An intelligent inspection and control method and system for a distribution substation based on robot collaboration

Abstract

The present invention provides an intelligent inspection and control method and system for a distribution substation based on robot collaboration. The method is to obtain the device information of all distribution devices in the distribution substation, and based on the device information, generate a device inspection sequence according to a preset inspection priority algorithm. Then, according to the movement capabilities of the unmanned vehicle and the robot dog, allocate inspection tasks based on the device inspection sequence to generate the corresponding target inspection routes for the unmanned vehicle and the robot dog and push them to the corresponding robot control terminals respectively. And the robot control terminals control the unmanned vehicle and the robot dog to execute the corresponding inspection tasks, and optimize and adjust the target inspection routes corresponding to the unmanned vehicle and the robot dog according to the dynamically monitored obstacle information and electromagnetic interference information until the inspection task is completed. The present invention can achieve efficient and reliable collaboration in task allocation and path planning among different machines, perceive environmental changes in real time and adaptively adjust the inspection route, effectively improving the inspection efficiency and inspection effect.

Description

Technical Field

[0001] The present invention relates to the technical field of inspection of distribution rooms, and particularly to an intelligent inspection control method and system for distribution rooms based on robot collaboration. Background Art

[0002] The distribution room is a key unit in the power system. The operating states of various power distribution devices in the distribution room will directly affect the stability and safety of power supply distribution. It is necessary to pay attention to the operating states of various power distribution devices in the distribution room to timely detect potential operating hazards and ensure the safe and stable operation of the power system.

[0003] In the existing intelligent inspection scenarios for distribution rooms, the inspection efficiency is mainly improved and the inspection cost is reduced by introducing inspection robots. However, there are many distribution cabinets in the actual distribution room, and the dense electrical equipment makes the passage in the distribution room narrow and there is an irregular distribution of obstacles. These factors make the mobile path planning of the inspection robots extremely complex, resulting in a reduction in the inspection efficiency in actual applications. Although in theory, using multiple robots to perform inspection tasks simultaneously can further improve the inspection efficiency, there are significant differences in the movement capabilities, sensing ranges, and task execution methods of different robots. How to convert fuzzy inspection tasks into specific robot operation instructions to ensure the efficient collaboration of different robots when performing inspection tasks, and then effectively improve the intelligent inspection efficiency and inspection effect of the distribution room has become a technical problem that urgently needs to be solved in the field. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent inspection control method for distribution rooms based on robot collaboration. By combining a task sorting mechanism for generating an inspection sequence based on a comprehensive analysis of the device operating state and device location information, a spatial dimension inspection task allocation mechanism designed based on the movement ability differences of unmanned vehicles and robot dogs and the regional location of power equipment, and an inspection line dynamic adjustment mechanism based on dynamic obstacle and electromagnetic interference information, it can not only achieve efficient and reliable collaboration in task allocation and path planning among different machines, but also adaptively adjust the inspection route by real-time sensing of the dynamic changes in the inspection environment, thereby effectively improving the inspection efficiency and inspection effect and providing a reliable decision-making basis for equipment maintenance.

[0005] To achieve the above object, an intelligent inspection control method and system for distribution rooms based on robot collaboration are provided.

[0006] In the first aspect, an embodiment of the present invention provides an intelligent inspection control method for distribution rooms based on robot collaboration. The robot includes an unmanned vehicle and a robot dog. The method includes the following steps:

[0007] Obtain the device information of all power distribution devices in the power distribution room, and based on the device information, generate a device inspection sequence according to a preset inspection priority algorithm; the device information includes the device operation status and the device location information;

[0008] According to the movement capabilities of the unmanned vehicle and the robot dog, based on the device inspection sequence, allocate inspection tasks, respectively generate the target inspection routes corresponding to the unmanned vehicle and the robot dog, and push the target inspection routes corresponding to the unmanned vehicle and the robot dog to the corresponding robot control terminals respectively;

[0009] The robot control terminals simultaneously control the unmanned vehicle and the robot dog to execute the corresponding inspection tasks, and optimize and adjust the target inspection routes corresponding to the unmanned vehicle and the robot dog according to the dynamically monitored dynamic obstacle information and electromagnetic interference information until the inspection tasks are completed.

[0010] Further, the step of generating a device inspection sequence based on a preset inspection priority algorithm according to the device information includes:

[0011] According to the device operation status in the device information of each power distribution device, generate the corresponding device inspection priority based on a preset inspection priority evaluation model; the preset inspection priority evaluation model is constructed based on fuzzy logic;

[0012] According to the device inspection priorities of each power distribution device, sort all power distribution devices based on the priority descending order method to obtain an initial device inspection sequence;

[0013] Judge whether there are power distribution devices with the same device inspection priority in the initial device inspection sequence;

[0014] If there are, then according to the device location information of the power distribution devices with the same device inspection priority, use the shortest path algorithm to sort and optimize the power distribution devices with the same device inspection priority in the initial device inspection sequence to obtain the device inspection sequence;

[0015] If not, then use the initial device inspection sequence as the device inspection sequence.

[0016] Further, the movement capabilities include movement speed, endurance time, and minimum turning radius; the step of allocating inspection tasks based on the device inspection sequence according to the movement capabilities of the unmanned vehicle and the robot dog, and respectively generating the target inspection routes corresponding to the unmanned vehicle and the robot dog includes:

[0017] According to the movement capabilities of the unmanned vehicle and the robot dog, obtain device area classifications; the device area classifications include open area types and narrow area types;

[0018] Based on the device location information of all power distribution devices in the device inspection sequence, perform spatial clustering based on the device area classification to generate an open area device cluster and a narrow area device cluster;

[0019] According to the open area device cluster and the narrow area device cluster, construct corresponding unmanned vehicle inspection device sets and legged robot inspection device sets respectively;

[0020] According to the device inspection sorting in the device inspection sequence, generate an unmanned vehicle inspection task sequence and a legged robot inspection task sequence corresponding to the unmanned vehicle inspection device set and the legged robot inspection device set respectively;

[0021] Respectively taking the unmanned vehicle inspection task sequence and the legged robot inspection task sequence as sequential constraints, based on a preset path optimization algorithm and a substation grid map, generate the corresponding target inspection routes for the unmanned vehicle and the legged robot.

[0022] Further, the step of respectively taking the unmanned vehicle inspection task sequence and the legged robot inspection task sequence as sequential constraints, and generating the corresponding target inspection routes for the unmanned vehicle and the legged robot based on a preset path optimization algorithm and a substation grid map includes:

[0023] According to the preset path planning constraint conditions, based on the A* algorithm and the substation grid map, generate the target inspection route of the unmanned vehicle; the preset path planning constraint conditions include sequential constraints, device distribution constraints, and channel width constraints;

[0024] According to the preset path planning constraint conditions, based on the breadth-first search algorithm and the substation grid map, generate the target inspection route of the legged robot.

[0025] Further, the step of optimizing and adjusting the corresponding target inspection routes of the unmanned vehicle and the legged robot according to the dynamically monitored dynamic obstacle information and electromagnetic interference information includes:

[0026] When it is determined according to the dynamic obstacle information that the dynamic obstacle needs to be avoided, take the activity area of the dynamic obstacle as the current robot path dangerous area;

[0027] When it is determined according to the electromagnetic interference information that the electromagnetic interference area needs to be avoided, take the electromagnetic interference area as the current robot path dangerous area;

[0028] According to the current robot path dangerous area information and the current robot position information, with the goal of minimizing the path length, minimizing the number of turns, and avoiding the path dangerous area, perform path optimization on the corresponding target inspection route, and update the adjusted and optimized inspection route to the corresponding robot control terminal.

[0029] Further, the method further includes:

[0030] According to the battery capacity of the robotic dog and a preset power consumption rate, the remaining power of the robotic dog is predicted in real time, and according to the current position information of the robotic dog and the remaining inspection route, the remaining task required power is obtained;

[0031] When it is determined that the remaining power is less than the remaining task required power, based on the current position information and the position information of the charging equipment in the power distribution room, an optimal charging path for the robotic dog is generated based on the Dijkstra algorithm and pushed to the corresponding robot control terminal, and after the robotic dog is charged to reach the remaining task required power, the corresponding target inspection route is readjusted according to the latest position information of the robotic dog to perform the remaining inspection tasks.

[0032] Further, the obtaining step of the preset power consumption rate includes:

[0033] Obtain the current environmental temperature, motion state and workload of the robotic dog, and according to the current environmental temperature, the motion state, the workload and a preset power consumption rate prediction model, obtain the preset power consumption rate; the preset power consumption rate prediction model is obtained based on a linear regression analysis of the historical power consumption rate data of the robotic dog under different temperatures, different motion states and different workloads.

[0034] Further, the method further includes:

[0035] During the process of the unmanned vehicle and the robotic dog performing the inspection task, the operation state monitoring data of each power distribution device in the corresponding target inspection route is obtained in real time, and when it is determined that the operation state monitoring data is abnormal, a fault alarm information is generated, and the inspection priority of the corresponding power distribution device is increased and the corresponding target inspection route is readjusted.

[0036] Further, the method further includes:

[0037] In response to the completion of the inspection tasks of the unmanned vehicle and the robotic dog, according to the inspection data corresponding to the unmanned vehicle and the robotic dog, a power distribution room equipment status report is generated, and according to the power distribution room equipment status report, a corresponding power distribution room operation and maintenance strategy is generated.

[0038] In a second aspect, an embodiment of the present invention provides a power distribution room intelligent inspection control system based on robot collaboration, where the robot includes an unmanned vehicle and a robotic dog; the system includes:

[0039] The patrol sequence generation module is used to obtain the device information of all power distribution devices in the substation, and generate a device patrol sequence based on the preset patrol priority algorithm according to the device information; the device information includes the device operation status and the device location information;

[0040] The patrol task allocation module is used to allocate patrol tasks based on the device patrol sequence according to the movement capabilities of the unmanned vehicle and the robot dog, respectively generate the target patrol routes corresponding to the unmanned vehicle and the robot dog, and push the target patrol routes corresponding to the unmanned vehicle and the robot dog to the corresponding robot control terminals respectively;

[0041] The patrol task execution module is used to control the unmanned vehicle and the robot dog to execute the corresponding patrol tasks simultaneously by the robot control terminal, and optimize and adjust the target patrol routes corresponding to the unmanned vehicle and the robot dog according to the dynamically monitored dynamic obstacle information and electromagnetic interference information until the patrol task is completed.

[0042] The present invention provides a method and system for intelligent patrol control of a substation based on robot collaboration. By the method, after obtaining the device information including the device operation status and the device location information of all power distribution devices in the substation, generating a device patrol sequence based on the preset patrol priority algorithm according to the device information, allocating patrol tasks based on the device patrol sequence according to the movement capabilities of the unmanned vehicle and the robot dog, respectively generating the target patrol routes corresponding to the unmanned vehicle and the robot dog, pushing the target patrol routes corresponding to the unmanned vehicle and the robot dog to the corresponding robot control terminals respectively, and controlling the unmanned vehicle and the robot dog to execute the corresponding patrol tasks simultaneously by the robot control terminal, and optimizing and adjusting the target patrol routes corresponding to the unmanned vehicle and the robot dog according to the dynamically monitored dynamic obstacle information and electromagnetic interference information until the patrol task is completed. Compared with the prior art, this intelligent patrol control method for a substation based on robot collaboration combines a task sorting mechanism for generating a patrol sequence based on a comprehensive analysis of the device operation status and the device location information, a spatial dimension patrol task allocation mechanism designed based on the movement capabilities of the unmanned vehicle and the robot dog and the regional location of power equipment, and a patrol route dynamic adjustment mechanism based on dynamic obstacle and electromagnetic interference information. It can not only achieve efficient and reliable collaboration in task allocation and path planning among different robots based on the patrol task priority, but also perceive the dynamic changes of the patrol environment in real time to adaptively adjust and optimize the patrol route, thereby effectively improving the patrol efficiency and patrol effect and providing a reliable decision-making basis for equipment maintenance. Description of the Drawings

[0043] Figure 1 is a schematic flowchart of the method for intelligent patrol control of a substation based on robot collaboration in an embodiment of the present invention;

[0044] Figure 2 It is a schematic structural diagram of the intelligent inspection and control system for a power distribution room based on robot collaboration in an embodiment of the present invention. Specific implementation manners

[0045] In order to make the objectives, technical solutions and beneficial effects of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Obviously, the following described embodiments are part of the embodiments of the present invention and are only used to illustrate the present invention, but not to limit the scope of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts fall within the scope of protection of the present invention.

[0046] The intelligent inspection and control method for a power distribution room based on robot collaboration provided by the present invention can be understood as an intelligent inspection control solution that can effectively improve the inspection efficiency and inspection effect through efficient collaboration based on task allocation and path planning between different robots, in view of the current application situation where only a single robot is used for intelligent inspection of a power distribution room and there is insufficient inspection efficiency. The following embodiments will detail the intelligent inspection and control method for a power distribution room based on robot collaboration of the present invention.

[0047] In one embodiment, as Figure 1 shown, an intelligent inspection and control method for a power distribution room based on robot collaboration is provided. The robot includes an unmanned vehicle and a robot dog, that is, efficient intelligent inspection of the power distribution room is realized through task collaborative control between the unmanned vehicle and the robot dog. The corresponding intelligent inspection and control method for the power distribution room includes the following steps:

[0048] S11. Obtain the device information of all power distribution devices in the power distribution room, and based on the device information, generate a device inspection sequence based on a preset inspection priority algorithm; wherein, the device information includes the device operation state and the device location information, and the device operation state information can be understood as the key parameters that can reflect the device operation state obtained through a state monitoring sensor network pre-deployed in the power distribution room, such as current, voltage, temperature and other information, which is not specifically limited here; the corresponding device location information can be understood as the geographical location coordinate information of each power distribution device in the power distribution room.

[0049] The preset inspection priority algorithm in this embodiment can be understood as an algorithm for determining the inspection urgency of each power distribution device. In order to ensure the efficiency and scientificity of generating the device inspection sequence, this embodiment preferably adopts a method of determining the device inspection priority with the device operation state as the main factor and the device location information as the auxiliary factor. Specifically, the step of generating a device inspection sequence based on the device information and the preset inspection priority algorithm includes:

[0050] According to the device operation status in the device information of each power distribution device, a corresponding device inspection priority is generated based on a preset inspection priority evaluation model; among them, the preset inspection priority evaluation model can be understood as a model for predicting the inspection priority based on device operation status data constructed based on the principle that different device operation statuses correspond to different device health statuses, and different device health statuses correspond to different device inspection priorities; considering the diversity and uncertainty of device operation status data in practical applications, in order to solve the problem of insufficient judgment of the "intermediate state" by the traditional threshold method (such as "slight overload" or "critical high temperature") to ensure the rationality of the inspection priority evaluation, in this embodiment, a corresponding preset inspection priority evaluation model is preferably constructed based on fuzzy logic, that is, by performing fuzzy analysis on parameters such as current, voltage, and temperature, quantifying the device health status, and then outputting the corresponding inspection priority (such as high / medium / low, etc.) based on the device health status. The specific model construction process can be implemented with reference to existing fuzzy logic theories. For example: first, define the fuzzy sets corresponding to each parameter respectively, and design the corresponding membership function (such as a trapezoidal function) to convert each parameter value in the device operation status into the corresponding fuzzy membership degree; then, formulate fuzzy rules driven by expert experience, and optimize the membership function parameters or rule weights based on relevant historical fault data combined with a neuro-fuzzy model (such as an adaptive-network-based fuzzy inference system, ANFIS), and dynamically adjust the rule contribution degree to perform weighted aggregation on the priority output of the triggered rules to obtain the corresponding fuzzy output; finally, use the centroid method to convert the fuzzy output into an exact value to obtain the required inspection priority.

[0051] According to the device inspection priorities of each power distribution device, all power distribution devices are sorted based on the priority descending order method to obtain an initial device inspection sequence; that is, the initial device inspection sequence can be understood as an inspection task execution sequence generated by sorting the power distribution devices based on the principle that the higher the device inspection priority, the more forward the inspection task is sorted.

[0052] Judge whether there are power distribution devices with the same device inspection priority in the initial device inspection sequence; that is, check whether there is a situation in the initial device inspection sequence where the generation of the device inspection order is unreasonable due to the same device inspection priority.

[0053] If it exists, according to the device location information of the power distribution devices with the same device inspection priority, the shortest path algorithm is used to sort and optimize the power distribution devices with the same device inspection priority in the initial device inspection sequence to obtain the device inspection sequence; among them, the process of using the shortest path algorithm to sort and optimize the power distribution devices with the same device inspection priority in the initial device inspection sequence can be understood as the process of finding the shortest non-repeating open path of all power distribution devices with the same device inspection priority according to the device location based on dynamic programming or approximate / heuristic algorithms. Specifically, it can be implemented with reference to the existing shortest open path acquisition technology based on fixed-position path nodes. For example, if the number of power distribution devices with the same device inspection priority is too large (such as more than 20) in practical applications, the heuristic algorithm is preferably used; otherwise, the dynamic programming algorithm can be selected, which will not be elaborated here. After obtaining the sub-inspection sequences corresponding to each power distribution device with the same device inspection priority, use each sub-inspection sequence to replace the inspection sorting of the corresponding power distribution device in the initial device inspection sequence, and finally obtain the required device inspection sequence.

[0054] If it does not exist, the initial device inspection sequence is used as the device inspection sequence; that is, when the inspection priorities of all power distribution devices in the initial device inspection sequence are different, the obtained initial device inspection sequence can be used as the required device inspection sequence without adjustment.

[0055] S12. According to the motion capabilities of the unmanned vehicle and the robot dog, based on the device inspection sequence, the inspection tasks are allocated, and the target inspection routes corresponding to the unmanned vehicle and the robot dog are respectively generated, and the target inspection routes corresponding to the unmanned vehicle and the robot dog are respectively pushed to the corresponding robot control terminals; among them, the motion capabilities can be understood as the characteristics that can reflect the differences between the applicable motion areas of the unmanned vehicle and the robot dog, preferably including motion speed, endurance time, and minimum turning radius. For example, the unmanned vehicle has a higher running speed and a longer endurance time, and is suitable for being responsible for the inspection of devices in the open area of the substation, while the robot dog has a smaller minimum turning radius, has flexibility and the ability to pass through narrow channels, and is suitable for being responsible for the inspection of devices in the narrow channel area of the substation.

[0056] To ensure that the respective motion ability advantages of the unmanned vehicle and the robot dog can be fully utilized, better realize the cooperation of inspection tasks, and improve the execution efficiency of the overall substation inspection task as much as possible, this embodiment preferably allocates the inspection tasks of the power distribution devices involved in the device inspection sequence based on the types of inspection areas suitable for the unmanned vehicle and the robot dog to ensure the task cooperation in space between the two; specifically, the steps of allocating the inspection tasks based on the motion capabilities of the unmanned vehicle and the robot dog and generating the target inspection routes corresponding to the unmanned vehicle and the robot dog based on the device inspection sequence include:

[0057] Based on the movement capabilities of the unmanned vehicle and the robotic dog, the equipment area classification is obtained; among them, the equipment area classification can be understood as the types of areas around power distribution equipment that are respectively applicable to the inspection of the unmanned vehicle and the robotic dog based on their movement characteristics, including the open area type and the narrow area type. The open area type can be understood as the area type where the width of the movable passage around the power distribution equipment is greater than or equal to the preset width threshold, and the corresponding narrow area type can be understood as the area type where the width of the movable passage around the power distribution equipment is less than the preset width threshold; it should be noted that the preset width threshold can be determined based on the minimum turning radius of the robotic dog. For example, if the minimum turning radius of the robotic dog is 0.3 meters, the preset width threshold can be set to 0.8 meters, and no specific limitation is made here.

[0058] Based on the equipment location information of all power distribution equipment in the equipment inspection sequence, spatial clustering is performed based on the equipment area classification to generate an open area equipment cluster and a narrow area equipment cluster; among them, spatial clustering can be understood as DBSCAN clustering based on the spatial distribution density of the equipment in the substation. The specific process of generating the open area equipment cluster and the narrow area equipment cluster includes: calculating the Euclidean distance matrix between equipment according to the equipment location information (plane two-dimensional coordinates) of each power distribution equipment; calculating the average distance from each power distribution equipment to the nearest several neighbor equipment according to the Euclidean distance matrix between equipment. If the average distance is greater than the preset width threshold, it is considered that the power distribution equipment is located in the open area, otherwise, it is located in the narrow area; using the preset width threshold as the neighborhood radius, after determining the minimum number of neighbors, clustering analysis is performed to obtain the required open area equipment cluster (high-density cluster) and narrow area equipment cluster (low-density cluster).

[0059] Based on the open area equipment cluster and the narrow area equipment cluster, the corresponding unmanned vehicle inspection equipment set and robotic dog inspection equipment set are respectively constructed; among them, the power distribution equipment in the unmanned vehicle inspection equipment set belongs to the power distribution equipment in the open area equipment cluster, and the corresponding power distribution equipment in the robotic dog inspection equipment set belongs to the power distribution equipment in the narrow area equipment cluster. That is, through spatial clustering that combines the movement ability characteristics of the robot, a reliable division of the equipment inspection task inside the substation is realized, which is convenient for improving the execution efficiency of the inspection task.

[0060] Based on the equipment inspection sorting in the equipment inspection sequence, the corresponding unmanned vehicle inspection task sequence and robotic dog inspection task sequence are respectively generated for the unmanned vehicle inspection equipment set and the robotic dog inspection equipment set; among them, the unmanned vehicle inspection task sequence can be understood as the priority order of the unmanned vehicle inspection equipment obtained according to the sorting relationship of each power distribution equipment in the unmanned vehicle inspection equipment set in the equipment inspection sequence; the corresponding robotic dog inspection task sequence can be understood as the priority order of the robotic dog inspection equipment obtained according to the sorting relationship of each power distribution equipment in the robotic dog inspection equipment set in the equipment inspection sequence.

[0061] Taking the unmanned vehicle inspection task sequence and the robot dog inspection task sequence as order constraints respectively, based on a preset path optimization algorithm and a substation grid map, generate the corresponding target inspection routes for the unmanned vehicle and the robot dog respectively; wherein, the substation grid map can be understood as an environmental map obtained by dividing the substation into multiple grid cells based on a preset grid cell size (such as 0.5m * 0.5m), and the corresponding target inspection routes for the unmanned vehicle and the robot dog can be understood as the equipment inspection task execution routes generated based on the environmental map as the analysis basis and executing the path optimization algorithm under the condition of order constraints.

[0062] In order to adapt to different equipment area types to ensure that both the unmanned vehicle and the robot dog can complete all the substation equipment inspection tasks in the corresponding inspection task sequence in the shortest time, this embodiment preferably uses different path optimization algorithms for different robot path finding; specifically, the steps of taking the unmanned vehicle inspection task sequence and the robot dog inspection task sequence as order constraints respectively, and generating the corresponding target inspection routes for the unmanned vehicle and the robot dog based on a preset path optimization algorithm and a substation grid map include:

[0063] According to the preset path planning constraint conditions, based on the A* algorithm and the substation grid map, generate the target inspection route of the unmanned vehicle; the preset path planning constraint conditions include order constraints, equipment distribution constraints, and channel width constraints; wherein, the equipment distribution constraints can be understood as information for avoiding static obstacles in the substation, and the channel width constraints can be understood as information for screening suitable channels for the unmanned vehicle to move; the specific process of obtaining the target inspection route of the unmanned vehicle can be implemented with reference to the execution process of the existing A* algorithm, and may include: on the substation grid map, according to the position information of all equipment in the substation and the marked fixed obstacle cells (not passable) such as walls, and further mark the non-passable areas of the substation grid map based on the channel width constraint applicable to the unmanned vehicle (channels with a width less than 0.8m are not passable) to obtain the marked substation grid map (the information on whether each grid cell is allowed to pass can be stored in the form of a two-dimensional matrix); then, based on the order constraints, divide the target inspection route to be searched into multiple sub-paths, and based on the marked substation grid map, independently run the A* algorithm on each sub-path with the grid distance that the sub-path needs to move as the actual cost and the Euclidean distance as the heuristic function to obtain the corresponding initial path, and then perform B-spline interpolation on the generated initial path to generate the required target sub-path; finally, merge the various target sub-paths according to the endpoint information of each sub-path to obtain the required target inspection route of the unmanned vehicle.

[0064] According to the preset path planning constraint conditions, based on the breadth-first search algorithm and the grid map of the distribution room, generate the target inspection route of the robot dog; wherein, the preset path planning constraint conditions include sequential constraint, equipment distribution constraint, and passage width constraint as described above. The equipment distribution constraint can be understood as information for avoiding static obstacles in the distribution room, and the passage width constraint can be understood as information for screening passages suitable for the movement of the robot dog; the specific process of obtaining the target inspection route of the unmanned vehicle can be implemented with reference to the execution process of the existing breadth-first search algorithm (BFS), and may include: on the grid map of the distribution room, fix the obstacle cells (marked as impassable) according to the position information of all equipment in the distribution room and marks such as walls, and further mark the impassable areas of the grid map of the distribution room based on the passage width constraint applicable to the movement of the robot dog (such as determining the inspection width mark based on the passage width, the width of the robot dog, and the safety threshold), to obtain the marked grid map of the distribution room (the marked information corresponding to each grid cell can be stored in the form of a two-dimensional matrix); then, divide the target inspection route to be searched into multiple sub-paths based on the sequential constraint, and based on the marked grid map of the distribution room, independently run the breadth-first search algorithm on each sub-path (allowing the setting of 8 neighborhood directions including diagonals to simulate the flexible movement of the robot dog, and adding the verification of the narrow passage width with the marked inspection width) to obtain the corresponding initial path, and then perform B-spline interpolation on the generated initial path to generate the required target sub-path; finally, merge the various target sub-paths according to the endpoint information of each sub-path to obtain the required target inspection route of the robot dog.

[0065] S13. The robot control terminal simultaneously controls the unmanned vehicle and the robot dog to perform corresponding inspection tasks, and optimally adjusts the target inspection routes corresponding to the unmanned vehicle and the robot dog according to the dynamically monitored obstacle information and electromagnetic interference information until the inspection task is completed; wherein, the robot control terminal simultaneously controlling the unmanned vehicle and the robot dog to perform corresponding inspection tasks can be understood as the control terminals of the unmanned vehicle and the robot dog respectively performing corresponding inspection tasks based on the corresponding target inspection routes. To ensure the safety and reliability of the unmanned vehicle and the robot dog in performing inspection tasks, in this embodiment, preferably in combination with the analysis of the environmental complexity of the distribution room, during the process of the unmanned vehicle and the robot dog performing inspection tasks, they respectively sense the environmental information in real time based on the lidar, infrared sensor, electromagnetic interference detector, etc. configured on themselves, so as to adaptively optimize and adjust the inspection path. Specifically, the step of optimally adjusting the target inspection routes corresponding to the unmanned vehicle and the robot dog according to the dynamically monitored obstacle information and electromagnetic interference information includes:

[0066] When it is determined according to the dynamic obstacle information that a dynamic obstacle needs to be avoided, the activity area of the dynamic obstacle is used as the current robot path dangerous area; wherein, the dynamic obstacle information may include information such as the position, moving speed, and moving direction of the dynamic obstacle, and its motion trajectory can be predicted based on the timing information of the dynamic obstacle, and the shortest distance between the robot and the motion trajectory of the dynamic obstacle can be calculated according to the current position information of the robot; at the same time, the expected collision duration is calculated according to the relative speed between the dynamic obstacle and the robot, and the minimum safe distance of the robot is calculated according to the braking ability (maximum deceleration) of the robot, and then it is determined whether the expected collision duration is greater than the preset safety duration threshold and the shortest distance is greater than the minimum safe distance. If so, the activity area of the dynamic obstacle (the activity area can be determined with the dynamic obstacle as the center and the minimum safe distance as the radius) is used as the dangerous area that needs to be avoided, and the path replanning mechanism is immediately triggered. It should be noted that when determining the mobile obstacle information, in order to ensure the accuracy of environmental perception, the Kalman Filter can be used to fuse the radar monitoring data and the infrared sensor data to improve the accuracy and robustness of the target state estimation. The specific processing can be implemented with reference to the relevant existing technologies and will not be elaborated here.

[0067] When it is determined according to the electromagnetic interference information that the electromagnetic interference area needs to be avoided, the electromagnetic interference area is used as the current robot path dangerous area; wherein, the electromagnetic interference information includes the electromagnetic interference intensity. If the electromagnetic interference intensity exceeds the preset intensity threshold, the path replanning mechanism is immediately triggered.

[0068] According to the current robot path dangerous area information and the current robot position information, with the optimization objectives of minimizing the path length, minimizing the number of turns, and avoiding the path dangerous area, the corresponding target inspection route is optimized, and the adjusted and optimized inspection route is updated to the corresponding robot control terminal; optimizing the target inspection route can be understood as optimizing the path of the remaining inspection tasks on the target inspection route based on the current robot position information while keeping the task order unchanged. The specific implementation process may include: constructing a corresponding fitness function with the optimization objectives of minimizing the path length (to improve the inspection efficiency), minimizing the number of turns (the number of direction changes to avoid the problem of spinning in place), and avoiding the path dangerous area (the distance between the path and the obstacle is greater than the safety threshold). After determining the fitness function, first use the A* algorithm to generate an initial feasible path, and then based on the NSGA-II algorithm, optimize and adjust the initial feasible path according to the fitness function to obtain the required adjusted and optimized inspection route; preferably, the fitness function in this embodiment is expressed as:

[0069]

[0070] In the formula, , and respectively represent the path length, the number of turns, and the normalized data corresponding to the proximity to the dangerous area of the inspection route . The specific calculation expressions can be obtained with reference to the existing related technologies and are not specifically limited here; , and represent weight coefficients, which can be determined according to actual application requirements.

[0071] It should be noted that based on the NSGA-II algorithm, the process of optimizing and adjusting the initial feasible path according to the fitness function can be implemented with reference to the existing related technologies for path optimization based on the NSGA-II algorithm and will not be elaborated here.

[0072] The technical solution provided by the embodiment of the present invention to obtain the device information including the device operation status and device location information of all power distribution devices in the power distribution room, generate a device inspection sequence based on a preset inspection priority algorithm according to the device information, and then perform inspection task allocation based on the device inspection sequence according to the movement capabilities of the unmanned vehicle and the robot dog, respectively generate the target inspection routes corresponding to the unmanned vehicle and the robot dog, push the target inspection routes corresponding to the unmanned vehicle and the robot dog to the corresponding robot control terminals respectively, and control the unmanned vehicle and the robot dog to execute the corresponding inspection tasks by the robot control terminals at the same time, and optimize and adjust the target inspection routes corresponding to the unmanned vehicle and the robot dog according to the dynamically monitored dynamic obstacle information and electromagnetic interference information until the inspection task is completed can not only achieve efficient and reliable cooperation of task allocation and path planning among different machines based on the inspection task priority, but also can perceive the dynamic changes of the inspection environment in real time to adaptively adjust and optimize the inspection route, thereby effectively improving the inspection efficiency and inspection effect and providing a reliable decision-making basis for equipment maintenance.

[0073] Considering that the battery life of the robot dog is not high, in order to ensure that the robot dog can normally complete the inspection tasks in the responsible area, this embodiment preferably predicts the remaining power of the robot dog in real time to evaluate whether energy needs to be replenished; specifically, the method further includes:

[0074] According to the battery capacity and the preset power consumption rate of the robotic dog, the remaining power of the robotic dog is predicted in real time, and according to the current position information and the remaining inspection route of the robotic dog, the remaining task required power is obtained; wherein, the remaining power of the robotic dog can be understood as the available power value calculated based on the battery capacity, the power consumption rate, and the inspected duration. Considering that the actual power consumption rate of the robotic dog is related to the environmental temperature, the motion state, and the workload, in order to ensure the accuracy of the remaining power prediction, in this embodiment, preferably, a comprehensive analysis is performed based on the environmental temperature, the motion state, and the load condition of the robotic dog during operation to obtain a reasonable power consumption rate for use in the remaining power prediction; specifically, the steps for obtaining the preset power consumption rate include:

[0075] Obtain the current environmental temperature, motion state, and workload of the robotic dog, and according to the current environmental temperature, the motion state, the workload, and the preset power consumption rate prediction model, obtain the preset power consumption rate; the preset power consumption rate prediction model is obtained based on a linear regression analysis of the historical power consumption rate data of the robotic dog under different temperatures, different motion states, and different workloads; wherein, the motion state includes fast walking state, slow walking state, rest state, etc., and the workload can be understood as the enabling situation of the sensors on the robotic dog, such as only using one of the infrared sensor, lidar, and electromagnetic interference detector, or combining any number of sensors for use, etc., corresponding to different load states. In practical applications, the power consumption rate data of the robotic dog under different temperatures, different motion states, and different workloads can be pre-obtained, and a linear regression analysis is performed with the environmental temperature, motion state, and workload as independent variables (the motion state and workload are quantitatively represented by numerical variables) and the power consumption rate as the dependent variable to construct the required preset power consumption rate prediction model; during the inspection process of the robotic dog, by obtaining the corresponding environmental temperature, motion state, and workload in real time and inputting them into the corresponding preset power consumption rate prediction model for power consumption rate prediction, the preset power consumption rate that conforms to the current state of the robotic dog can be obtained.

[0076] When it is determined that the remaining power is less than the remaining task required power, based on the current position information and the position information of the charging equipment in the power distribution room, the optimal charging path for the robot dog is generated based on the Dijkstra algorithm and pushed to the corresponding robot control terminal. After the robot dog is charged to reach the remaining task required power, the corresponding target inspection route is readjusted according to the latest position information of the robot dog to execute the remaining inspection tasks; among them, the remaining task required power can be analyzed based on the inspection path length of the remaining task combined with the corresponding power consumption rate. For example, the estimated inspection duration required can be obtained based on the inspection path length combined with the movement speed of the robot dog. Then, the remaining task required power can be estimated based on the product of the estimated inspection duration required and the power consumption rate. When the remaining power of the robot dog is not enough to support the completion of the remaining inspection tasks and charging is required, it is necessary to search for the charging equipment closest to it in the power distribution room according to the current position information of the robot dog, and use the path planning algorithm with the closest charging equipment as the target point and avoiding obstacles and electromagnetic interference areas as the constraint conditions to generate the optimal charging path for the robot dog to move forward to the closest charging equipment. And when the robot dog is charged to a certain degree that the power exceeds the remaining required power, the charging can be ended, and the inspection path corresponding to the remaining inspection tasks is replanned according to the current charging position and the equipment position information corresponding to the remaining inspection tasks. After optimizing the target inspection route stored in the current control terminal, the corresponding inspection tasks are continued based on the adjusted target inspection path. The specific method for obtaining the optimal charging path and the method for optimizing and obtaining the target inspection path can both be realized based on the process of optimizing and adjusting the target inspection route according to the dynamic obstacle information and electromagnetic interference information mentioned above, which will not be elaborated here.

[0077] It should be noted that during the charging process of the robotic dog in this embodiment, the patrol task of the robotic dog is in a stagnant state, which will to a certain extent affect the execution efficiency of the overall patrol task. To avoid the lag in the execution of emergency patrol tasks caused by the charging of the robotic dog, in this embodiment, after the robotic dog starts the charging program, the current remaining patrol tasks of the robotic dog and the unmanned vehicle are respectively obtained, and based on the task priority, the current remaining patrol tasks of the two are merged (when merging, it is necessary to add a robot mark to which the task belongs, and optimize the sorting of devices with the same patrol priority according to the location information of the devices). According to the combined comprehensive remaining patrol task sequence, the patrol route of the unmanned vehicle is readjusted. After the robotic dog finishes charging, the remaining patrol tasks in the comprehensive remaining patrol tasks are divided into two remaining patrol task sequences corresponding to the unmanned vehicle and the robotic dog according to the corresponding robot marks, and the patrol routes of the two remaining patrol task sequences are planned based on the aforementioned path planning algorithm, and the newly generated patrol routes are respectively pushed to the unmanned vehicle and the robotic dog, so that they continue to execute the remaining patrol tasks according to the latest patrol routes, so as to effectively ensure the continuity of the patrol task and improve the overall patrol efficiency in a seamless connection manner in terms of the patrol time of the unmanned vehicle and the robotic dog.

[0078] In addition, considering that in actual applications, there may be a situation where grid equipment that has not been patrolled suddenly malfunctions. To facilitate the timely confirmation of the abnormal state and maintenance of the abnormal equipment and ensure power supply safety, preferably in this embodiment, during the process of the unmanned vehicle and the robotic dog collaborating to execute the patrol task, the operation status of the grid equipment to be patrolled in the remaining patrol tasks is also monitored to timely detect equipment abnormalities and adaptively adjust the corresponding patrol task priorities to avoid potential safety risks; specifically, the method further includes:

[0079] During the execution of the inspection tasks by the driverless vehicle and the robotic dog, the operation status monitoring data of each power distribution device in the corresponding target inspection route is obtained in real time. When it is determined that the operation status monitoring data is abnormal, a fault alarm message is generated, the inspection priority of the corresponding power distribution device is increased, and the corresponding target inspection route is readjusted. Among them, the operation status monitoring data can be understood to include data such as current, voltage, and temperature. Based on the operation status monitoring data obtained in real time, it is analyzed whether each data is within the preset normal range. If the above conditions cannot be met, for example, the rated current is 10A while the actually collected current value is 15A, or the equipment temperature safety threshold is 80°C while the actually collected surface temperature of the equipment is 80°C, then the operation status monitoring data can be analyzed for abnormalities based on the pre-constructed Bayesian network model, and it is determined whether the power distribution device has an abnormal operation according to the corresponding output device failure probability. When it is determined that there is an abnormal operation, the corresponding fault alarm mechanism is triggered. At the same time, the inspection priority of the corresponding power distribution device is adjusted to the highest level, and according to the adjusted inspection priority, the aforementioned shortest path algorithm (such as using the Dijkstra algorithm) is combined with the real-time environment map to regenerate the remaining inspection task route and push it to the corresponding control terminal. The Bayesian network model in this embodiment can be trained and constructed based on the historical data of different operation status monitoring data and the corresponding failure probabilities. The specific construction process can be implemented with reference to relevant existing technologies and will not be elaborated here. In addition, in order to avoid potential equipment operation risks, this embodiment can also predict potential faulty equipment in advance by learning the historical fault data through the support vector machine algorithm. For example, if it is predicted from the historical data that a certain disconnector is prone to failure in a high-temperature environment, the inspection priority of the corresponding equipment can be further increased, so that the inspection task of the corresponding equipment is executed in advance.

[0080] The method provided by the embodiment of the present invention for real-time monitoring and analyzing the operation status data of power distribution devices and timely adjusting the inspection task level and updating the inspection task execution order based on the corresponding device status analysis results can ensure that abnormal situations of power distribution devices are promptly responded to and processed, thereby effectively guaranteeing the stable operation of the power system.

[0081] In addition, in order to further improve the safety and stability of the operation of the substation, this embodiment preferably performs a status analysis on each power distribution device based on the corresponding inspection data of the driverless vehicle and the robotic dog, makes a maintenance decision according to the corresponding analysis results, and generates a corresponding substation operation and maintenance strategy. Specifically, the method further includes:

[0082] In response to the completion of the inspection tasks of the driverless vehicle and the robotic dog, generate a power distribution room equipment status report based on the inspection data corresponding to the driverless vehicle and the robotic dog, and generate a corresponding power distribution room operation and maintenance strategy based on the power distribution room equipment status report; among them, the inspection data varies according to different power distribution equipment. For example, for a transformer, the corresponding inspection data includes winding temperature, oil temperature, operating noise, vibration amplitude, oil level, bushing and insulator status, etc.; for high-voltage switchgear, it includes electrical parameters, mechanical status and insulation performance, etc.; the power distribution room equipment status report generated based on the analysis of inspection data may include the inspection data of each power distribution equipment, the equipment status analysis results generated based on the analysis of inspection data, maintenance suggestions and maintenance priorities. The process of analyzing the status of the corresponding power distribution equipment based on the inspection data can be understood as first performing preprocessing such as cleaning, denoising and normalization on the inspection data to be analyzed of each power distribution equipment, and then using a neural network model constructed based on relevant historical data for status identification. After obtaining the corresponding status analysis results, the status analysis results of each power distribution equipment can be analyzed based on the existing association rule analysis algorithm (Apriori algorithm) to generate corresponding maintenance suggestions and maintenance priorities; after generating the power distribution room equipment status report based on the analysis of inspection data, the maintenance suggestions and maintenance priorities of all power distribution equipment in the power distribution room equipment status report can also be comprehensively analyzed based on the reinforcement learning model constructed in advance based on relevant historical data to generate the corresponding power distribution room operation and maintenance strategy; it should be noted that the association rules used in the association rule analysis algorithm and the construction of the reinforcement learning model in this embodiment can be implemented with reference to relevant existing technologies according to actual application requirements, which will not be elaborated here.

[0083] Through the efficient and intelligent analysis of the inspection data in this embodiment, not only can the accurate assessment of the operating status of power distribution equipment be realized, but also a reliable power distribution room operation and maintenance strategy can be automatically generated based on the operating status of the equipment in the entire power distribution room, providing a guiding basis for the timely operation and maintenance of the power distribution room, and thus providing a strong guarantee for the safe and reliable operation of the entire power distribution room.

[0084] It should be noted that although the steps in the above flow chart are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders.

[0085] In one embodiment, as Figure 2 shown, a power distribution room intelligent inspection control system based on robot collaboration is provided. The robot includes a driverless vehicle and a robotic dog; the system includes:

[0086] The patrol sequence generation module 1 is configured to obtain the device information of all power distribution devices in the power distribution room, and generate a device patrol sequence based on the preset patrol priority algorithm according to the device information; the device information includes the device operation status and the device location information;

[0087] The patrol task allocation module 2 is configured to allocate patrol tasks based on the device patrol sequence according to the movement capabilities of the unmanned vehicle and the robotic dog, generate the target patrol routes corresponding to the unmanned vehicle and the robotic dog respectively, and push the target patrol routes corresponding to the unmanned vehicle and the robotic dog to the corresponding robot control terminals respectively;

[0088] The patrol task execution module 3 is configured to control the unmanned vehicle and the robotic dog to execute the corresponding patrol tasks simultaneously by the robot control terminal, and optimize and adjust the target patrol routes corresponding to the unmanned vehicle and the robotic dog according to the dynamically monitored dynamic obstacle information and electromagnetic interference information until the patrol task is completed.

[0089] For the specific limitations of the intelligent patrol control system for the power distribution room based on robot collaboration, reference can be made to the limitations of the intelligent patrol control method for the power distribution room based on robot collaboration in the above text, and the corresponding technical effects can also be equivalently obtained, which will not be elaborated here. Each module in the above intelligent patrol control system for the power distribution room based on robot collaboration can be implemented in whole or in part by software, hardware and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to the above modules.

[0090] In summary, the intelligent patrol control method and system for a distribution room based on robot collaboration provided by the embodiments of the present invention implement obtaining device information including the device operation status and device location information of all power distribution devices in the distribution room, and based on the device information, generating a device patrol sequence based on a preset patrol priority algorithm. Then, according to the movement capabilities of the unmanned vehicle and the robot dog, the patrol tasks are assigned based on the device patrol sequence, and the corresponding target patrol routes for the unmanned vehicle and the robot dog are generated respectively. The corresponding target patrol routes for the unmanned vehicle and the robot dog are pushed to the corresponding robot control terminals respectively, and the robot control terminals simultaneously control the unmanned vehicle and the robot dog to execute the corresponding patrol tasks. And according to the dynamically monitored dynamic obstacle information and electromagnetic interference information, the target patrol routes corresponding to the unmanned vehicle and the robot dog are optimized and adjusted until the patrol task is completed. The technical solution of this method combines a task sorting mechanism for generating a patrol sequence based on the comprehensive analysis of the device operation status and device location information, a spatial dimension patrol task allocation mechanism designed based on the movement capabilities of the unmanned vehicle and the robot dog and the regional location of power equipment, and a patrol route dynamic adjustment mechanism based on dynamic obstacle and electromagnetic interference information. It can not only achieve efficient and reliable collaboration in task allocation and path planning among different robots based on the patrol task priority, but also can perceive the dynamic changes of the patrol environment in real time to adaptively adjust and optimize the patrol route, thereby effectively improving the patrol efficiency and patrol effect, and providing a reliable decision-making basis for equipment maintenance.

[0091] Each embodiment in this specification is described in a progressive manner. For parts that are the same or similar in each embodiment, they can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0092] The above embodiments only represent several preferred implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be pointed out that for those of ordinary skill in the art in this technical field, without departing from the technical principle of the present invention, several improvements and substitutions can still be made, and these improvements and substitutions should also be regarded as the protection scope of the present invention. Therefore, the protection scope of the invention patent should be subject to the protection scope of the claims.

Claims

1. An intelligent inspection and control method for a distribution substation based on robot collaboration, characterized in that The robot includes an autonomous vehicle and a robotic dog; the method includes the following steps: Obtain the device information of all power distribution devices in the power distribution room, and based on the device information, generate a device inspection sequence based on a preset inspection priority algorithm; the device information includes the device operation status and the device location information; Based on the movement capabilities of the autonomous vehicle and the robotic dog, perform inspection task allocation based on the device inspection sequence, respectively generate the target inspection routes corresponding to the autonomous vehicle and the robotic dog, and push the target inspection routes corresponding to the autonomous vehicle and the robotic dog to the corresponding robot control terminals respectively; The robot control terminal simultaneously controls the autonomous vehicle and the robotic dog to perform the corresponding inspection tasks, and based on the dynamically monitored obstacle information and electromagnetic interference information, optimize and adjust the target inspection routes corresponding to the autonomous vehicle and the robotic dog until the inspection tasks are completed; Among them, the step of generating a device inspection sequence based on a preset inspection priority algorithm according to the device information includes: Based on the device operation status in the device information of each power distribution device, generate the corresponding device inspection priority based on a preset inspection priority evaluation model; the preset inspection priority evaluation model is constructed based on fuzzy logic; Based on the device inspection priorities of all power distribution devices, sort all power distribution devices using the priority descending sorting method to obtain an initial device inspection sequence; Judge whether there are power distribution devices with the same device inspection priority in the initial device inspection sequence; If so, based on the device location information of the power distribution devices with the same device inspection priority, use the shortest path algorithm to sort and optimize the power distribution devices with the same device inspection priority in the initial device inspection sequence to obtain the device inspection sequence; If not, use the initial device inspection sequence as the device inspection sequence; The movement capabilities include movement speed, endurance time, and minimum turning radius; the step of performing inspection task allocation based on the device inspection sequence according to the movement capabilities of the autonomous vehicle and the robotic dog, and respectively generating the target inspection routes corresponding to the autonomous vehicle and the robotic dog includes: Based on the movement capabilities of the autonomous vehicle and the robotic dog, obtain device area classifications; the device area classifications include open area types and narrow area types; Based on the device location information of all power distribution devices in the device inspection sequence, perform spatial clustering based on the device area classifications to generate an open area device cluster and a narrow area device cluster; Based on the open area device cluster and the narrow area device cluster, respectively construct corresponding autonomous vehicle inspection device sets and robotic dog inspection device sets; Based on the device inspection sorting in the device inspection sequence, respectively generate an autonomous vehicle inspection task sequence and a robotic dog inspection task sequence corresponding to the autonomous vehicle inspection device set and the robotic dog inspection device set; Respectively using the autonomous vehicle inspection task sequence and the robotic dog inspection task sequence as sequence constraints, based on a preset path optimization algorithm and a power distribution room grid map, respectively generate the target inspection routes corresponding to the autonomous vehicle and the robotic dog.

2. The intelligent inspection and control method for a distribution substation based on robot cooperation according to claim 1, characterized in that, The steps of respectively generating the target inspection routes corresponding to the unmanned vehicle and the robot dog based on the preset path optimization algorithm and the substation grid map, with the inspection task sequences of the unmanned vehicle and the robot dog as the sequence constraints, include: Generating the target inspection route of the unmanned vehicle based on the A* algorithm and the substation grid map according to the preset path planning constraint conditions; the preset path planning constraint conditions include sequence constraints, equipment distribution constraints, and channel width constraints; Generating the target inspection route of the robot dog based on the breadth-first search algorithm and the substation grid map according to the preset path planning constraint conditions.

3. The intelligent inspection and control method for a distribution substation based on robot collaboration according to claim 1, wherein, The steps of optimizing and adjusting the target inspection routes corresponding to the unmanned vehicle and the robot dog according to the dynamically monitored dynamic obstacle information and electromagnetic interference information include: When it is determined according to the dynamic obstacle information that the dynamic obstacle needs to be avoided, the activity area of the dynamic obstacle is used as the current robot path dangerous area; When it is determined according to the electromagnetic interference information that the electromagnetic interference area needs to be avoided, the electromagnetic interference area is used as the current robot path dangerous area; According to the current robot path dangerous area information and the current robot position information, with the goals of minimizing the path length, minimizing the number of turns, and avoiding the path dangerous area, the corresponding target inspection route is optimized, and the adjusted and optimized inspection route is updated to the corresponding robot control terminal.

4. The intelligent inspection and control method for a distribution substation based on robot collaboration according to claim 1, characterized in that, The method further includes: According to the battery capacity of the robot dog and the preset power consumption rate, the remaining power of the robot dog is predicted in real time, and according to the current position information and the remaining inspection route of the robot dog, the remaining task required power is obtained; When it is determined that the remaining power is less than the remaining task required power, based on the Dijkstra algorithm, the optimal charging path corresponding to the robot dog is generated according to the current position information and the substation charging equipment position information and pushed to the corresponding robot control terminal, and after the robot dog is charged to reach the remaining task required power, the corresponding target inspection route is readjusted according to the latest position information of the robot dog to execute the remaining inspection tasks.

5. The intelligent inspection and control method for a power distribution room based on robot collaboration according to claim 4, characterized in that The steps of obtaining the preset power consumption rate include: Obtaining the current environmental temperature, motion state, and workload of the robot dog, and according to the current environmental temperature, the motion state, the workload, and the preset power consumption rate prediction model, obtaining the preset power consumption rate; the preset power consumption rate prediction model is obtained based on the linear regression analysis of the historical power consumption rate data of the robot dog under different temperatures, different motion states, and different workloads.

6. The intelligent inspection and control method for a power distribution room based on robot collaboration according to claim 1, wherein The method further includes: During the process of the unmanned vehicle and the robot dog performing the inspection tasks, the operation state monitoring data of each power distribution equipment in the corresponding target inspection route is obtained in real time, and when it is determined that the operation state monitoring data is abnormal, a fault alarm message is generated, and the inspection priority of the corresponding power distribution equipment is increased and the corresponding target inspection route is readjusted.

7. The intelligent inspection control method for a power distribution room based on robot collaboration according to claim 1, wherein The method further includes: In response to the completion of the inspection tasks of the driverless vehicle and the robotic dog, generate an equipment status report for the power distribution room based on the inspection data corresponding to the driverless vehicle and the robotic dog, and generate a corresponding operation and maintenance strategy for the power distribution room according to the equipment status report of the power distribution room.

8. An intelligent inspection and control system for a distribution substation based on robot collaboration, characterized in that The robot includes a driverless vehicle and a robotic dog; Apply the intelligent inspection control method for the power distribution room based on robot collaboration according to claim 1, the system includes: An inspection sequence generation module, configured to obtain the equipment information of all power distribution equipment in the power distribution room, and generate an equipment inspection sequence based on a preset inspection priority algorithm according to the equipment information; the equipment information includes the equipment operation status and the equipment location information; An inspection task allocation module, configured to allocate inspection tasks based on the equipment inspection sequence according to the movement capabilities of the driverless vehicle and the robotic dog, respectively generate target inspection routes corresponding to the driverless vehicle and the robotic dog, and push the target inspection routes corresponding to the driverless vehicle and the robotic dog to the corresponding robot control terminals respectively; An inspection task execution module, configured to simultaneously control the driverless vehicle and the robotic dog by the robot control terminal to execute the corresponding inspection tasks, and optimize and adjust the target inspection routes corresponding to the driverless vehicle and the robotic dog according to the dynamically monitored obstacle information and electromagnetic interference information until the inspection tasks are completed.

Images