Busway detection apparatus motion control system and method

Abstract

The application discloses a bus duct detection equipment motion control system and method, and belongs to the technical field of power distribution equipment, wherein the system comprises: a monitoring configuration module for obtaining a multi-dimensional sensing array; a fault framing module for obtaining a fault detection interval; an equipment matching module for obtaining a plurality of real-time detection equipment; a strategy optimization module for obtaining a target multi-modal diagnostic path of a target detection equipment; and a fault recognition module for outputting a real-time fault node and a real-time fault type. The application solves the technical problems of incomplete coverage of a single detection method due to the complex structure of the bus duct, thereby leading to inaccurate fault area positioning and low detection efficiency, and achieves the technical effects of effectively improving the positioning accuracy of the fault area, realizing efficient and accurate fault holographic recognition through multi-sensor data fusion, dynamic path planning and intelligent switching of equipment.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of power distribution equipment, in particular to a bus duct detection equipment motion control system and method. BACKGROUND

[0002] With the rapid development of power systems, the monitoring and fault detection of bus ducts, as an important power distribution equipment, are particularly important. However, the existing bus duct detection technology often relies on manual inspection or single sensor detection methods, which have certain limitations in practical application. First, the traditional detection method is difficult to accurately locate the fault area in real time, and has poor adaptability to obstacles or spatial layout in complex environments, and cannot intelligently adjust the detection path of the equipment according to the actual situation. Secondly, the single sensor monitoring method often cannot comprehensively reflect the fault information of the bus duct, causing the risk of missed detection or false detection. Therefore, an efficient and intelligent detection equipment is needed, which can realize accurate positioning and intelligent detection of bus duct faults through multi-sensor data fusion, real-time path planning and equipment motion control. SUMMARY

[0003] The application provides a bus duct detection equipment motion control system and method, aiming to solve the technical problem of inaccurate fault area positioning and low detection efficiency caused by incomplete coverage of single detection method due to the complex structure of bus ducts, and achieve the technical effect of efficient and accurate fault holographic recognition through multi-sensor data fusion, dynamic path planning and intelligent switching of equipment.

[0004] The first aspect of the application provides a bus duct detection equipment motion control system, which comprises: a monitoring configuration module for multi-dimensional monitoring coverage configuration of a bus power distribution network according to a monitoring scene, obtaining a multi-dimensional sensor array; a fault framing module for a patrol control center to preliminarily frame a fault range according to a multi-dimensional monitoring time array returned by the multi-dimensional sensor array, obtaining a fault detection interval, wherein the fault detection interval is identified by a real-time characteristic parameter; a device matching module for matching a hot standby detection device according to the fault detection interval and the real-time characteristic parameter, obtaining a plurality of real-time detection devices; a strategy optimization module for mobile strategy optimization according to the fault detection interval and a plurality of real-time device positions of the plurality of real-time detection devices, to screen a target multi-modal diagnostic path of a target detection device; and a fault recognition module for driving the target detection device to perform fault holographic recognition in the fault detection interval according to the real-time characteristic parameter after controlling the target detection device to move to the fault detection interval by the target multi-modal diagnostic path, and outputting a real-time fault node and a real-time fault type.

[0005] In another aspect of the present application, a bus duct detection device motion control method is provided, which comprises: performing multi-dimensional monitoring coverage configuration on a bus power distribution network according to a monitoring scene to obtain a multi-dimensional sensing array; performing preliminary fault range framing by a patrol control center according to a multi-dimensional monitoring time sequence array returned by the multi-dimensional sensing array to obtain a fault detection interval, wherein the fault detection interval is identified by a real-time characteristic parameter; performing hot backup detection device matching according to the fault detection interval and the real-time characteristic parameter to obtain a plurality of real-time detection devices; performing mobile strategy optimization according to the fault detection interval and a plurality of real-time device positions of the plurality of real-time detection devices to screen a target multi-modal diagnostic path of a target detection device; and after controlling the target detection device to move to the fault detection interval by using the target multi-modal diagnostic path, driving the target detection device to perform fault holographic identification in the fault detection interval according to the real-time characteristic parameter to output a real-time fault node and a real-time fault type.

[0006] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0007] Due to the multi-dimensional monitoring coverage configuration on the bus power distribution network according to the monitoring scene by the monitoring configuration module to obtain the multi-dimensional sensing array, the preliminary fault range framing by the patrol control center according to the multi-dimensional monitoring time sequence array returned by the multi-dimensional sensing array to obtain the fault detection interval, the identification of the fault detection interval by the real-time characteristic parameter, the hot backup detection device matching according to the fault detection interval and the real-time characteristic parameter to obtain the plurality of real-time detection devices, the mobile strategy optimization according to the fault detection interval and the plurality of real-time device positions of the plurality of real-time detection devices to screen the target multi-modal diagnostic path of the target detection device, and the driving of the target detection device to perform the fault holographic identification in the fault detection interval according to the real-time characteristic parameter to output the real-time fault node and the real-time fault type after controlling the target detection device to move to the fault detection interval by using the target multi-modal diagnostic path, the technical problems of the incomplete coverage of a single detection method due to the complex bus duct structure and the low positioning accuracy and detection efficiency of a fault area are solved, and the technical effects of effectively improving the positioning accuracy of a fault area and realizing efficient and accurate fault holographic identification through multi-sensor data fusion, dynamic path planning and intelligent switching of devices are achieved.

[0008] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the specific embodiments of the present application can be implemented in accordance with the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0009] Figure 1 A structural schematic diagram of a bus duct detection equipment motion control system is provided for the embodiments of the present application.

[0010] Figure 2 A flowchart of a bus duct detection equipment motion control method is provided for the embodiments of the present application.

[0011] Reference signs: monitoring configuration module 11, fault framing module 12, equipment matching module 13, strategy optimization module 14, fault identification module 15. DETAILED DESCRIPTION

[0012] The present application provides a bus duct detection equipment motion control system and method, which solves the technical problem of inaccurate fault area positioning and low detection efficiency due to the complex structure of bus ducts and the incomplete coverage of single detection method, and achieves the technical effect of efficient and accurate fault holographic identification through multi-sensor data fusion, dynamic path planning and intelligent switching of equipment, effectively improving the positioning accuracy of the fault area.

[0013] Hereinafter, the technical solutions in the present application will be described clearly and completely with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, not all embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the present application. In addition, it should be noted that, for convenience of description, only parts related to the present application are shown in the drawings, not all.

[0014] Embodiment one, as shown in the present application, a bus duct detection equipment motion control system is provided, which comprises: Figure 1

[0015] The monitoring configuration module 11 is used for multi-dimensional monitoring coverage configuration of the bus power distribution network according to the monitoring scene, and a multi-dimensional sensor array is obtained.

[0016] ​Specifically, in the monitoring configuration module 11, the system terminal first conducts comprehensive monitoring coverage of the busbar distribution network according to the requirements of the monitoring scene. This process involves configuring multiple different types of sensors according to different detection requirements and environmental conditions (such as straight sections, bends, shafts, or high-pressure enclosed spaces), forming a multi-dimensional sensor array. Specifically, the sensor array combines different sensor types, such as infrared thermal imagers, ultrasonic sensors, partial discharge detectors, current transformers, etc., to comprehensively monitor various operating parameters and states of the busbar trunking. Each sensor is placed at different positions in the busbar distribution network according to its specific function, enabling multi-dimensional monitoring of the busbar trunking. Among them, the infrared thermal imager is mainly responsible for monitoring temperature abnormalities, the ultrasonic sensor is used to detect mechanical vibrations and internal defects, the partial discharge detector captures insulation deterioration signals, and the current transformer monitors real-time load current fluctuations. Through such comprehensive configuration, any possible fault signals or abnormal states can be ensured not to be missed during the monitoring process, and accurate data support is provided for subsequent fault location and diagnosis. Ultimately, these data will constitute a highly integrated multi-dimensional sensor array, enabling precise monitoring and data acquisition in any area of the busbar trunking.

[0017] The fault framing module 12 is used to preliminarily frame the fault range according to the multi-dimensional monitoring time sequence array returned by the multi-dimensional sensor array, and obtain a fault detection interval, wherein the fault detection interval is identified by real-time characteristic parameters.

[0018] Specifically, in the fault framing module 12, the inspection control center receives monitoring data from the multi-dimensional sensor array, which is returned in a time sequence manner, containing various monitoring information of the bus duct at different time points. The control center processes and analyzes these returned data, first comprehensively evaluates the characteristic parameters collected at each time, and then determines the potential fault area. Specifically, the control center uses the time sequence information in the multi-dimensional monitoring data to identify faults from two aspects of signal range and data trend. In terms of signal range, the signal value of the sensor is detected to determine whether it appears a signal value outside the standard range in continuous multiple return periods (which can be set according to actual needs, such as two return periods). For example, for an infrared sensor, its output value should be within a standard temperature range. If the returned data exceeds the standard range in continuous periods, it means that the monitoring area corresponding to the multi-dimensional sensor array may have a fault. In terms of data trend, it is checked whether the change of the returned data is significantly different from the normal operating state. Under normal circumstances, the data should show a certain trend change (such as temperature gradually changing at a permitted change rate). If it is found that in continuous multiple return periods, the data has abnormal temperature rise, sudden change of vibration frequency, or sharp increase of current load, i.e., the change rate exceeds the permitted change rate, it means that the monitoring area corresponding to the multi-dimensional sensor array may have a fault. Through data comparison and trend analysis, the area where the fault may occur is preliminarily identified, and the monitoring range of the sensor corresponding to the abnormal data in the multi-dimensional sensor array is used to preliminarily frame the fault range, so as to determine the fault detection interval. Each fault detection interval has a real-time characteristic parameter identifier, i.e., the key parameters such as temperature, vibration, and current that exist abnormally are used for calibration. These characteristic parameters are helpful for more accurate diagnosis and detection in the future, so as to ensure that the bus duct detection equipment (such as a magnetic multi-legged composite robot) can locate the specific position of the fault in the shortest time, and provide basic data support for subsequent accurate detection and maintenance.

[0019] The device matching module 13 is used for matching the hot standby detection equipment according to the fault detection interval and the real-time characteristic parameter, to obtain multiple real-time detection equipment.

[0020] Specifically, in the device matching module 13, according to the fault detection interval determined in the previous step and the real-time characteristic parameter, the inspection control center will perform the matching work of the hot standby detection device. Specifically, the control center will input the real-time characteristic parameter identifier of the fault detection interval into the pre-constructed fault parameter correlation topology for reverse deduction to determine the possible fault types existing in the fault detection interval, and then determine the functional features required by the suitable detection device according to these possible fault types. Subsequently, with the fault detection interval as the constraint, the detection device that can complete the detection of the fault detection interval within the response time is selected from all bus duct detection devices. The bus duct detection device is a composite robot combining magnetic attraction and multi-legged gripping functions, which is specially used for accurate fault detection and diagnosis of bus ducts. The design goal of the device is to automatically complete the inspection task of the bus duct by flexibly moving in the complex power distribution environment through magnetic attraction or climbing. Different bus duct detection devices have different detection functions and moving speeds due to the differences in the sensors carried. Then, the functions of the selected bus duct detection device are matched with the functional features required by the hot standby detection device (in standby state, can be dispatched to the fault area at any time, maintains real-time availability, does not need preheating or starting time) obtained by the previous analysis, so as to obtain multiple real-time detection devices, providing sufficient device resources for subsequent fault detection and analysis.

[0021] Further, the device matching module 13 comprises:

[0022] a first demand reverse deduction unit for detecting demand reverse deduction according to the real-time characteristic parameter, and matching detection device functional features according to the reverse deduction result; a first device matching unit for matching hot standby detection devices in the bus power distribution network according to the detection device functional features and the fault detection interval, and obtaining multiple real-time detection devices.

[0023] In a preferred embodiment, the first demand backstepping unit extracts real-time characteristic parameters from the fault detection interval, which includes temperature, vibration, current, thermal imaging and other characteristic parameters. The system terminal analyzes these real-time characteristic parameters based on the fault parameter correlation topology to identify the possible fault types of the current bus duct. Subsequently, according to the corresponding relationship between fault types and fault detection requirements, the possible fault types are backstepped and matched to determine the detection device functional characteristics, which reflect the capabilities that the bus duct detection device should have. For example, if the temperature sensor detects an abnormal temperature rise, it may indicate that there is an overheating fault, and an infrared detection function of the bus duct detection device is backstepped to further diagnose the specific fault of the temperature abnormal area. Then, the first device matching unit processes the detection device functional characteristics and the fault detection interval jointly to match the hot standby detection device from the response and function aspects, ensuring that the selected detection device is suitable for the current monitoring range and detection requirements. Finally, the first device matching unit selects multiple real-time detection devices according to the above matching rules, which do not need to be preheated or started up, and can be dispatched to the fault area for detection at any time, ensuring efficient and comprehensive detection of the fault area and providing protection for subsequent fault location and diagnosis.

[0024] Further, the first demand backstepping unit includes:

[0025] a historical record calling unit for locally calling historical detection records of the bus power distribution network; a historical record clustering unit for clustering the historical detection records based on feature parameter-fault type mapping features to obtain multiple sample fault type sets of multiple sample feature parameters; an association topology construction unit for constructing a fault parameter association topology according to the fault coincidence relationship of the multiple sample fault type sets and the multiple sample feature parameters and sample fault type sets; a fault backstepping unit for inputting the real-time characteristic parameters into the fault parameter association topology to backstep the fault type to obtain K associated fault types; a second demand backstepping unit for backstepping the detection requirements according to the K associated fault types to obtain K sample device functions, wherein the K sample device functions constitute the detection device functional characteristics.

[0026] In a preferred embodiment, the system terminal retrieves historical detection data of the busbar distribution network from the local database or storage system through a historical record calling unit, which includes previous fault detection records, operating parameters (such as temperature, current, vibration, etc.), and fault types, etc. These historical data help provide references and backgrounds for the current detection task. Subsequently, a historical record clustering unit maps the feature parameters (such as temperature, vibration, current, etc.) in the historical detection data to the corresponding fault types, and based on these mapping relationships, aggregates feature parameters belonging to the same fault type together, for example, if certain feature parameters are associated with the same fault type in multiple historical records, these records will be clustered into a sample set. Through clustering, multiple sample fault type sets are generated, each containing sample feature parameters related to the type, which can help the system terminal understand the association between different feature parameters and different fault types. Then, an association topology construction unit uses the fault overlap relationships in the sample fault type sets (whether different fault types appear under similar feature parameter conditions, such as current increase may lead to the occurrence of overheating fault and overload fault) to construct a fault parameter association topology, which reflects the internal association and mutual influence between different fault types and feature parameters. Topology construction can be achieved through a graph model, where each node represents a feature parameter (such as temperature, current, vibration, etc.) and a fault type (such as overload, poor contact, etc.), and the edge represents the association between them. For each pair of associated feature parameters and fault types, if they appear in similar patterns between multiple fault type sets, an edge will be created between them, and the weight of the edge can be set according to the degree of overlap between them, for example, if current and overload fault often appear together, a larger weight will be assigned to the edge between them (the ratio of the number of times current and overload fault appear together to the number of times overload fault appears), indicating a strong association between them. Then, the fault backtracking unit inputs real-time feature parameters (such as real-time temperature, vibration, current, etc.) into the already constructed fault parameter association topology, and performs reverse reasoning through the guidance of the edges, thereby deducing K associated fault types according to the location of the real-time feature parameters in the topology. After obtaining K associated fault types, the second demand backtracking unit will backtrace the corresponding detection requirements based on these fault types, and each fault type may require a specific detection method, such as ultrasonic detection, infrared thermal imaging, etc. During the backtracking process, the system terminal determines the most suitable detection method for each fault type based on its characteristic requirements and industry knowledge, including the detection equipment functions required for each fault type during detection, for example, if a fault type needs to detect temperature changes, a busbar slot detection equipment with infrared thermal imaging function will be selected; if the vibration of mechanical structures needs to be detected, a busbar slot detection equipment with ultrasonic sensors may be selected.In this way, the system terminal obtains K sample device functions according to the requirements of K fault types, and forms the final detection device function features by using these sample device function features. Through the above steps, the fault type and the corresponding detection requirement can be intelligently deduced based on the historical detection records and real-time data, and then the appropriate detection device is matched for each fault type, so as to ensure the accuracy and efficiency of fault detection.

[0027] Further, the first device matching unit comprises:

[0028] The detection device interaction unit is configured to interactively obtain a standard inspection speed of the bus duct detection device; the response scale calculation unit is configured to calculate a fault detection response scale according to the fault detection response window of the bus power distribution network and the standard inspection speed; the device frame selection unit is configured to select a backup detection device frame according to the fault detection response scale, taking the fault detection interval as a starting point, to obtain a plurality of candidate detection devices; and the second device matching unit is configured to match the backup detection device by using the detection device function features to traverse the device function descriptions of the plurality of candidate detection devices, to obtain the plurality of real-time detection devices.

[0029] In a preferred embodiment, the detection equipment interaction unit obtains the standard inspection speed of the equipment by interacting with the bus duct detection equipment, which is usually determined by the technical parameters and design requirements of the equipment and reflects the movement speed of the equipment when performing the inspection task. After obtaining the standard inspection speed, the working efficiency and movement ability of the equipment during the inspection process can be determined. Then, the response scale calculation unit calculates the standard inspection speed of the equipment based on the fault detection response window of the bus power distribution network. The fault detection response window refers to a pre-set time range for determining how quickly the equipment needs to respond and detect when a fault occurs. Based on the standard inspection speed and the fault detection response window, the response scale calculation unit multiplies the standard inspection speed by the fault detection response window to calculate a fault detection response scale, which reflects the detection range that the bus duct detection equipment can cover within a specified time, determines the working area and movement path of the equipment, and helps ensure that the detection equipment can timely and effectively complete the fault detection task in the specified area. After that, the equipment bounding box unit will select all hot standby detection equipment based on the fault detection interval as the starting point. In this process, the bounding box unit will start from the fault detection interval and find equipment that can reach and complete the detection of the interval within the response scale. By comparing the distance from the fault detection interval to the equipment location and the fault detection response scale, all equipment that meet the requirements are selected as multiple candidate detection equipment. Then, the second equipment matching unit determines the functional characteristics of each equipment based on the multiple candidate detection equipment obtained by bounding box, such as infrared thermal imaging and ultrasonic detection. The system terminal compares the functional characteristics of the multiple candidate detection equipment with the detection equipment functional characteristics to select equipment that meets the current detection requirements, which are matched as multiple real-time detection equipment and prepared for the fault detection process to effectively perform the detection task of the fault interval, improving the accuracy and automation of fault diagnosis.

[0030] The strategy optimization module 14 is configured to perform mobile strategy optimization based on the fault detection interval and the multiple real-time equipment positions of the multiple real-time detection equipment to screen a target multi-modal diagnosis path of a target detection equipment.

[0031] Specifically, in the strategy optimization module 14, the system terminal first performs global motion planning according to the previously determined fault detection interval and the current positions of the plurality of real-time detection devices (which can be obtained through preset position information), with the goal of generating a plurality of feasible diagnosis paths to enable the device to cover the fault detection interval while avoiding known obstacles. After obtaining a plurality of fault diagnosis paths, the system terminal evaluates each path, taking into account not only the energy consumption and time required of the path, but also the feasibility of the path, such as whether the device can flexibly switch between motion modes (magnetic attraction mode and multi-legged mode) to adapt to different environments. After evaluation, the most suitable target multi-modal diagnosis path is selected. This path not only ensures that the device covers all detection requirements within the fault detection interval, but also flexibly switches between motion modes to adapt to the complexity of the bus duct environment, thereby achieving accurate diagnosis of bus duct faults.

[0032] Further, the strategy optimization module 14 includes:

[0033] a diagnosis path fitting unit for fitting a diagnosis path in the bus power distribution network according to the fault detection interval and a plurality of real-time device positions of the plurality of real-time detection devices, to obtain a plurality of fault diagnosis paths; a first mobile strategy evaluation unit for evaluating the mobile strategy of the plurality of fault diagnosis paths to select a target multi-modal diagnosis path of the target detection device.

[0034] In a preferred embodiment, the diagnostic path fitting unit first considers the spatial layout of the fault detection interval, which is a region determined according to the actual situation of the bus duct, combined with sensor data, and this region not only contains horizontal sections but also may include vertical bus duct sections. In the diagnostic path fitting process, first define the starting point (such as the initial position of the device), represented as a three-dimensional coordinate, and define the target point (such as the center position of the fault detection region), represented as a three-dimensional coordinate. Then, establish a tree at the starting point position, where the nodes of the tree represent the positions of the device in three-dimensional space. When initializing the tree, the root node of the tree is the starting point. Subsequently, define the three-dimensional working space of the bus duct, including all possible regions, divided into obstacle regions (such as devices, supports, walls, etc.) and free space (walkable regions). Then set the sampling range and step size, for example, randomly sample points within a certain range and ensure that the sampling points are within the walkable regions of the bus duct, including vertical and horizontal sections, considering the z-direction motion range. At the same time, set the expansion step size of the tree, which is the maximum distance from the current node to the new node in the tree. In three-dimensional space, the step size affects the expansion speed and path accuracy of the tree. After the environmental parameters are set, randomly sample a point in three-dimensional space. This point can be a completely random position or use a target attraction strategy (i.e., the sampling point is closer to the target point). The target attraction strategy helps to expand faster towards the target point, enhancing the fitting efficiency. In three-dimensional path planning, the coordinates of the sampling point will include x, y, and z, especially on the vertical section path, the change of z coordinate is crucial to path planning. To expand the tree, the system terminal needs to find the nearest node in the tree to the sampling point. The expansion of the tree is to generate a new node along the direction from the nearest node to the sampling point, with a maximum step size. If the generated new node exceeds the distance from the sampling point to the nearest node, it expands to the sampling point. In three-dimensional space, special consideration is needed for the change in the z direction during the expansion process to ensure smooth transitions between vertical sections. At the same time, collision detection is needed during expansion to ensure that the path does not collide with obstacles, especially in vertical section path planning to ensure stable movement of the device during vertical motion. Once the tree is expanded near the target point, the path is backtracked, and the backtracking process traces the parent nodes in the tree until it returns to the starting point, obtaining the path from the starting point to the target point, i.e., the fault diagnosis path. After obtaining a fault diagnosis path, the fitting is started again from the starting point for the remaining nodes until each node participates in the fitting. In this way, multiple possible fault diagnosis paths can be obtained. After obtaining multiple possible fault diagnosis paths, the first mobile strategy evaluation unit begins to evaluate these paths, focusing on the energy consumption, switching, and time consumption of each path. Through the evaluation of multiple fault diagnosis paths, the system terminal selects the optimal path as the target multi-modal diagnosis path. This path meets the requirements of efficient and accurate detection, while considering the switching of the device's motion mode between vertical and horizontal bus duct sections.For example, when the device needs to transition from a horizontal section to a vertical section, it needs to switch to magnetic attraction mode, and after completing detection in the vertical section, switch back to the multi-legged mode to continue moving in the horizontal section. Through the above process, it can be ensured that the device can flexibly cope with the switching between vertical and horizontal sections in a complex bus duct structure, thereby efficiently performing the detection task and improving the accuracy of fault identification.

[0035] Further, the first mobile strategy evaluation unit comprises:

[0036] The diagnostic path decomposition unit is configured to decompose the first fault diagnosis path according to the path vector switching to obtain M local diagnosis paths; the second mobile strategy evaluation unit is configured to evaluate the M local diagnosis paths according to the path space characteristics to obtain a first mobile strategy coefficient; the third mobile strategy evaluation unit is configured to evaluate the plurality of fault diagnosis paths by analogy to obtain a plurality of mobile strategy coefficients; and the diagnostic path screening unit is configured to screen the target multi-modal diagnosis path from the plurality of fault diagnosis paths according to the plurality of mobile strategy coefficients, and locate the target detection device according to the path device correspondence relationship.

[0037] In a preferred embodiment, the diagnostic path decomposition unit performs path vector switching on the first fault diagnosis path, which refers to decomposing the entire fault diagnosis path according to certain rules, mainly based on the direction change, spatial layout and movement mode of the equipment. For example, if the first fault diagnosis path transitions from a horizontal segment to a vertical segment, the path will be cut into two local paths, one for the horizontal segment and the other for the vertical segment, or if the direction of the first fault diagnosis path changes, it will be cut at the turning point. Through path decomposition, M local diagnosis paths are generated. Each local path represents a stage or a specific movement trajectory, for example, there may be a path dedicated to the movement of the equipment on the horizontal segment of the bus duct, and another path dedicated to the movement of the equipment on the vertical segment. Each local path has a clear target and can ensure the equipment to complete the fault diagnosis task smoothly. Subsequently, the second mobile strategy evaluation unit evaluates each local diagnosis path based on its spatial characteristics, including path shape, equipment movement mode, etc. Through analysis in terms of energy consumption characteristics, switching characteristics, time consumption characteristics, etc., and then using the preset evaluation rules to weight and fuse the analysis results, the first mobile strategy coefficient of the first fault diagnosis path is obtained, which measures the adaptability and execution difficulty of the path. After the second mobile strategy evaluation unit evaluates the local paths, the third mobile strategy evaluation unit evaluates the mobile strategy of all diagnosis paths to assess their adaptability in the overall movement of the equipment. In this stage, a mobile strategy coefficient will be assigned to each fault diagnosis path, and the strategy coefficients of multiple paths will be aggregated. This process ensures that the complexity of each path is considered comprehensively, and the most suitable movement strategy for the equipment is selected to ensure the execution efficiency and accuracy of the path. Then, the diagnostic path screening unit selects the optimal fault diagnosis path according to the above multiple mobile strategy coefficients, i.e. the one with the largest mobile strategy coefficient. This fault diagnosis path will be the target multi-modal diagnosis path to ensure that the detection task can be completed in the shortest time and the risk of equipment movement can be minimized. After the target multi-modal diagnosis path is selected, the system terminal determines the final selected target detection equipment according to the correspondence between the path and the equipment. Each path is matched with a specific detection equipment to ensure that the equipment can adapt to the changes in the path and spatial requirements. Through this method, each bus duct detection equipment can perform fault diagnosis according to the most suitable path, making it smoothly execute the detection task.

[0038] Further, the second mobile strategy evaluation unit includes:

[0039] a motion pattern matching unit configured to perform motion pattern matching on the M local diagnosis paths to obtain M local motion patterns; a motion switching analysis unit configured to perform motion switching analysis on the M local motion patterns to output M motion pattern features; a control energy consumption analysis unit configured to, after loading the M local diagnosis paths and the M motion pattern features into a motion energy consumption prediction model to perform control energy consumption analysis and obtaining M local energy consumption features, add the M local energy consumption features to obtain a first energy consumption feature; a switching statistics unit configured to perform motion control switching statistics according to the M local motion patterns to obtain a first switching feature; a time consumption feature calculation unit configured to calculate a first time consumption feature according to the M local diagnosis paths and the M motion pattern features; and a feature weighting unit configured to, based on a preset evaluation rule, weight and fuse the first energy consumption feature, the first switching feature, and the first time consumption feature as the first mobile strategy coefficient.

[0040] In a preferred embodiment, after the diagnosis path decomposition, the motion pattern matching unit performs motion pattern matching analysis on each local diagnosis path, and the goal of motion pattern matching is to determine the motion mode that the device should adopt according to the spatial characteristics (such as the shape, direction, etc. of the path) of each local path, for example, if the path is a horizontal segment, the device can walk in a multi-legged mode; if the path is a vertical segment, the device may need to switch to a magnetic attraction mode to ensure stable adhesion on the bus duct surface. Through the matching process, the system terminal assigns an appropriate motion mode to each local diagnosis path, obtaining M local motion modes, which describe the specific motion mode that the device should adopt on different local paths. Subsequently, the motion switching analysis unit analyzes the M local motion modes to determine the impact of switching from one mode to another, and in this process, the M local motion modes are matched with the sample motion modes to obtain the motion speed of each local motion mode, and these motion speeds are taken as M motion mode features, corresponding to the M local motion modes. Then, the M local diagnosis paths and M motion mode features are loaded into the motion energy consumption prediction model for analysis, and before loading, the local diagnosis paths and motion mode features are standardized, that is, the coordinate points and running speeds are scaled to a fixed range (such as [0, 1]) through Min-Max Scaling. The energy consumption prediction model will analyze the control energy consumption of the received local diagnosis paths and motion mode features based on the learned mapping relationship, thereby obtaining M local energy consumption features. This motion energy consumption prediction model can be based on a fully connected neural network, or based on a convolutional neural network, a random forest regression, etc. Taking a fully connected neural network as an example, the system terminal will input the sample diagnosis paths and sample running speeds into the network, and perform iterative training through forward propagation, loss calculation, back propagation, parameter optimization, etc. When the iteration end condition (such as the maximum number of iterations, loss function convergence) is reached, the current network will be tested using test data, and the accuracy and other indicators obtained through testing will be used for evaluation. When the evaluation result is greater than or equal to the preset value, the current network is saved as the motion energy consumption prediction model, otherwise, the learning rate, training batch size, and other hyperparameters are adjusted to further improve the prediction performance of the network. According to the model analysis, the system terminal will calculate the energy consumption features of each local path, and each energy consumption feature reflects the energy consumption of the device when executing the path. By accumulating these energy consumption features, the first energy consumption feature representing the first fault diagnosis path can be obtained. Then, the switching statistics unit will count the number of times the device switches from one mode to another according to the M local motion modes, and take the counted number of times as the first switching feature to describe the switching frequency and complexity of the device when executing the path.Further, the time-consuming feature calculation unit calculates the time required by the device on each local path by dividing the local diagnostic path by the motion mode feature, and takes this time as the time-consuming feature. By accumulating all the time-consuming features, the first time-consuming feature is obtained, which represents the total time required by the device to complete all local diagnostic paths. Finally, the feature weighting unit weights and fuses the first energy consumption feature, the first switching feature and the first time-consuming feature according to the preset evaluation rule. The evaluation rule includes assigning different weights to different features according to different requirements of the task. For example, if energy saving is the priority, the weight of the energy consumption feature may be higher, and if efficiency is more important, the weight of the time-consuming feature may be greater. Through weighted fusion, the first mobile strategy coefficient is obtained, which reflects the comprehensive performance of the device in executing the task, including energy efficiency, switching complexity and time required to complete the task. The goal is to evaluate the overall mobile strategy of the device through this coefficient to ensure the efficiency and stability of path execution while avoiding unnecessary path switching or energy waste.

[0041] Further, the motion mode feature includes sample motion mode and sample motion speed.

[0042] In a preferred embodiment, the motion mode feature includes sample motion mode and sample motion speed, wherein the sample motion mode is reference data for comparison with the local motion mode, providing some predefined motion modes such as magnetic attraction mode, multi-legged mode. The sample motion speed is a parameter associated with each sample motion mode, which describes the motion speed of the device in a specific motion mode. These motion speed values help the system terminal more accurately evaluate the performance and motion efficiency of the device on each local diagnostic path.

[0043] The fault recognition module 15 is used to drive the target detection device to perform fault holographic recognition in the fault detection interval according to the real-time feature parameters after the target multimodal diagnostic path is used to control the movement of the target detection device to the fault detection interval, and output real-time fault nodes and real-time fault types.

[0044] Specifically, in the fault recognition module 15, under the control of the target multi-modal diagnostic path, the target detection equipment will move according to the pre-planned path until it reaches the designated fault detection interval. When the equipment reaches the fault detection interval, it will operate based on real-time characteristic parameters such as temperature, vibration, current, and other sensor-collected data, which provide key data about the current operating state of the bus duct, helping the equipment determine whether there is an anomaly. Specifically, the target detection equipment will start the fault holographic recognition process, which means that the equipment not only simply detects faults but also comprehensively identifies and analyzes the type and location of the fault. The fault holographic recognition process usually involves the joint use of multi-modal sensors (such as ultrasonic waves, infrared thermographs, etc.), capturing data from multiple angles and levels to provide comprehensive information about the fault. In this process, the target detection equipment outputs the real-time fault node (i.e., the specific location where the fault occurs) and the real-time fault type (i.e., the specific nature of the fault, such as overload, short circuit, poor contact, etc.) to the system terminal, which will make judgments based on this information and take appropriate subsequent measures such as alarm, maintenance scheduling, etc. for subsequent decision-making, ensuring that sufficient information support can be provided for subsequent maintenance and optimization.

[0045] Further, the fault recognition module includes:

[0046] a detection mode activation unit for activating a multi-modal joint detection mode according to the K sample equipment functions, wherein the multi-modal joint detection mode includes K single-modal detection modes; a first fault recognition unit for driving the target detection equipment to perform fault holographic recognition in the fault detection interval using the multi-modal joint detection mode.

[0047] In a preferred embodiment, the detection mode activation unit activates a corresponding multi-modal joint detection mode according to K kinds of sample device functions, where the K kinds of sample device functions are based on the previous demand reverse deduction and device matching process, which may include infrared thermal imaging, ultrasonic detection, vibration monitoring, etc., each of which is suitable for different types of fault detection. The multi-modal joint detection mode is realized by combining multiple single-modal detection modes, each of which represents an independent detection technology. For example, in a fault diagnosis task, an infrared thermal imaging mode may be used to identify temperature anomalies, while an ultrasonic detection mode may be used to identify mechanical faults such as poor contact. The system terminal combines these single-modal modes to ensure multi-angle and all-around detection of faults. After activating the multi-modal joint detection mode, the first fault recognition unit drives the target detection device to perform holographic recognition, which means combining multiple detection methods and information to comprehensively diagnose and analyze the fault area. In this process, the target detection device performs different detection tasks within the fault detection interval, using the activated single-modal detection modes to analyze the device state in detail. For example, the device may first use the ultrasonic mode for fault positioning, then use the infrared mode to collect data in the positioning area, and then input these data into the fault type discriminator for analysis. The fault type discriminator is constructed in a similar manner as described above. The device will perform effective diagnosis within the fault detection interval according to the multi-modal joint detection mode. Multi-modal detection can simultaneously capture different types of fault signals (such as temperature, pressure, vibration, etc.), providing more comprehensive and accurate fault recognition results. Through this joint detection, the risk of missed detection and misjudgment is reduced, and the overall fault diagnosis accuracy is improved.

[0048] In summary, the bus duct detection device motion control system provided by the embodiments of the present application has the following technical effects:

[0049] The monitoring configuration module 11 is configured to perform multi-dimensional monitoring coverage configuration on the bus power distribution network according to a monitoring scene, to obtain a multi-dimensional sensing array; the fault framing module 12 is configured to perform preliminary fault range framing according to a multi-dimensional monitoring time sequence array returned by the multi-dimensional sensing array, to obtain a fault detection interval, wherein the fault detection interval is identified by a real-time characteristic parameter; the device matching module 13 is configured to perform hot backup detection device matching according to the fault detection interval and the real-time characteristic parameter, to obtain a plurality of real-time detection devices; the strategy optimization module 14 is configured to perform mobile strategy optimization according to the fault detection interval and a plurality of real-time device positions of the plurality of real-time detection devices, to screen a target multi-modal diagnostic path of a target detection device; and the fault identification module 15 is configured to perform fault holographic identification in the fault detection interval according to the real-time characteristic parameter after the target detection device is controlled to move to the fault detection interval by using the target multi-modal diagnostic path, to output a real-time fault node and a real-time fault type. Through the above steps, the technical problem of inaccurate fault area positioning and low detection efficiency caused by the fact that a single detection method cannot cover all due to the complex structure of the bus duct is solved, and the technical effect of effectively improving the positioning accuracy of the fault area and realizing efficient and accurate fault holographic identification through multi-sensor data fusion, dynamic path planning and intelligent switching of devices is achieved.

[0050] In the second embodiment, based on the same inventive concept as the bus duct detection device motion control system in the foregoing embodiments, as shown in the accompanying drawings, the embodiment of the present application provides a bus duct detection device motion control method, which comprises the following steps: Figure 2

[0051] The monitoring configuration module 11 is configured to perform multi-dimensional monitoring coverage configuration on the bus power distribution network according to a monitoring scene, to obtain a multi-dimensional sensing array; the fault framing module 12 is configured to perform preliminary fault range framing according to a multi-dimensional monitoring time sequence array returned by the multi-dimensional sensing array, to obtain a fault detection interval, wherein the fault detection interval is identified by a real-time characteristic parameter; the device matching module 13 is configured to perform hot backup detection device matching according to the fault detection interval and the real-time characteristic parameter, to obtain a plurality of real-time detection devices; the strategy optimization module 14 is configured to perform mobile strategy optimization according to the fault detection interval and a plurality of real-time device positions of the plurality of real-time detection devices, to screen a target multi-modal diagnostic path of a target detection device; and the fault identification module 15 is configured to perform fault holographic identification in the fault detection interval according to the real-time characteristic parameter after the target detection device is controlled to move to the fault detection interval by using the target multi-modal diagnostic path, to output a real-time fault node and a real-time fault type. Through the above steps, the technical problem of inaccurate fault area positioning and low detection efficiency caused by the fact that a single detection method cannot cover all due to the complex structure of the bus duct is solved, and the technical effect of effectively improving the positioning accuracy of the fault area and realizing efficient and accurate fault holographic identification through multi-sensor data fusion, dynamic path planning and intelligent switching of devices is achieved.

[0052] Further, the method comprises:

[0053] ​According to the real-time characteristic parameter, a detection demand is inversely deduced, and a detection equipment function characteristic is matched according to an inverse deduction result; according to the detection equipment function characteristic and a fault detection interval, a hot backup detection equipment is matched in the bus power distribution network, and a plurality of real-time detection equipments are obtained.

[0054] Further, the method comprises:

[0055] A historical record calling unit is configured to locally call historical detection records of the bus power distribution network, cluster the historical detection records based on a characteristic parameter-fault type mapping feature to obtain a plurality of sample fault type sets of a plurality of sample characteristic parameters, construct a fault parameter correlation topology according to the plurality of sample fault type sets and the plurality of sample characteristic parameters based on a fault coincidence relationship of the plurality of sample fault type sets, input the real-time characteristic parameter into the fault parameter correlation topology to inversely deduce a fault type, and obtain K associated fault types; according to the K associated fault types, a detection demand is inversely deduced to obtain K sample equipment functions, wherein the K sample equipment functions constitute the detection equipment function characteristic.

[0056] Further, the method comprises:

[0057] According to the K sample equipment functions, a multi-modal joint detection mode is activated, wherein the multi-modal joint detection mode comprises K single-modal detection modes; and the multi-modal joint detection mode is used to drive the target detection equipment to perform fault holographic identification in the fault detection interval.

[0058] Further, the method comprises:

[0059] According to the fault detection interval and a plurality of real-time equipment positions of a plurality of real-time detection equipments, diagnostic path fitting is performed in the bus power distribution network to obtain a plurality of fault diagnostic paths; and the plurality of fault diagnostic paths are subjected to mobile strategy evaluation to screen a target multi-modal diagnostic path of the target detection equipment.

[0060] Further, the method comprises:

[0061] According to a path vector switching, a first fault diagnostic path is decomposed to obtain M local diagnostic paths; according to path space features, the M local diagnostic paths are subjected to stage-by-stage mobile strategy evaluation to obtain a first mobile strategy coefficient; and in the same way, the plurality of fault diagnostic paths are subjected to mobile strategy evaluation to obtain a plurality of mobile strategy coefficients; according to the plurality of mobile strategy coefficients, the target multi-modal diagnostic path is screened from the plurality of fault diagnostic paths, and the target detection equipment is located according to a path equipment correspondence relationship.

[0062] Further, the method comprises:

[0063] The M local diagnosis paths are subjected to motion pattern matching to obtain M local motion patterns; the M local motion patterns are subjected to motion switching analysis to output M motion pattern features; after the M local diagnosis paths and the M motion pattern features are loaded into a motion energy consumption prediction model for control energy consumption analysis to obtain M local energy consumption features, the first energy consumption feature is obtained by summing the M local energy consumption features; the first switching feature is obtained by performing motion control switching statistics according to the M local motion patterns; the first time consumption feature is calculated according to the M local diagnosis paths and the M motion pattern features; and the first energy consumption feature, the first switching feature and the first time consumption feature are weighted and fused as the first mobile strategy coefficient based on a preset evaluation rule.

[0064] Further, the method comprises:

[0065] The motion pattern features comprise sample motion patterns and sample motion speeds.

[0066] Further, the method comprises:

[0067] The standard inspection speed of the bus duct detection device is interactively obtained; the fault detection response scale is calculated according to the fault detection response window of the bus power distribution network and the standard inspection speed; the hot backup detection device is framed according to the fault detection response scale with the fault detection interval as the starting point to obtain a plurality of alternative detection devices; and the detection device function features are used to traverse the device function descriptions of the plurality of alternative detection devices to perform hot backup detection device matching to obtain the plurality of real-time detection devices.

[0068] Any step of the above method can be stored as computer instructions or programs in an unrestricted computer memory and can be called and recognized by an unrestricted computer processor to implement any method in the embodiments of the present application, and no redundant limitation is made herein.

[0069] Further, the first or the second described above may not only represent an order relationship, but also may represent a certain specific concept, and / or refer to the selection of multiple elements individually or collectively. Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. Busway detection apparatus motion control system, characterized in that, The system comprises: A monitoring configuration module is configured to perform multi-dimensional monitoring coverage configuration on the bus power distribution network according to a monitoring scene, and obtain a multi-dimensional sensing array; A fault framing module is configured to perform preliminary fault range framing on a multi-dimensional monitoring time sequence array returned by the multi-dimensional sensing array according to a patrol control center, and obtain a fault detection interval, wherein the fault detection interval is identified by using real-time characteristic parameters; A device matching module is configured to perform hot backup detection device matching according to the fault detection interval and real-time characteristic parameters, and obtain a plurality of real-time detection devices; A strategy optimization module is configured to perform mobile strategy optimization according to the fault detection interval and a plurality of real-time device positions of the plurality of real-time detection devices, and screen a target multi-modal diagnosis path of a target detection device; A fault identification module is configured to perform fault holographic identification in the fault detection interval according to the real-time characteristic parameters after controlling the target detection device to move to the fault detection interval by using the target multi-modal diagnosis path, and output real-time fault nodes and real-time fault types; The strategy optimization module comprises: A first mobile strategy evaluation unit is configured to screen a target multi-modal diagnosis path of a target detection device by performing mobile strategy evaluation on a plurality of fault diagnosis paths; The first mobile strategy evaluation unit comprises: A diagnosis path decomposition unit is configured to decompose a first fault diagnosis path according to path vector switching, and obtain M local diagnosis paths; A second mobile strategy evaluation unit is configured to perform stage-by-stage mobile strategy evaluation on the M local diagnosis paths according to path space features, and obtain a first mobile strategy coefficient; A third mobile strategy evaluation unit is configured to perform mobile strategy evaluation on the plurality of fault diagnosis paths by analogy, and obtain a plurality of mobile strategy coefficients; A diagnosis path screening unit is configured to screen the target multi-modal diagnosis path from the plurality of fault diagnosis paths according to the plurality of mobile strategy coefficients, and locate the target detection device according to a path-device correspondence relationship.

2. The busway detection apparatus motion control system of claim 1, wherein, The device matching module comprises: A first demand backstepping unit is configured to perform detection demand backstepping according to the real-time characteristic parameters, and match detection device functional features according to a backstepping result; A first device matching unit is configured to perform hot backup detection device matching on the bus power distribution network according to the detection device functional features and the fault detection interval, and obtain a plurality of real-time detection devices.

3. The busway detection apparatus motion control system of claim 2, wherein, The first demand backstepping unit comprises: A historical record calling unit is configured to locally call historical detection records of the bus power distribution network; A historical record clustering unit is configured to cluster the historical detection records based on feature parameter-fault type mapping features, and obtain a plurality of sample fault type sets of a plurality of sample characteristic parameters; An association topology construction unit is configured to construct a fault parameter association topology according to the plurality of sample fault type sets, the plurality of sample characteristic parameters, and a plurality of sample fault type sets according to a fault coincidence relationship of the plurality of sample fault type sets; A fault backstepping unit is configured to input the real-time characteristic parameters into the fault parameter association topology to perform fault type backstepping, and obtain K associated fault types; A second demand backstepping unit is configured to backstep the detection demand according to the K associated fault types, to obtain K sample device functions, wherein the K sample device functions constitute the detection device function feature.

4. The busway detection apparatus motion control system of claim 3, wherein, The fault identification module comprises: A detection mode activation unit is configured to activate a multi-modal joint detection mode according to the K sample device functions, wherein the multi-modal joint detection mode comprises K single-modal detection modes; A first fault identification unit is configured to drive the target detection device to perform fault holographic identification in the fault detection interval by using the multi-modal joint detection mode.

5. The busway detection apparatus motion control system of claim 1, wherein, The strategy optimization module comprises: A diagnostic path fitting unit is configured to fit diagnostic paths in the busbar power distribution network according to the fault detection interval and real-time device positions of multiple real-time detection devices, to obtain multiple fault diagnostic paths.

6. The busway detection apparatus motion control system of claim 1, wherein, The second mobile strategy evaluation unit comprises: A motion mode matching unit is configured to match motion modes of the M local diagnostic paths, to obtain M local motion modes; A motion switching analysis unit is configured to analyze the M local motion modes, to output M motion mode features; A control energy consumption analysis unit is configured to analyze control energy consumption by loading the M local diagnostic paths and M motion mode features into a motion energy consumption prediction model, to obtain M local energy consumption features, and then sum the M local energy consumption features to obtain a first energy consumption feature; A switching statistical unit is configured to statistically analyze motion control switching according to the M local motion modes, to obtain a first switching feature; A time consumption feature calculation unit is configured to calculate a first time consumption feature according to the M local diagnostic paths and M motion mode features; A feature weighting unit is configured to fuse the first energy consumption feature, the first switching feature, and the first time consumption feature by weighting based on a preset evaluation rule, as the first mobile strategy coefficient.

7. The busway detection apparatus motion control system of claim 6, wherein, The motion mode features comprise sample motion modes and sample motion speeds.

8. The busway detection apparatus motion control system of claim 2, wherein, The first device matching unit comprises: A detection device interaction unit is configured to interactively obtain a standard patrol speed of a busbar trunking detection device; A response scale calculation unit is configured to calculate a fault detection response scale according to a fault detection response window of the busbar power distribution network and the standard patrol speed; A device framing unit is configured to frame backup detection devices according to the fault detection response scale, with the fault detection interval as a starting point, to obtain multiple candidate detection devices; A second device matching unit is configured to match backup detection devices by traversing device function descriptions of the multiple candidate detection devices using the detection device function feature, to obtain the multiple real-time detection devices.

9. A busway detection apparatus motion control method characterized by, The method is executed by the busbar trunking detection device motion control system of any one of claims 1 to 8, comprising: performing multi-dimensional monitoring coverage configuration on the busbar power distribution network according to a monitoring scene, to obtain a multi-dimensional sensor array; performing preliminary fault range framing according to a multi-dimensional monitoring time sequence array returned by the multi-dimensional sensor array, to obtain a fault detection interval, wherein the fault detection interval is identified by using real-time feature parameters; According to the fault detection interval and the real-time characteristic parameter, a hot backup detection device is matched to obtain a plurality of real-time detection devices; According to the fault detection interval and a plurality of real-time device positions of the plurality of real-time detection devices, a mobile strategy is optimized to screen a target multi-modal diagnostic path of a target detection device; After the target multi-modal diagnostic path is used to control the target detection device to move to the fault detection interval, the target detection device is driven to perform fault holographic identification in the fault detection interval according to the real-time characteristic parameter, and a real-time fault node and a real-time fault type are output.

Images