Partial discharge and positioning device based on nuclear phase hole of switch cabinet

Abstract

The invention discloses a partial discharge and positioning device based on a nuclear phase hole of a switch cabinet, which is characterized in that the nuclear phase hole of the switch cabinet is connected through each link and then forms a path with a cable, and when an insulation defect is generated in the cable, an excited traveling wave signal can be transmitted along the cable and finally transmitted to the nuclear phase hole. The defect positioning device is inserted into a nuclear phase hole through three pins to capture traveling wave signals caused by insulation defects in the cable, and signal quality is improved through a series of signal processing steps such as filtering and amplifying. Subsequently, deep analysis is carried out on the processed traveling wave signals, a PRPD atlas is generated to judge the type of partial discharge so as to carry out defect type qualification, a defect type positioning atlas is further generated to position the defect position of the cable, and a defect distance is output, so that powerful technical support is provided for maintenance of a power system.

Description

Technical Field

[0001] The present application relates to the field of intelligent positioning, and more specifically, to a partial discharge and positioning device based on a core phase hole of a switch cabinet. Background Art

[0002] With the development of power systems, the safe operation of power equipment has become one of the key factors to ensure the stability of the power grid. As an important electrical equipment in the power system, the safety and reliability of switchgear directly affect the normal operation of the entire power system. Partial discharge (PD) is a phenomenon in which the insulation material inside high-voltage electrical equipment undergoes local breakdown. It is one of the important reasons for the degradation of insulation performance, accelerated aging and even failure of electrical equipment. Therefore, timely and effective monitoring and positioning of partial discharge is of great significance for preventing accidents and improving the safe operation level of the power system.

[0003] However, although traditional partial discharge detection methods such as ultrasonic detection and high-frequency current transformer (HFCT) can provide certain detection information, it may be difficult to effectively capture weak partial discharge signals due to insufficient sensor sensitivity or large environmental noise interference, affecting the accuracy of detection. In addition, traditional partial discharge positioning methods mainly rely on the time difference method (TDOA) or amplitude comparison method. These methods are greatly affected by factors such as multipath effects and signal attenuation, which can easily lead to increased positioning errors, especially in complex electromagnetic environments, making it difficult to achieve high-precision positioning. In addition, traditional switchgear partial discharge detection and positioning methods usually require experienced technicians to perform on-site operations and interpret results, and have a low degree of automation, which not only increases labor costs, but may also lead to misjudgments due to human factors, affecting the accuracy of partial discharge positioning, and making it difficult to meet the high requirements of modern power systems for equipment status monitoring and maintenance.

[0004] Therefore, an optimized switch cabinet partial discharge and positioning device is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. An embodiment of the present application provides a partial discharge and positioning device based on the core phase hole of the switch cabinet, which forms a passage with the cable by connecting the core phase hole of the switch cabinet through various links. When an insulation defect occurs in the cable, the excited traveling wave signal will propagate along the cable and finally be transmitted to the core phase hole. The defect positioning device is inserted into the core phase hole through three pins to capture the traveling wave signal caused by the insulation defect inside the cable, and undergoes a series of signal processing steps, such as filtering, amplification, etc., to improve the signal quality. Subsequently, by conducting an in-depth analysis of the processed traveling wave signal, a PRPD spectrum is generated to determine the type of partial discharge for defect type characterization, and a defect type positioning spectrum is further generated to locate the cable defect position, and the defect distance is output, thereby providing strong technical support for the maintenance of the power system.

[0006] According to one aspect of the present application, a partial discharge and positioning device based on a core phase hole of a switch cabinet is provided, comprising: A traveling wave signal acquisition module, used for acquiring traveling wave signals; A traveling wave signal preprocessing module, used for filtering and amplifying the traveling wave signal to obtain a preprocessed traveling wave signal; A signal analysis module, used for performing signal analysis on the pre-processed traveling wave signal to obtain a PRPD spectrum of the traveling wave signal; A defect type qualitative module, used for qualitatively determining the defect type based on the traveling wave signal PRPD spectrum to obtain a defect type positioning spectrum; The cable defect location positioning module is used to determine the location data of the cable defect location based on the defect type positioning map.

[0007] Compared with the prior art, the present application provides a partial discharge and positioning device based on the core phase hole of the switch cabinet, which forms a path with the cable by connecting the core phase hole of the switch cabinet through various links. When an insulation defect occurs in the cable, the excited traveling wave signal will propagate along the cable and finally be transmitted to the core phase hole. The defect positioning device is inserted into the core phase hole through three pins to capture the traveling wave signal caused by the insulation defect inside the cable, and undergoes a series of signal processing steps, such as filtering and amplification, to improve the signal quality. Subsequently, by conducting an in-depth analysis of the processed traveling wave signal, a PRPD spectrum is generated to determine the type of partial discharge for defect type characterization, and a defect type positioning spectrum is further generated to locate the cable defect position, and the defect distance is output, thereby providing strong technical support for the maintenance of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0009] Figure 1 It is a block diagram of a partial discharge and positioning device based on a core phase hole of a switch cabinet according to an embodiment of the present application; Figure 2 A schematic diagram of data flow of a partial discharge and positioning device based on a core phase hole of a switch cabinet according to an embodiment of the present application; Figure 3 It is a block diagram of a defect type qualitative module in a partial discharge and positioning device based on a core phase hole of a switch cabinet according to an embodiment of the present application. DETAILED DESCRIPTION

[0010] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0011] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0012] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0013] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0014] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0015] Traditional partial discharge detection methods such as ultrasonic detection and high-frequency current transformer (HFCT) can provide certain detection information, but due to insufficient sensor sensitivity or large environmental noise interference, weak partial discharge signals are difficult to be effectively captured, affecting the accuracy of detection. In addition, traditional partial discharge positioning methods mainly rely on the time difference method (TDOA) or amplitude comparison method. These methods are greatly affected by factors such as multipath effects and signal attenuation, which can easily lead to increased positioning errors, especially in complex electromagnetic environments, making it difficult to achieve high-precision positioning. In addition, traditional switch cabinet partial discharge detection and positioning methods usually require experienced technicians to perform on-site operations and interpret results, and have a low degree of automation, which not only increases labor costs, but may also lead to misjudgments due to human factors, affecting the accuracy of partial discharge positioning, and making it difficult to meet the high requirements of modern power systems for equipment status monitoring and maintenance.

[0016] In recent years, with the development of information technology and intelligent algorithms, online monitoring technology for partial discharge based on traveling wave signals has gradually become a research hotspot. Traveling wave signals refer to high-frequency pulse signals generated by partial discharge. These signals propagate between conductors of power equipment or between conductors and the ground, and can carry information about the location and characteristics of partial discharge. Based on this, in the technical solution of this application, a partial discharge and positioning device based on the core phase hole of the switch cabinet is proposed, aiming to solve the problems existing in the prior art.

[0017] In the technical solution of the present application, a partial discharge and positioning device based on the core phase hole of the switch cabinet is proposed. Figure 1 It is a block diagram of a partial discharge and positioning device based on a core phase hole of a switch cabinet according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of a partial discharge and positioning device based on a core phase hole of a switch cabinet according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to the embodiment of the present application, a partial discharge and positioning device 300 based on the core phase hole of the switch cabinet includes: a traveling wave signal acquisition module 310, which is used to collect traveling wave signals; a traveling wave signal preprocessing module 320, which is used to filter and amplify the traveling wave signals to obtain preprocessed traveling wave signals; a signal analysis module 330, which is used to perform signal analysis on the preprocessed traveling wave signals to obtain a traveling wave signal PRPD spectrum; a defect type qualitative module 340, which is used to qualitatively characterize the defect type based on the traveling wave signal PRPD spectrum to obtain a defect type positioning spectrum; a cable defect position positioning module 350, which is used to determine the positioning data of the cable defect position based on the defect type positioning spectrum.

[0018] In particular, the traveling wave signal acquisition module 310 is used to collect traveling wave signals. The traveling wave signals are excited by insulation defects inside the cable. When there are insulation defects inside the cable, local discharges will be generated at the defects, and these local discharges will excite traveling wave signals. It should be understood that when the traveling wave signal propagates in the cable, it will carry information about the location of the defect. By analyzing the propagation time and path of the traveling wave signal, the location of the defect can be accurately located.

[0019] In particular, the traveling wave signal preprocessing module 320 is used to filter and amplify the traveling wave signal to obtain a preprocessed traveling wave signal. It should be understood that in actual environments, traveling wave signals are often interfered with by various noises, such as electromagnetic interference, environmental noise, etc. These noises will seriously affect the quality of the signal, resulting in a decrease in the accuracy of subsequent analysis. Traveling wave signals usually have a specific frequency range, and noise may exist in different frequency ranges. Through filtering, unnecessary frequency components can be removed and useful signals can be retained. In addition, since the traveling wave signal may attenuate during propagation, especially after long-distance transmission or multiple reflections, the amplitude of the signal may become very small. Amplifying the traveling wave signal can enhance the amplitude of the signal. By amplifying the signal, the sensitivity of the detection system can be improved, enabling it to detect weaker signals and thus discover smaller defects.

[0020] In particular, the signal analysis module 330 is used to perform signal analysis on the pre-processed traveling wave signal to obtain a PRPD spectrum of the traveling wave signal. It should be understood that partial discharge is a common manifestation of insulation defects inside the cable, and different types of partial discharge have different characteristics. In order to accurately determine the type of insulation defects inside the cable, it is necessary to perform a detailed analysis of the traveling wave signal to extract the characteristic information of the partial discharge. In particular, the PRPD spectrum is a commonly used partial discharge analysis tool, which generates a two-dimensional or three-dimensional spectrum by recording the phase and amplitude of the partial discharge pulse. In one example, in the two-dimensional spectrum, the abscissa represents the power phase (from 0° to 360°), and the ordinate represents the discharge intensity (such as voltage or current). Each point in the figure represents a partial discharge event, and the color or size of the point can be used to represent the intensity or frequency of the event. In the three-dimensional spectrum, in addition to the power phase and the discharge intensity, the third dimension is used to represent the frequency of partial discharge (i.e., the number of discharges occurring in a specific phase interval). This kind of graph can more intuitively show the density of discharge activities.

[0021] In particular, the defect type qualitative module 340 is used to qualitatively characterize the defect type based on the traveling wave signal PRPD spectrum to obtain a defect type location spectrum. Specifically, the process of determining the location data of the cable defect position based on the defect type location spectrum includes the following steps: first, extracting all the marked defect point information from the defect type location spectrum, including the local discharge type, PRPD spectrum characteristics and specific position coordinates of each defect point on the spectrum. Then, according to the propagation speed of the traveling wave signal (usually 60%-80% of the speed of light), combined with the timestamp of the traveling wave signal of each defect point reaching the core phase hole, the actual distance of each defect point relative to the core phase hole is calculated. The formula is: actual distance = traveling wave speed × time difference / 2, where the timestamp is half of the time difference from the generation of the traveling wave signal to the capture by the core phase hole. Finally, the calculated actual distance of each defect point is combined with the cable path map to generate a visualization of the cable defect position. And the specific location data, local discharge type and severity of each defect point are marked in the visualization map, so that maintenance personnel can intuitively understand the health status of the cable. In particular, if there are multiple core phase holes or multiple detection points, the positioning data of each defect point can be further corrected by multi-point positioning technology, such as triangulation positioning or multilateral positioning, to improve the positioning accuracy of the cable defect position.

[0022] In particular, in the above-mentioned partial discharge and positioning device based on the core phase hole of the switch cabinet, the step of qualitatively characterizing the defect type based on the traveling wave signal PRPD spectrum is crucial, because the partial discharge of the switch cabinet can be caused by a variety of different defects, such as air gap discharge, surface discharge, internal discharge, etc. Each discharge type has different characteristics on the PRPD spectrum. By analyzing the PRPD spectrum, the specific discharge type can be accurately identified, thereby helping maintenance personnel to better understand the nature of the fault. In addition, different discharge types cause different degrees of damage to cables and equipment. Through qualitative analysis, different types of defects can be prioritized, those defects that have a greater impact on the system can be given priority, and maintenance resources can be reasonably allocated. Not only that, through accurate characterization of defect types, potential fault sources can be discovered early, and preventive measures can be taken to prevent further deterioration of the fault, thereby extending the service life of the equipment and reducing maintenance costs.

[0023] Based on this, in the process of qualitatively characterizing the defect type, the technical concept of the present application is to extract the local traveling wave signal PRPD spectrum of each defect point from the traveling wave signal PRPD spectrum, and then introduce a data processing and analysis algorithm based on artificial intelligence and deep learning at the back end to analyze the local traveling wave signal PRPD spectrum of these defect points, so as to capture the PRPD spectrum characteristics of each defect point, and identify the local discharge types of different defect points based on these characteristics, and then mark the local discharge types of these defect points on the traveling wave signal PRPD spectrum to generate a defect type location spectrum, so as to facilitate the subsequent determination of the location data of the cable defect position to carry out corresponding fault handling and switch cabinet maintenance work. In this way, the defect type location spectrum can be automatically generated based on the traveling wave signal PRPD spectrum characteristics of different defect points, which helps to achieve more intelligent switch cabinet partial discharge detection and positioning, and provide a reliable basis for subsequent maintenance work. Specifically, in a specific example of the present application, such as Figure 3As shown, the defect type qualitative module 340 includes: a local traveling wave signal PRPD spectrum extraction unit 341, which is used to extract the local traveling wave signal PRPD spectrum of each defect point from the traveling wave signal PRPD spectrum to obtain a set of local traveling wave signal PRPD spectra of the defect point; a defect point PRPD spectrum feature extraction unit 342, which is used to extract spectrum features of the set of local traveling wave signal PRPD spectra of the defect point to obtain a set of defect point PRPD spectrum feature maps; a defect point PRPD space enhancement unit 343, which is used to perform spatial flexible feature attention enhancement processing on each defect point PRPD spectrum feature map in the set of defect point PRPD spectrum feature maps to obtain a set of defect point PRPD spatial dimension enhanced spectrum feature maps; a discharge type identification unit 344, which is used to perform discharge type identification based on the set of defect point PRPD spatial dimension enhanced spectrum feature maps to determine a set of local discharge types of the defect point, and mark the set of local discharge types of the defect point on the traveling wave signal PRPD spectrum to obtain the defect type positioning spectrum.

[0024] Specifically, the local traveling wave signal PRPD spectrum extraction unit 341 is used to extract the local traveling wave signal PRPD spectrum of each defect point from the traveling wave signal PRPD spectrum to obtain a set of local traveling wave signal PRPD spectra of the defect point. It should be understood that the traveling wave signal PRPD spectrum presents a global view, showing all local discharge signals of the entire cable segment. These signals may interfere with each other during the analysis process, affecting the accuracy of the analysis. By extracting the local traveling wave signal PRPD spectrum of each defect point, the analysis scope can be narrowed to the specific defect point, so as to analyze the characteristics of each defect point in more detail, reduce mutual interference, and improve the accuracy of analysis and positioning.

[0025] Specifically, the defect point PRPD spectrum feature extraction unit 342 is used to extract spectrum features from the set of defect point local traveling wave signal PRPD spectra to obtain a set of defect point PRPD spectrum feature maps. In a specific example of the present application, each defect point local traveling wave signal PRPD spectrum in the set of defect point local traveling wave signal PRPD spectrum is input into a spectrum feature extractor based on a hole convolutional neural network model to obtain a set of defect point PRPD spectrum feature maps. That is, each defect point local traveling wave signal PRPD spectrum in the set of defect point local traveling wave signal PRPD spectrum is input into a spectrum feature extractor based on a hole convolutional neural network model for feature mining, so as to extract the implicit feature distribution information in each defect point local traveling wave signal PRPD spectrum, thereby obtaining a set of defect point PRPD spectrum feature maps. It is worth mentioning that, since the traditional convolution operation may lose some small features, the hole convolution can extract more detailed features while maintaining spatial information, which is very important for the feature recognition of partial discharge.

[0026] Specifically, the defect point PRPD spatial enhancement unit 343 is used to perform spatial flexible feature attention enhancement processing on each defect point PRPD spectrum feature map in the set of defect point PRPD spectrum feature maps to obtain a set of defect point PRPD spatial dimension enhanced spectrum feature maps. It should be understood that in practical applications, local discharge may occur in different parts of the cable, and the discharge characteristics of each part may be different. However, in the case of complex background, the traditional feature extraction method may be difficult to distinguish the local discharge signal characteristics in different spaces. Therefore, in the technical solution of the present application, each defect point PRPD spectrum feature map in the set of defect point PRPD spectrum feature maps is further subjected to spatial flexible feature attention enhancement processing to obtain a set of defect point PRPD spatial dimension enhanced spectrum feature maps. In particular, the spatial flexible feature attention enhancement processing method can extract the fine features of the PRPD spectrum of each local discharge point, give differentiated attention levels to different areas in the feature map, and help the model to more accurately locate and identify specific targets to distinguish different discharge types (air gap discharge, surface discharge, etc.). Moreover, the spatial flexible feature attention enhancement processing can also weaken the influence of non-critical parts that are irrelevant to the subsequent discharge type recognition and labeling tasks to a certain extent, while highlighting the PRPD map features of key defect points that are relatively stable even under adverse conditions, thereby enhancing the model's ability to cope with various complex situations and providing a more accurate basis for the discharge type recognition task.

[0027] In an embodiment of the present application, the specific process of performing spatial flexible feature attention enhancement processing on each defect point PRPD spectrum feature map in the set of defect point PRPD spectrum feature maps includes: first, performing feature fine-grained spatial decoupling along the channel dimension on the defect point PRPD spectrum feature map to obtain a set of defect point PRPD spectrum pixel granularity feature vectors; obtaining more refined defect point PRPD spectrum feature information by performing feature fine-grained spatial decoupling along the channel dimension on the defect point PRPD spectrum feature map, this process is similar to decomposing the traveling wave signal PRPD spectrum into different frequency components, so that the fine features and local details of each local discharge point can be analyzed independently. Then, the spatial position of each defect point PRPD spectrum pixel granularity feature vector in the set of defect point PRPD spectrum pixel granularity feature vectors is annotated to obtain a set of defect point PRPD spectrum spatial position data; here, in order to maintain the spatial relationship between the defect point PRPD spectrum pixel granularity features of these local details and the original image, it is necessary to add the exact position information of each pixel-level feature vector in the original image. This kind of position annotation is very important for understanding the relative spatial relationship between the pixel granularity features of the defect point PRPD map, because even if two pixels have similar features, they may represent completely different meanings if they are located in different positions. Accordingly, maintaining good spatial structure information helps to accurately capture the subtle differences between the pixel granularity features of these defect point PRPD maps in subsequent processing, and provides a basis for the enhanced expression of key features. Furthermore, any two defect point PRPD map pixel granularity feature vectors in the set of the defect point PRPD map pixel granularity feature vectors are input into the semantic association score measurement network to obtain a set of semantic association score vectors between the pixel granularities of the defect point PRPD map; and based on the defect point PRPD map spatial position data of any two defect point PRPD map pixel granularity feature vectors in the set of the defect point PRPD map spatial position data, the spatial span value of the any two pixel granularity feature vectors is determined to obtain a set of spatial span values ​​of the defect point PRPD map. That is, on this basis, a specially trained network is used to estimate the semantic connection strength between the pixel granularity of any two defect point PRPD maps. This network learns to identify which types of local features tend to co-occur on the same type of discharge. For example, air gap discharge and surface discharge may have similarities in some local features, but they can be distinguished by semantic association scores.At the same time, it is also necessary to consider the actual distance between the two graph pixel granularity features in the feature map space, that is, to calculate their spatial span values. Combining the semantic association score with the spatial span value can help filter out the pixel granularity parts of the defect point PRPD graph that look similar but are actually irrelevant, further improving the expressiveness of the feature map and the ability to identify specific objects, that is, having a better recognition effect on different types of partial discharges. Further, based on the spatial span values ​​of each defect point PRPD graph in the set of the spatial span values ​​of the defect point PRPD graph, the semantic association score vectors between the pixel granularities of each defect point PRPD graph in the set of the semantic association score vectors between the pixel granularities of the defect point PRPD graph are subjected to semantic association attenuation adaptive attenuation modulation to obtain a set of semantic association score vectors between the pixel granularities of the defect point PRPD graph spatial attenuation. That is, based on the spatial span values ​​of each defect point PRPD graph in the set of the spatial span values ​​of the defect point PRPD graph, the semantic association score vectors between the pixel granularities of each defect point PRPD graph in the set of the semantic association score vectors between the pixel granularities of the defect point PRPD graph are subjected to semantic association attenuation adaptive attenuation modulation. When the pixel granularity features of the two defect point PRPD maps show strong semantic correlation and are close to each other, their importance will be maximized; on the contrary, if the two have high semantic similarity but are far apart, their influence will be appropriately weakened. Such dynamic adjustment ensures that the semantic association weight map between the spatial attenuation pixel granularity of the defect point PRPD map generated in the end can accurately reflect the most noteworthy part of the local traveling wave signal PRPD map of the Meigel defect point, providing a basis for the subsequent defect type identification and positioning. Then, after arranging the set of semantic association score vectors between the spatial attenuation pixel granularity of the defect point PRPD map as a semantic association score feature map between the spatial attenuation pixel granularity of the defect point PRPD map, the semantic association score feature map between the spatial attenuation pixel granularity of the defect point PRPD map is input into the channel dimension modulation layer based on the point convolution layer and the weight distribution network based on the Sigmoid function to obtain the semantic association weight map between the spatial attenuation pixel granularity of the defect point PRPD map; through the application of the Sigmoid function, it is ensured that all weights are in the range of 0 to 1, which is convenient for direct application to the original feature map. This helps to apply directly to the original feature map, and through element-by-element multiplication operations, selectively amplify the important defect point PRPD map pixel granularity feature information and suppress insignificant or noisy pixel granularity areas. This not only retains all the advantages of the original feature map, but also further highlights those defect point PRPD map pixel granularity information fragments that are considered to be the most critical for the current defect type identification and location tasks. Finally, the semantic association weight map between the spatial attenuation pixel granularity of the defect point PRPD map and the defect point PRPD map feature map is calculated by multiplying the position points to obtain the defect point PRPD spatial dimension enhancement map feature map.

[0028] Among them, the specific steps of inputting any two defect point PRPD map pixel granularity feature vectors in the set of defect point PRPD map pixel granularity feature vectors into the semantic association score measurement network include: cascading any two defect point PRPD map pixel granularity feature vectors in the set of defect point PRPD map pixel granularity feature vectors to obtain a defect point PRPD map pixel granularity cascade feature vector; multiplying the defect point PRPD map pixel granularity cascade feature vector with a weight matrix, and then adding the offset vector by position to obtain a semantic association score vector between defect point PRPD map pixel granularities.

[0029] More specifically, the specific steps of determining the spatial span value of any two pixel granularity feature vectors of the defect point PRPD map spatial position data based on the defect point PRPD map pixel granularity feature vectors in the set of the defect point PRPD map spatial position data include: calculating the Euclidean distance between the defect point PRPD map spatial position data of the any two defect point PRPD map pixel granularity feature vectors to obtain the defect point PRPD map spatial span value.

[0030] More specifically, based on the spatial span values ​​of each defect point PRPD map in the set of spatial span values ​​of the defect point PRPD map, the specific steps of performing semantic association attenuation adaptive attenuation modulation on the semantic association score vectors between pixel granularities of each defect point PRPD map in the set of semantic association score vectors between pixel granularities of the defect point PRPD map include: dividing the spatial span value of the defect point PRPD map by a double semantic association attenuation hyperparameter to obtain a semantic association attenuation modulation factor of the spatial span of the defect point PRPD map; calculating a natural exponential function value with the natural constant e as the base and the negative number of the semantic association attenuation modulation factor of the spatial span of the defect point PRPD map as the exponential power to obtain a spatial span class support modulation factor of the defect point PRPD map; calculating the positional division between the semantic association score vectors between pixel granularities of each corresponding defect point PRPD map and the spatial span class support modulation factor of the defect point PRPD map to obtain a semantic association score vector between pixel granularities of the defect point PRPD map spatial attenuation.

[0031] In summary, in the above embodiment, each defect point PRPD spectrum feature map in the set of defect point PRPD spectrum feature maps is subjected to spatial flexible feature attention enhancement processing to obtain a set of defect point PRPD spatial dimension enhanced spectrum feature maps, including: performing spatial flexible feature attention enhancement processing on each defect point PRPD spectrum feature map in the set of defect point PRPD spectrum feature maps using the following feature attention enhancement formula to obtain a set of defect point PRPD spatial dimension enhanced spectrum feature maps; wherein the feature attention enhancement formula is: ;in, is each defect point PRPD spectrum feature graph in the set of defect point PRPD spectrum feature graphs, It is the feature fine-grained spatial decoupling process. is the pixel granularity feature vector of each defect point PRPD map in the set of pixel granularity feature vectors of the defect point PRPD map, , is the number of channels of the PRPD spectrum characteristic diagram of the defect point, is a real number, for Spatial location data, , is the spatial position coordinate represented by the spatial position data, is a collection of spatial location data, and are any two defect point PRPD map pixel granularity feature vectors in the set of defect point PRPD map pixel granularity feature vectors, For vector concatenation, is the weight matrix, is the bias vector, for and The semantic association score vector between the pixel granularity of the defect point PRPD map, is the set of semantic association score vectors between pixel granularities of the defect point PRPD map, for Spatial location data, for Spatial location data, for and The space span value between is a set of spatial span values, is the semantic relevance decay hyperparameter, for and The semantic association score vector between the defect point PRPD map space attenuation pixel granularity, is the set of semantic association score vectors between the spatial attenuation pixel granularities of the defect point PRPD map, is a feature graph arrangement of a vector set, is a semantic association score feature map between the spatial attenuation pixel granularities of the defect point PRPD map obtained by arranging the set of semantic association score vectors between the spatial attenuation pixel granularities of the defect point PRPD map, It is the point convolution channel modulation processing, It is the feature map of semantic association scores between pixel granularities of the spatial attenuation of the PRPD map of the channel dimension modulation defect point. is the Sigmoid function, is the semantic association weight map between the spatial attenuation pixel granularity of the defect point PRPD map, For point-by-position multiplication, Enhance the atlas feature map for the PRPD spatial dimension of the defect point.

[0032] Specifically, the discharge type identification unit 344 is used to identify the discharge type based on the set of the defect point PRPD spatial dimension enhancement map feature map to determine the set of local discharge types of the defect point, and mark the set of local discharge types of the defect point on the traveling wave signal PRPD map to obtain the defect type positioning map. In a specific example of the present application, each defect point PRPD spatial dimension enhancement map feature map in the set of defect point PRPD spatial dimension enhancement map feature maps is respectively input into the classifier-based discharge type identification module to obtain a set of local discharge types of the defect point. That is, the PRPD spatial dimension enhancement map features of each defect point are used to perform classification processing respectively, so as to identify the local discharge type of each defect point. In one example, in the local discharge detection of power equipment, the discharge type is generally used to describe different types of local discharge phenomena. Common discharge types generally include: internal discharge, surface discharge, suspended discharge, etc. Internal discharge occurs inside the insulating material, usually due to bubbles, cracks or other defects in the insulating material. The PRPD spectrum shows that the discharge pulse is concentrated in a certain interval of the phase and has a high amplitude. Surface discharge occurs on the surface of the insulating material, usually due to surface contamination, moisture or surface unevenness. The PRPD spectrum shows that the discharge pulse is widely distributed, covering multiple phase intervals, and has a medium amplitude. Suspended discharge occurs between the conductor and the grounding component, usually due to loose conductors or poor contact. The PRPD spectrum shows that the discharge pulse is concentrated in a certain interval of the phase and has a high amplitude, similar to internal discharge. Furthermore, the set of local discharge types of the defect points is marked on the traveling wave signal PRPD spectrum to obtain the defect type location spectrum, so as to facilitate the subsequent determination of the location data of the cable defect position to carry out corresponding fault handling and switch cabinet maintenance. In this way, the defect type location spectrum can be automatically generated based on the traveling wave signal PRPD spectrum characteristics of different defect points, which helps to achieve more intelligent switch cabinet partial discharge detection and positioning, and provide a reliable basis for subsequent maintenance work.

[0033] Preferably, the defect point PRPD spatial dimension enhancement spectrum feature map is input into a classifier-based discharge type identification module to obtain the defect point local discharge type, including: Determine the number of zero eigenvalues ​​in the defect point PRPD spatial dimension enhancement map feature map, and respectively calculate the reciprocal of the logarithm of the number of zero eigenvalues ​​with base two and the exponential value of the reciprocal of the number of zero eigenvalues ​​with base natural constants to obtain a first defect point PRPD map feature representation value and a second defect point PRPD map feature representation value; Expand the defect point PRPD spatial dimension enhancement map feature map into a defect point PRPD spatial dimension enhancement map feature vector; Calculate each eigenvalue of the defect point PRPD space dimension enhancement map feature vector as a power function with the inverse of the number of zero eigenvalues ​​as an exponent to obtain the defect point PRPD space dimension enhancement map structure state vector; Performing a dot multiplication of the defect point PRPD space dimension enhancement map structure state vector and the first defect point PRPD space dimension enhancement map representation value to obtain a defect point PRPD space dimension enhancement map information representation vector; The autocorrelation matrix of the defect point PRPD space dimension reinforcement map structure state vector is respectively multiplied by the second defect point PRPD space dimension reinforcement map representation value and weight hyperparameter to obtain the defect point PRPD space dimension reinforcement map regression understanding matrix; Performing matrix multiplication of the defect point PRPD space dimension enhancement map information representation vector and the defect point PRPD space dimension enhancement map regression understanding matrix to obtain an optimized defect point PRPD space dimension enhancement map feature vector; and The optimized defect point PRPD spatial dimension enhanced atlas feature vector is input into a discharge type recognition module based on a classifier to obtain the defect point local discharge type.

[0034] That is, the first defect point PRPD space dimension enhancement map representation value and the second defect point PRPD space dimension enhancement map representation value are expressed as: ;in, is the number of zero eigenvalues ​​in the PRPD spatial dimension enhancement spectrum feature map of the defect point, is the representation value of the PRPD spatial dimension enhancement map of the first defect point, is the value of the PRPD spatial dimension enhancement spectrum representation of the second defect point, represents the logarithm operation with base 2, Represents exponential operation; And the optimized defect point PRPD spatial dimension enhancement map feature vector It is expressed as: ; wherein the defect point PRPD spatial dimension strengthens the atlas feature vector is a row vector, It represents a power function that calculates each eigenvalue of the eigenvector of the PRPD spatial dimension enhancement spectrum of the defect point with the inverse of the number of zero eigenvalues ​​as the exponent, is the representation vector of the defect point PRPD spatial dimension enhanced graph information, is the weight hyperparameter, is the defect point PRPD spatial dimension reinforcement map regression understanding matrix, represents matrix multiplication, represents positional multiplication, It is the optimized defect point PRPD space dimension enhancement map feature vector.

[0035] Here, the defect point PRPD spatial dimension enhanced spectrum feature map represents the spectrum semantic coding features of the local traveling wave signal PRPD spectrum of each defect point. After inputting it into the feature attention enhancement module based on spatial flexible modulation, its feature significance in the spatial dimension will be enhanced, but this will also weaken its feature distribution consistency in the spatial dimension, making the defect point PRPD spatial dimension enhanced spectrum feature map have a long-range class regression understanding representation imbalance, affecting the accuracy of the classification result.

[0036] Based on this, the applicant of the present application realizes effective long-range modeling of the defect point PRPD spatial dimensional enhanced spectrum feature map under the field linear complexity by modeling the vector field zero dimension of the defect point PRPD spatial dimensional enhanced spectrum feature vector obtained after the expansion of the defect point PRPD spatial dimensional enhanced spectrum feature map as the global field redundant dependency. In this way, in the feature regression process of the defect point PRPD spatial dimensional enhanced spectrum feature map, the structured correlation self-distillation between information representation and class understanding is performed based on the selective state space of feature distribution, so as to maintain the overall information complexity of the feature distribution of the defect point PRPD spatial dimensional enhanced spectrum feature map while promoting the long-range dynamic class understanding balance of its feature convergence, so as to improve the class understanding consistency between the classification regression and feature extraction processes, and improve the accuracy of the defect point local discharge type obtained by the discharge type recognition module based on the classifier input of the defect point PRPD spatial dimensional enhanced spectrum feature map. In this way, the local discharge types of these defect points are marked on the traveling wave signal PRPD spectrum to generate a defect type positioning spectrum, so as to more accurately determine the positioning data of the cable defect position in the future to carry out corresponding fault processing and switch cabinet maintenance work, and provide an intelligent level of switch cabinet partial discharge detection and positioning.

[0037] In particular, the cable defect location positioning module 350 is used to determine the location data of the cable defect location based on the defect type location map, thereby providing powerful technical support for the maintenance of the power system by accurately locating the defect location.

[0038] As described above, the partial discharge and positioning device 300 based on the core phase hole of the switch cabinet according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a partial discharge and positioning algorithm based on the core phase hole of the switch cabinet. In a possible implementation, the partial discharge and positioning device 300 based on the core phase hole of the switch cabinet according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the partial discharge and positioning device 300 based on the core phase hole of the switch cabinet can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the partial discharge and positioning device 300 based on the core phase hole of the switch cabinet can also be one of the many hardware modules of the wireless terminal.

[0039] Alternatively, in another example, the partial discharge and positioning device 300 based on the core phase hole of the switch cabinet and the wireless terminal may also be separate devices, and the partial discharge and positioning device 300 based on the core phase hole of the switch cabinet may be connected to the wireless terminal via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0040] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A partial discharge and positioning device based on the core phase hole of a switch cabinet, characterized in that: include: A traveling wave signal acquisition module, used for acquiring traveling wave signals; A traveling wave signal preprocessing module, used for filtering and amplifying the traveling wave signal to obtain a preprocessed traveling wave signal; A signal analysis module, used for performing signal analysis on the pre-processed traveling wave signal to obtain a PRPD spectrum of the traveling wave signal; A defect type qualitative module, used for qualitatively determining the defect type based on the traveling wave signal PRPD spectrum to obtain a defect type positioning spectrum; The cable defect location positioning module is used to determine the location data of the cable defect location based on the defect type positioning map.

2. The partial discharge and positioning device based on the core phase hole of the switch cabinet according to claim 1 is characterized in that: The defect type qualitative module includes: A local traveling wave signal PRPD spectrum extraction unit, used to extract the local traveling wave signal PRPD spectrum of each defect point from the traveling wave signal PRPD spectrum to obtain a set of local traveling wave signal PRPD spectra of the defect point; A defect point PRPD spectrum feature extraction unit, used for extracting spectrum features from the set of local traveling wave signal PRPD spectra of the defect point to obtain a set of defect point PRPD spectrum feature maps; A defect point PRPD spatial enhancement unit, used for performing spatial flexible feature attention enhancement processing on each defect point PRPD spectrum feature map in the set of defect point PRPD spectrum feature maps to obtain a set of defect point PRPD spatial dimension enhanced spectrum feature maps; A discharge type identification unit is used to identify the discharge type based on the set of PRPD spatial dimension enhancement spectrum feature maps of the defect point to determine the set of local discharge types of the defect point, and mark the set of local discharge types of the defect point on the traveling wave signal PRPD spectrum to obtain the defect type positioning spectrum.

3. The partial discharge and positioning device based on the core phase hole of the switch cabinet according to claim 2 is characterized in that: The defect point PRPD spectrum feature extraction unit is used to: input each defect point local traveling wave signal PRPD spectrum in the set of defect point local traveling wave signal PRPD spectrum into a spectrum feature extractor based on a void convolutional neural network model to obtain a set of defect point PRPD spectrum feature maps.

4. The partial discharge and positioning device based on the core phase hole of the switch cabinet according to claim 3 is characterized in that: The defect point PRPD space enhancement unit comprises: A feature fine-grained spatial decoupling subunit is used to perform feature fine-grained spatial decoupling along the channel dimension on the defect point PRPD map feature map to obtain a set of pixel granularity feature vectors of the defect point PRPD map; A spatial position labeling subunit is used to label the spatial position of each defect point PRPD map pixel granularity feature vector in the set of defect point PRPD map pixel granularity feature vectors to obtain a set of defect point PRPD map spatial position data; A semantic association score measurement subunit, used for inputting any two defect point PRPD map pixel granularity feature vectors in the set of defect point PRPD map pixel granularity feature vectors into a semantic association score measurement network to obtain a set of semantic association score vectors between defect point PRPD map pixel granularities; A spatial span calculation subunit, used to determine the spatial span value of any two pixel granularity feature vectors of the defect point PRPD spectrum based on the defect point PRPD spectrum spatial position data of the pixel granularity feature vectors of any two defect point PRPD spectrum in the set of the defect point PRPD spectrum spatial position data to obtain a set of defect point PRPD spectrum spatial span values; A semantic association attenuation adaptive attenuation modulation subunit is used to perform semantic association attenuation adaptive attenuation modulation on each defect point PRPD map pixel granularity semantic association score vector in the set of the defect point PRPD map pixel granularity semantic association score vectors based on each defect point PRPD map spatial span value in the set of the defect point PRPD map spatial span value to obtain a set of defect point PRPD map spatial attenuation pixel granularity semantic association score vectors; A weight allocation subunit is used to arrange the set of semantic association score vectors between spatial attenuation pixel granularities of the defect point PRPD map into a semantic association score feature map between spatial attenuation pixel granularities of the defect point PRPD map, and then input the semantic association score feature map between spatial attenuation pixel granularities of the defect point PRPD map into a channel dimension modulation layer based on a point convolution layer and a weight allocation network based on a sigmoid function to obtain a semantic association weight map between spatial attenuation pixel granularities of the defect point PRPD map; The defect point PRPD spatial dimension enhancement subunit is used to calculate the semantic association weight map between the spatial attenuation pixel granularity of the defect point PRPD map and the defect point PRPD map feature map by position point multiplication to obtain the defect point PRPD spatial dimension enhancement map feature map.

5. The partial discharge and positioning device based on the core phase hole of the switch cabinet according to claim 4 is characterized in that: The semantic association score measurement subunit is used to: Cascading any two defect point PRPD map pixel granularity feature vectors in the set of defect point PRPD map pixel granularity feature vectors to obtain a defect point PRPD map pixel granularity cascade feature vector; After the pixel granularity cascade feature vector of the defect point PRPD map is multiplied by the weight matrix, it is added to the bias vector by position to obtain the semantic association score vector between the pixel granularities of the defect point PRPD map.

6. The partial discharge and positioning device based on the core phase hole of the switch cabinet according to claim 5 is characterized in that: The spatial span calculation subunit is used to calculate the Euclidean distance between the defect point PRPD map spatial position data of the pixel granularity feature vectors of any two defect point PRPD maps to obtain the defect point PRPD map spatial span value.

7. The partial discharge and positioning device based on the core phase hole of the switch cabinet according to claim 6 is characterized in that: The semantic association attenuation adaptive attenuation modulation subunit is used to: Divide the defect point PRPD map space span value by a doubled semantic association attenuation hyperparameter to obtain a defect point PRPD map space span semantic association attenuation modulation factor; Calculate the natural exponential function value with the natural constant e as the base and the negative number of the semantic association attenuation modulation factor of the defect point PRPD spectrum space span as the exponential power to obtain the defect point PRPD spectrum space span class support modulation factor; Calculate the positional division between the semantic association score vector between pixel granularities of the defect point PRPD map and the spatial span class support modulation factor of the defect point PRPD map corresponding to each group to obtain the semantic association score vector between the spatial attenuation pixel granularities of the defect point PRPD map.

8. The partial discharge and positioning device based on the core phase hole of the switch cabinet according to claim 7 is characterized in that: The discharge type identification unit is used to: Inputting each defect point PRPD spatial dimension enhancement spectrum feature map in the set of defect point PRPD spatial dimension enhancement spectrum feature maps into a discharge type recognition module based on a classifier to obtain a set of defect point partial discharge types; The set of local discharge types of the defect points is marked on the traveling wave signal PRPD spectrum to obtain the defect type positioning spectrum.

Images