to the collected data of the ultrasonic sensor, the first correlation characterizes the correlation between the collected signal of the ultrasonic sensor and a first signal frequency threshold, the second correlation characterizes the correlation between the collected signal of the ultrasonic sensor and a second signal frequency threshold, and the second signal frequency threshold is greater than the first signal frequency threshold; using the output of the current BP neural network model related to the collected data of the ultrasonic sensor as the second local discharge score.

[0110] Optionally, the current BP neural network model is used to process the collected data of the transient ground voltage sensor in the final collected data to obtain a third partial discharge score: the transient ground voltage return value of the collected data of the transient ground voltage sensor in the final collected data is used as the input of the current BP neural network model related to the collected data of the transient ground voltage sensor; the output of the current BP neural network model related to the collected data of the transient ground voltage sensor is used as the third partial discharge score.

[0111] Optionally, the current BP neural network model is used to process the collected data of the UHF sensor in the final collected data to obtain a fourth local discharge score: the effective value, maximum value and frequency of the collected data of the UHF sensor in the final collected data are used as inputs of the current BP neural network model related to the collected data of the UHF sensor; the output of the current BP neural network model related to the collected data of the UHF sensor is used as the fourth local discharge score.

[0112] Optionally, the current partial discharge level is one of the following: a first partial discharge level, a second partial discharge level and a third partial discharge level, which characterize the real-time partial discharge capability of the switch cabinet in descending order. After the final collected data is processed by the current BP neural network model to obtain the current partial discharge level of the switch cabinet, the method further includes: when the current partial discharge level is the first partial discharge level, determining that the real-time partial discharge capability of the switch cabinet is qualified; when the current partial discharge level is the second partial discharge level, generating a first prompt message to prompt that the switch cabinet needs to be inspected within a preset time; when the current partial discharge level is the third partial discharge level, generating a second prompt message to prompt that the switch cabinet needs to be repaired.

[0113] An embodiment of the present invention provides a processor, and the processor is used to run a program, wherein the program executes the method for detecting the partial discharge capability of the switch cabinet when running.

[0114] The embodiment of the present invention provides a device, the device includes a processor, a memory and a program stored in the memory and can be run on the processor, and when the processor executes the program, at least the following steps are implemented: acquiring the data collected by the pulse current detector, the ultrasonic sensor, the transient ground voltage sensor and the ultra-high frequency sensor on the switch cabinet to obtain real-time collected data, the ultra-high frequency sensor is used to collect the ultra-high frequency data collected by the pulse current detector, the ultrasonic sensor and the transient ground voltage sensor; using the wavelet transform method to denoise the real-time collected data to obtain the final collected data; using the K-means clustering algorithm to classify the historical collected data to obtain the classification result, and using the classification result as a training set, using the training set to train the BP neural network model to obtain the current BP neural network model; using the current BP neural network model to process the final collected data to obtain the current partial discharge level of the switch cabinet, and the current partial discharge level represents the real-time partial discharge capability of the switch cabinet. The device in this article can be a server, a PC, a PAD, a mobile phone, etc.

[0115] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that is initialized with at least the following method steps: obtaining the data collected by a pulse current detector, an ultrasonic sensor, a transient ground voltage sensor and an ultra-high frequency sensor on a switch cabinet to obtain real-time collected data, wherein the ultra-high frequency sensor is used to collect the ultra-high frequency data collected by the pulse current detector, the ultrasonic sensor and the transient ground voltage sensor; performing denoising processing on the real-time collected data using a wavelet transform method to obtain final collected data; classifying the historical collected data using a K-means clustering algorithm to obtain a classification result, and using the classification result as a training set, and using the training set to train a BP neural network model to obtain a current BP neural network model; using the current BP neural network model to process the final collected data to obtain a current partial discharge level of the switch cabinet, wherein the current partial discharge level represents the real-time partial discharge capability of the switch cabinet.

[0116] The present application also provides a detection system for the partial discharge capability of a switch cabinet, comprising: a pulse current detector, an ultrasonic sensor, a transient ground voltage sensor, a UHF sensor and a controller, wherein the controller communicates with the pulse current detector, the ultrasonic sensor, the transient ground voltage sensor and the UHF sensor respectively, and the controller is used to execute any one of the above methods.

[0117] By acquiring the data collected by the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor on the switch cabinet, the real-time collected data is obtained; the wavelet transform method is used to denoise the real-time collected data to obtain the final collected data; the K-means clustering algorithm is used to classify the historical collected data to obtain the classification results, and the classification results are used as training sets to train the BP neural network model to obtain the current BP neural network model; the current BP neural network model is used to process the final collected data to obtain the current partial discharge level of the switch cabinet. The joint detection of partial discharge is realized, which solves the problem that most of the existing solutions are single detection methods. Since only two detection methods are used, the frequency band coverage is insufficient, which easily leads to low accuracy of the detection results.

[0118] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0119] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0120] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0121] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0122] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0123] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0124] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0125] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0126] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0127] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0128] 1) The detection method of partial discharge capability of the switch cabinet of the present application obtains the data collected by the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor on the switch cabinet to obtain real-time collected data; uses the wavelet transform method to denoise the real-time collected data to obtain the final collected data; uses the K-means clustering algorithm to classify the historical collected data to obtain the classification results, and uses the classification results as a training set to train the BP neural network model to obtain the current BP neural network model; uses the current BP neural network model to process the final collected data to obtain the current partial discharge level of the switch cabinet. The joint detection of partial discharge is realized, that is, the problem that most of the existing solutions are single detection methods, and because only two detection methods are used, the frequency band coverage is insufficient, which easily leads to low accuracy of the detection results.

[0129] 2) The detection device of the partial discharge capability of the switch cabinet of the present application obtains the data collected by the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor on the switch cabinet to obtain real-time collected data; uses the wavelet transform method to denoise the real-time collected data to obtain the final collected data; uses the K-means clustering algorithm to classify the historical collected data to obtain the classification results, and uses the classification results as a training set to train the BP neural network model to obtain the current BP neural network model; uses the current BP neural network model to process the final collected data to obtain the current partial discharge level of the switch cabinet. The joint detection of partial discharge is realized, that is, the problem that most of the existing solutions are single detection methods, and because only two detection methods are used, the frequency band coverage is insufficient, which easily leads to low accuracy of the detection results.

[0130] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting partial discharge capability of a switch cabinet, characterized in that: include: Acquire the data collected by the pulse current detector, ultrasonic sensor, transient ground voltage sensor and ultra-high frequency sensor on the switch cabinet to obtain real-time collected data, wherein the ultra-high frequency sensor is used to collect the ultra-high frequency data collected by the pulse current detector, the ultrasonic sensor and the transient ground voltage sensor; Using wavelet transform method to perform denoising on the real-time collected data to obtain final collected data; Using K-means clustering algorithm to classify historical collected data to obtain classification results, and using the classification results as training sets, using the training sets to train the BP neural network model to obtain the current BP neural network model; The final collected data is processed using the current BP neural network model to obtain a current partial discharge level of the switch cabinet, wherein the current partial discharge level represents a real-time partial discharge capability of the switch cabinet; The final collected data is processed using the current BP neural network model to obtain the current partial discharge level of the switch cabinet, including: Using the current BP neural network model to process the collected data of the pulse current detector in the final collected data to obtain a first partial discharge score; Using the current BP neural network model to process the collected data of the ultrasonic sensor in the final collected data to obtain a second partial discharge score; Using the current BP neural network model to process the collected data of the transient ground voltage sensor in the final collected data to obtain a third partial discharge score; Using the current BP neural network model to process the collected data of the ultra-high frequency sensor in the final collected data to obtain a fourth partial discharge score; A weighted sum process is performed on the first partial discharge score, the second partial discharge score, the third partial discharge scores and the fourth partial discharge score to obtain a final partial discharge score, and the current partial discharge level is determined according to a range in which the current partial discharge level is located.

2. The method according to claim 1, characterized in that The current BP neural network model is used to process the collected data of the pulse current detector in the final collected data to obtain a first partial discharge score, including: Using the discharge amount, discharge times and phase in the final collected data as inputs related to the collected data of the current BP neural network model and the pulse current detector; The output of the current BP neural network model related to the collected data of the pulse current detector is used as the first partial discharge score.

3. The method according to claim 1, characterized in that The current BP neural network model is used to process the collected data of the ultrasonic sensor in the final collected data to obtain a second partial discharge score, including: Using the effective value, maximum value, first correlation and second correlation of the collected data of the ultrasonic sensor in the final collected data as inputs related to the current BP neural network model and the collected data of the ultrasonic sensor, the first correlation represents the correlation between the collected signal of the ultrasonic sensor and a first signal frequency threshold, the second correlation represents the correlation between the collected signal of the ultrasonic sensor and a second signal frequency threshold, and the second signal frequency threshold is greater than the first signal frequency threshold; The output of the current BP neural network model related to the collected data of the ultrasonic sensor is used as the second partial discharge score.

4. The method according to claim 1, characterized in that: The current BP neural network model is used to process the collected data of the transient voltage sensor in the final collected data to obtain a third partial discharge score: Using the transient ground voltage return value of the collected data of the transient ground voltage sensor in the final collected data as the input of the current BP neural network model related to the collected data of the transient ground voltage sensor; The output of the current BP neural network model related to the collected data of the transient voltage sensor is used as the third partial discharge score.

5. The method according to claim 1, characterized in that The current BP neural network model is used to process the collected data of the ultra-high frequency sensor in the final collected data to obtain a fourth partial discharge score: Using the effective value, maximum value and frequency of the collected data of the ultra-high frequency sensor in the final collected data as inputs related to the current BP neural network model and the collected data of the ultra-high frequency sensor; The output of the current BP neural network model related to the collected data of the ultra-high frequency sensor is used as the fourth partial discharge score.

6. The method according to any one of claims 1 to 5, characterized in that The current partial discharge level is one of the following: a first partial discharge level, a second partial discharge level, and a third partial discharge level that characterize the real-time partial discharge capability of the switch cabinet in descending order. After the final collected data is processed using the current BP neural network model to obtain the current partial discharge level of the switch cabinet, the method further includes: When the current partial discharge level is the first partial discharge level, determining that the real-time partial discharge capability of the switch cabinet is qualified; When the current partial discharge level is the second partial discharge level, generating first prompt information to prompt that the switch cabinet needs to be inspected within a preset time; When the current partial discharge level is the third partial discharge level, second prompt information is generated to prompt that the switch cabinet needs to be repaired.

7. A device for detecting partial discharge capability of a switch cabinet, characterized in that: include: An acquisition unit is used to acquire the data collected by the pulse current detector, the ultrasonic sensor, the transient ground voltage sensor and the ultra-high frequency sensor on the switch cabinet to obtain real-time collected data, wherein the ultra-high frequency sensor is used to collect the ultra-high frequency data collected by the pulse current detector, the ultrasonic sensor and the transient ground voltage sensor; A first processing unit is used to perform denoising on the real-time collected data by using a wavelet transform method to obtain final collected data; The second processing unit is used to classify the historical collected data using a K-means clustering algorithm to obtain a classification result, and use the classification result as a training set to train the BP neural network model to obtain a current BP neural network model; A third processing unit is used to process the final collected data using the current BP neural network model to obtain a current partial discharge level of the switch cabinet, wherein the current partial discharge level represents a real-time partial discharge capability of the switch cabinet; The third processing unit includes a first processing module, a second processing module, a third processing module, a fourth processing module and a fifth processing module, wherein the first processing module is used to process the collected data of the pulse current detector in the final collected data by using the current BP neural network model to obtain a first partial discharge score; The second processing module is used to process the collected data of the ultrasonic sensor in the final collected data using the current BP neural network model to obtain a second partial discharge score; The third processing module is used to process the collected data of the transient ground voltage sensor in the final collected data by using the current BP neural network model to obtain a third partial discharge score; The fourth processing module is used to use the current BP neural network model to process the collected data of the ultra-high frequency sensor in the final collected data to obtain a fourth partial discharge score; the fifth processing module is used to perform weighted sum processing on the first partial discharge score, the second partial discharge score, the third partial discharge scores and the fourth partial discharge score to obtain a final partial discharge score, and determine the current partial discharge level according to the range of the current partial discharge level.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 6.

9. A detection system for partial discharge capability of a switch cabinet, characterized in that: include: A pulse current detector, an ultrasonic sensor, a transient ground voltage sensor, a UHF sensor and a controller, wherein the controller communicates with the pulse current detector, the ultrasonic sensor, the transient ground voltage sensor and the UHF sensor respectively, and the controller is used to execute the method described in any one of claims 1 to 6.

Images