Substation intelligent inspection robot monitoring method based on MABC algorithm

Abstract

The invention discloses an intelligent substation inspection robot monitoring method based on an MABC algorithm. The method comprises the steps that S1, a substation channel scene is virtualized; s2, subarea division and coverage of a transformer substation channel; s3, transformer substation channel area coverage; s11, carrying out obstacle feature extraction; s12, drawing a virtual map; s13, drawing a topological map; in the step S3, virtual painting is carried out on the divided sub-regions by using a four-color algorithm, and then modeling is carried out according to problems: the sub-regions in a virtual region are covered by different colors, each three different colors have a region boundary intersection point, and a surplus robot is placed at a cutting point position in a graph. The method comprises the following steps: firstly, realizing scene virtualization in a transformer substation by using a particle swarm algorithm, and reasonably dividing sub-regions; regional coverage of the sub-regions is realized by improving an artificial probability bee colony algorithm; and on the basis of a four-color algorithm, redundant robots are reasonably arranged, so that the emergency strain capacity of the system is enhanced, the elasticity and robustness of the system are improved, and all-directional monitoring of a substation scene is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of application of robots in the electric power industry, and in particular to a monitoring method of a substation intelligent inspection robot based on a MABC algorithm. Background Art

[0002] At present, the monitoring method of unmanned substations mostly adopts fixed-point monitoring by monitoring cameras, which has monitoring blind spots. In special circumstances, problems cannot be discovered in time, and equipment defects may threaten the safe operation of the power grid. Due to the poor global perception of the monitoring area of ​​the monitoring camera, the robot autonomous inspection intelligent auxiliary means are used for optimization. The existing robot inspection technology mostly adopts single robot fixed-point inspection and manually arranges the inspection route. It cannot adapt to the inspection needs after environmental changes. In addition, the inspection cycle of a single robot is relatively long during the inspection process, resulting in a monitoring blind spot in time during the inspection process. Therefore, the single robot manually specifies the inspection route, and the fixed-point inspection method ignores the factors of geographical environment changes and the blind spots in time during the inspection process, lacking flexibility and robustness; while the change in the number of multi-robot inspections avoids the monitoring blind spots in time, but the adaptive ability is still poor after environmental changes, and the flexibility and robustness cannot meet the work requirements.

[0003] Explanation of relevant terms:

[0004] M2M: Dialogue between robots to achieve information sharing.

[0005] Artificial Probabilistic Bee Colony Algorithm (MABC): M is a probability value, which indicates whether it is necessary to perform a neighborhood search on the adjacent area. The larger M is, the more dimensions of the neighborhood search, which is more conducive to local search and improving convergence speed; the smaller M is, the more conducive to global search. Summary of the invention

[0006] The technical problem to be solved by the present invention is to overcome the deficiencies of the above-mentioned prior art and provide a substation intelligent inspection robot monitoring method based on the MABC algorithm which is easy to operate and has good effects.

[0007] The present invention is implemented by the following technical solution: A substation intelligent inspection robot monitoring method based on MABC algorithm, comprising the following steps:

[0008] S1. Substation channel scene virtualization;

[0009] S2, substation channel sub-area division and coverage;

[0010] S3, Substation channel area coverage;

[0011] Step S1 includes the following sub-steps:

[0012] S11, obstacle feature extraction;

[0013] S12, virtual map drawing;

[0014] S13, topological map drawing;

[0015] In step S2, the steps of dividing the substation channel sub-areas are as follows:

[0016] S21. According to the shape of the target area, a point cloud of the virtual map is obtained using a particle swarm and Poisson disk sampling algorithm;

[0017] S22. Each point cloud in the figure represents a target area point, and each circle with a point as the center represents the maximum area that can be radiated by this point;

[0018] S23, a circle with the point as the center, each circle will have a cutting edge with other circles or the boundary of the covered area, and then connect each cutting point to preliminarily divide the shape of each sub-area; in step S3, the divided sub-areas are virtually painted using the four-color algorithm, and then modeled according to the problem: the sub-areas in the virtual area are covered with different colors, and every three different colors have an intersection point on the area boundary, and the surplus robots are placed at the cutting point positions in the figure.

[0019] The method further comprises: in step S11, a virtual coordinate system is established according to the substation channel to quantify the position information of the robot; the range of the obstacle area is calculated according to the quantified data; and the boundary information of the obstacle area is generalized.

[0020] Generalize the obstacle area boundary and use the following formula:

[0021]

[0022] Among them, S1 is the area of ​​the region formed by the obstacle and the area tangent point, S AOB is the area of ​​the robot position and the area intersection point, P is the area of ​​S1 and S AOB The ratio between them, area D is an obstacle, S(x i ,y i ,d1,d2) represents the robot position state, x i ,y i The unique information represents the position of the robot in the channel within the station. d1 and d2 represent the length of the tangent point from point S to the boundary of the area measured by the robot sensor. α1 and α2 represent the angles of the two tangent lines to the x-axis respectively.

[0023] In step S12, combined with fog computing, the block map model explored by the robot is integrated and summarized, and mapped to the mobile robot that will perform the task in the next step.

[0024] In step S13, a topological map is drawn based on the inherent obstacles and newly added obstacle information quantified from the geometric map.

[0025] In step S2, the problem is modeled. The shape and area of ​​region D are known. Assume that deployment n = D / D c The point clouds cover the area, and the divided areas are independent of each other. Let the position of the center of gravity i of the point cloud in the global coordinate system be p i (x i ,y i ), then the particle swarm algorithm is used to solve the point cloud E = {v i |i=1,2,...,n}, where v i The position is p i (x i ,y i ), so that each robot group covers an area of The target area can be covered to the maximum extent, that is, the coverage rate R ci =D i / D c The overall coverage is To solve the above problem using particle swarm algorithm, we need to:

[0026] (a) Position quantization

[0027] Construct a vector based on the position of the point cloud in the global coordinate system;

[0028] (b) Constraints

[0029] The location of the point cloud should not exceed the area to be covered;

[0030] (c) Constraint function

[0031] For the target area, a secant circle covering the entire target area is drawn with the point cloud as the circle point.

[0032] In step S2, the first nectar source placement algorithm for sub-area coverage adopts the traditional ABC algorithm, and the robot perception range threshold is set to r. The second and subsequent nectar source placement algorithms adopt the following formula:

[0033] x ij =x minj +r·(x maxj -x minj )

[0034] Among them, x i j represents the i-th nectar source x i The j-th dimension value of , i is taken from {1,2…,SN}, j is taken from {1,2.}, x min j ,x max jRepresent the minimum and maximum values ​​in the j-dimension respectively. Initializing the honey source is to randomly generate SN honey sources in a certain range in all dimensions of each honey source. Select a parameter M to change the exploration dimension information of the exploration peak. M is a probability value, which means whether it is necessary to search the neighboring area. The exploration peak makes a judgment on each dimension, that is, first generate a random number P ij , then P ij With M ij For comparison, if M ij >P ij , then the corresponding dimension will perform neighborhood search, otherwise it will not be performed. The judgment expression of M is as follows:

[0035]

[0036] Among them, α is the evaluation coefficient, F is the number of evaluations that tend to search the neighborhood, and Max is the number of global evaluations.

[0037] In step S3, the principle of placement follows the following method: the placement of the redundant robot group can be defined as a triplet<E,R,G> , the environment E of the redundant robots in the group represents the environmental conditions of the three-color area around each cut point, and whether each area can operate normally. Let P i is the probability that region i needs this robot to cooperate, i∈{1,2,3}, and each region is required to have only one redundant robot for monitoring. R represents a group of N redundant robots, and each robot R i Represents the input, behavior and collaboration of the robot; input mainly refers to the perception of external information, behavior mainly refers to the interaction between internal and external information, and collaboration mainly refers to the interaction between redundant robots and between redundant robots and robots that are performing tasks; G refers to the target area, indicating whether the robot takes action to perform tasks in the target area.

[0038] (a) Position quantization

[0039] According to the position of the cut point in the global coordinate system, the vector P(x j ,y j ){j|1,2,...,n};

[0040] (b) Constraints

[0041] The position of the surplus robot is placed at the cutting point, that is:

[0042] (c) Constraint function

[0043] Set Boolean variables N and M to indicate whether this node can be used to place a robot.

[0044]

[0045] Where P(x j ,y j ) represents the location of the cut point, A k Represents three regions around the cut point, Visite (A k ) indicates whether the three surrounding areas are covered, where k can only take three possible values. If it is True, the robot can be placed; if it is False, the robot cannot be placed.

[0046] The present invention has the following advantages: the intelligent inspection robot monitoring method for substations based on the MABC algorithm of the present invention first uses the particle swarm algorithm to realize the scene virtualization in the substation and reasonably divides the sub-areas; the regional coverage of the sub-areas is realized by improving the artificial probability bee colony algorithm; on the basis of the four-color algorithm, the redundant robots are reasonably placed to enhance the emergency response capability of the system, improve the elasticity and robustness of the system, and realize all-round monitoring of the substation scene. At the same time, the MABC algorithm is adopted to increase the local search and improve the convergence speed, thereby accelerating the process of the global search. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings are part of the present invention and are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but do not constitute an improper limitation of the present invention. Obviously, the accompanying drawings described below are only some embodiments. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0048] In the attached picture:

[0049] Figure 1 is a flow chart of the steps of the present invention;

[0050] Figure 2 It is a schematic diagram of obstacle extraction modeling coordinates of the present invention;

[0051] Figure 3 It is a schematic diagram of the steps of dividing the virtual map into sub-areas of the present invention;

[0052] Figure 4 It is a schematic diagram of the placement of the rich robot of the present invention;

[0053] Figure 5 is a switch room topology map of the present invention;

[0054] Figure 6 It is a point cloud distribution diagram of the switch room topology map of the present invention;

[0055] Figure 7 This is the path diagram of the movement trajectory of the switch room robot of the present invention.

[0056] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The following embodiments are used to illustrate the present invention but are not used to limit the scope of the present invention.

[0058] In the description of the present invention, it should be noted that the directions or positional relationships indicated by terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “vertical”, “inside” and “outside” are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0059] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0060] like Figures 1 to 4 A substation intelligent inspection robot monitoring method based on the MABC algorithm is shown, comprising the following steps:

[0061] S1. Virtualization of substation channel scenes: virtualize unknown sections in substations using a topological-geometric hybrid method, combine fog computing, use mobile robots within the geographical distribution range, add edge computing nodes, and use M2M to achieve obstacle feature extraction, virtual map drawing, and topological map drawing;

[0062] S2. Substation channel sub-area division and coverage; use particle swarm and Poisson disk sampling algorithms to obtain the point cloud of the virtual map, and achieve sub-interval coverage through artificial probabilistic bee colony algorithm (MABC); S3. Substation channel area coverage; use the four-color algorithm to place redundant robots to increase the robustness of the algorithm, and achieve substation channel interval coverage through robot information sharing.

[0063] like Figure 2The substation intelligent inspection robot monitoring method based on the MABC algorithm is shown in step S1, in which the substation area scene is virtualized, including the following sub-steps:

[0064] S11, obstacle feature extraction;

[0065] like Figure 2 As shown, it is known that area D is an obstacle, S(x i ,y i ,d1,d2) represents the robot position state, x i ,y i The unique information represents the position of the robot in the channel within the station. d1 and d2 represent the length of the tangent point from point S to the boundary of the area measured by the robot sensor. α1 and α2 represent the angles of the two tangent lines to the x-axis respectively.

[0066] To effectively extract the feature information of obstacles, you need to:

[0067] 1) Establish a virtual coordinate system based on the substation channel to quantify the robot's position information x i ,y i . 2) Use the sensor to obtain the distance d1, d2 from the tangent point to the robot, and calculate the angle α1, α2 based on the tangent line; the two tangent points and the robot position can be obtained from the robot's position status information. The area of ​​the triangle that can be obtained by combining the above information is:

[0068]

[0069] 3) Generalize the boundary information of the obstacle area. Based on the robot’s estimation of the tangent area, assume that the area of ​​the area formed by the obstacle and the tangent point is S1. Based on the bottom midline d, d1=d·p, the midline division position is recorded as O1, and AO1 and BO1 are connected to form the virtual area boundary.

[0070] S12, virtual map drawing;

[0071] After obtaining the specific characteristics of the obstacles, the block map model explored by the robot is integrated and summarized in combination with fog computing, and mapped to the mobile robot that will perform the task in the next step. Assuming that the number of robots performing the task is N = {1, 2, 3...n}, the robots will autonomously form a team P after entering the task interval, and the number of teams is Q, where 1≤P≤(N / Q), P in represents the obstacle information feature calculation input of the task robot after autonomous team formation, P p Represents the processing of task data, P outIt represents the final output of the task. Under the general framework of 3C, after processing the specific operation parameters such as input, processing, and output, multiple mobile robot groups can share resources and utilize the advantages of fog computing and edge computing to realize the integration and aggregation of the overall map geometric features.

[0072] S13, topological map drawing;

[0073] When the regional geometric features are obtained, it is necessary to use the topological map to draw the topological map based on the inherent obstacles and new obstacles quantified by the geometric map. After the commonality screening and feature extraction of the environmental information collected by the robot sensor, simple geometric information features (such as points, line segments, arcs or curves) are obtained. The geometric representation used with each topological node as the core can ensure the realization of local precise positioning. This representation method is simpler and convenient for obstacle pose information estimation and feature recognition. For solving positioning problems in large-scale and dynamic environments, it can ensure the accuracy of local positioning information and verify the consistency of global positioning information. In the scenario application of substations, obstacles are often substation equipment, and new obstacles are mostly movable devices in the station. From the perspective of the sensor, they are some long reflective surfaces. It is easy to obtain the node information of the topological map based on the geometric map method.

[0074] like Figure 3 In the substation intelligent inspection robot monitoring method based on the MABC algorithm shown in FIG, in step S2, the substation channel sub-area division steps are as follows:

[0075] S21. According to the shape of the target area, a point cloud of the virtual map is obtained using a particle swarm and Poisson disk sampling algorithm;

[0076] S22. Each point cloud in the figure represents a target area point, and each circle with a point as the center represents the maximum area that can be radiated by this point;

[0077] S23, the circle with the point as the center, each circle will have a cutting edge with other circles or the boundary of the coverage area, and then connect each cutting point to preliminarily divide the shape of each sub-interval.

[0078] Model the problem. Given the shape and area of ​​region D, assume that deployment n = D / D c The point clouds cover the area, and the divided areas are independent of each other. Let the position of the center of gravity i of the point cloud in the global coordinate system be p i (x i ,y i ), then the particle swarm algorithm is used to solve the point cloud E = {v i |i=1,2,...,n}, where v i The position is p i (xi ,y i ), so that each robot group covers an area of The target area can be covered to the maximum extent, that is, the coverage rate R ci =D i / D c The overall coverage is For the area boundary or the area with obstacles, the robot can be arranged for additional expansion. Since the particle swarm algorithm is usually used to find the minimum value, it is equivalent to finding E, so that the uncovered area is minimized, that is, the uncovered rate R u =1-R c Minimum.

[0079] To solve the above problem using particle swarm algorithm, we need to:

[0080] (a) Position quantization

[0081] The vector is formed according to the position of the point cloud in the global coordinate system: p i (x i ,y i ){i|1,2,...,n};

[0082] (b) Constraints

[0083] The position of the point cloud should not exceed the area D to be covered, that is: (x i ,y i )∈D, (c) Constraint function

[0084] Draw a secant circle covering the entire target area with the point cloud as the circle point. Assume that the target area is divided into m circles and mark the center of each circle as C. j ,{j|1,2,...,m}. When C j When the radius is large enough, it means that the circular area is covered by the point cloud. When all circles can be covered by the divided circles, it can be considered that the virtual map is covered, that is:

[0085] min(||c j -p i ||2)<r, The constraint function is:

[0086] g=1-m c / m,m≈R u .

[0087] like Figure 3 A substation intelligent inspection robot monitoring method based on the MABC algorithm is shown, in step S2, the substation sub-area coverage includes the sub-area coverage strategy and the MABC algorithm, wherein,

[0088] 1) Sub-area coverage strategy

[0089] Initialize the parameters of the ABC algorithm, which include the number of nectar sources SN, the number of times a nectar source is determined to be abandoned limit, and the number of iterations to terminate. In the standard ABC algorithm, the number of nectar sources SN is equal to the number of employed bees and the number of follower bees. The formula for generating a nectar source is as follows:

[0090] x ij =x minj +rand[0,1](x maxj -x minj )

[0091] where x i j represents the i-th nectar source x i The j-th dimension value of , i is taken from {1,2…,SN}, j is taken from {1,2.}, x min j ,x max j They represent the minimum and maximum values ​​in the j dimension respectively. Initializing the nectar source means randomly generating SN nectar sources in a certain range in all dimensions of each nectar source.

[0092] (a) Employed bees conduct a one-dimensional neighborhood search near a food source to generate a new food source. The search strategy is as follows:

[0093]

[0094] Where: v ij is a newly generated food source, k∈{1,2,…,SN} and j∈{1,2,…,D} are randomly selected, but k≠i, is a randomly generated number on [-1,1].

[0095] (b) The hired bee stage ends and the follower bee stage begins. In this stage, hired bees share nectar source information in the dance area. (The dance area is arranged at the location of the surplus robots). The follower bees analyze this information and use the roulette strategy to select nectar sources for tracking and mining, so as to ensure that the probability of mining nectar sources with higher fitness values ​​is greater.

[0096] The mining process of follower bees is the same as that of hired bees. They use the formula to find new nectar sources and keep the better adapted ones. The nectar source has a parameter T. When the nectar source update is retained, T is 0, otherwise, T increases by 1. Therefore, T can count the number of times a nectar source has not been updated.

[0097] (c) If a nectar source has not been updated after multiple exploitations, that is, the T value is too high and exceeds the predetermined threshold limit, then this nectar source needs to be abandoned and the exploration peak phase is initiated. This reflects the negative feedback and fluctuation properties of self-organization in ABC, and then new nectar sources need to be found according to the formula for finding nectar sources.

[0098] 2) The MABC algorithm aims to solve the problem of slow convergence of artificial bee colony. To speed up the convergence of the algorithm, the following two points should be achieved:

[0099] (1) Strategies to improve the location of nectar sources;

[0100] (2) Improve the exploration of peak dimensions.

[0101] (a) The first honey source placement algorithm for sub-area coverage adopts the traditional ABC algorithm, setting the robot's perception range threshold to r. The second and subsequent honey source placement algorithms use the following formula:

[0102] x ij =x minj +r·(x maxj -x minj )

[0103] Among them, x i j represents the i-th nectar source x i The j-th dimension value of , i is taken from {1,2…,SN}, j is taken from {1,2.}, x min j ,x max j They represent the minimum and maximum values ​​in the j dimension respectively. Initializing the nectar source means randomly generating SN nectar sources in a certain range in all dimensions of each nectar source.

[0104] In the above formula, the location information of the nectar source is not limited to the numbers between [0,1]. The location of the nectar source will appear randomly within the range of the perception threshold r. This not only reduces the number of nectar sources, but also maximizes the convergence speed of the algorithm.

[0105] (b) In the above statement (a), the leader bee and the follower bee are only in x i A one-dimensional neighborhood search is performed near j. This method easily leads to low search efficiency of the algorithm, and the selection of food source k is randomly generated in the range of [0,1]. To improve the search efficiency, a parameter M can be selected to change the exploration dimension information of the exploration peak. M is a probability value, which means whether it is necessary to perform a neighborhood search on the adjacent area. The exploration peak makes a judgment on each dimension, that is, a random number P is generated first. ij , then P ij With M ij For comparison, if M ij >P ij , then the corresponding dimension will perform neighborhood search, otherwise it will not. It can be seen that the larger M is, the more dimensions the leading bee and the following bee will perform neighborhood search, which is more conducive to local search and improving convergence speed; the smaller M is, the more conducive to global search. The judgment expression of M is as follows:

[0106]

[0107] Among them, α is the evaluation coefficient, F is the number of evaluations that tend to search the neighborhood, and Max is the number of global evaluations.

[0108] like Figure 4 In the embodiment of the monitoring method of a substation intelligent inspection robot based on the MABC algorithm, in step S3, the divided sub-areas are virtually colored using the four-color algorithm, and then modeled according to the problem: the sub-areas in the virtual area are covered with different colors, and every three different colors have an area boundary intersection. The surplus robots are placed at the cut point position in the figure, and the placement principle follows the following method:

[0109] The placement of the surplus robot group can be defined as a triplet<E,R,G> , where E represents the robot's location environment, R represents the redundant robot group relationship, and G represents the robot placement target.

[0110] The environment E of the redundant robots in the group represents the environmental conditions of the three-color area around each cut point, and whether each area can operate normally. Let P i is the probability that region i needs this robot to cooperate, i∈{1,2,3}, and each region is required to have only one redundant robot for monitoring, R represents a group of N redundant robots, and each robot R i Represents the input, behavior and collaboration of the robot; input mainly refers to the perception of external information, behavior mainly refers to the interaction between internal and external information, collaboration mainly refers to the interaction between redundant robots and between redundant robots and robots that are performing tasks; G refers to the target area, indicating whether the robot takes action to go to the target area to perform the task.

[0111] (a) Position quantization

[0112] According to the position of the cut point in the global coordinate system, the vector P(x j ,y j ){j|1,2,...,n};

[0113] (b) Constraints

[0114] The position of the surplus robot is placed at the cutting point, that is:

[0115] (c) Constraint function

[0116] Set Boolean variables N and M to indicate whether this node can be used to place a robot.

[0117] Where P(x j ,y j ) represents the location of the cut point, Ak Represents three regions around the cut point, Visite (A k ) indicates whether the three surrounding areas are covered, where k can only take three possible values. If it is True, the robot can be placed; if it is False, the robot cannot be placed.

[0118] Specific application examples:

[0119] like Figure 5 As shown in the figure, a 110 kV smart substation has a high level of digitization and intelligence, complete primary and secondary equipment functions, supports multiple signal transmission methods such as optical fiber communication and wireless communication, and completes topological mapping in a 10 kV switch room with the help of a group robot;

[0120] like Figure 6 As shown in the figure, in the topological map of the 10 kV switch room, the sensor radiation range of the robot is comprehensively considered, the particle swarm algorithm and Poisson disk algorithm are combined to divide the interval into sub-areas, and the point cloud distribution map of the topological map is drawn;

[0121] like Figure 7 As shown in the figure, the improved artificial bee colony algorithm is used to combine the distribution of nectar sources. Under the continuous trial and error of the exploration bees, sub-interval coverage is achieved. Finally, all the exploration bees share resources, and after integration and summary, a path map of the robot's action trajectory is obtained.

[0122] The robot's coverage of the interval is the standard for indicator evaluation. Using the algorithm given in this application, the 10 kV switch room is traversed and covered multiple times, as shown below:

[0123] Table 1 Statistics of traversal times and area coverage

[0124]

[0125] It can be seen from the above table that the topological area converges faster in the early stage, but becomes slower after a certain number of traversals. When traversed 20 times, the coverage rate reaches 99.60%.

[0126] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0127] In addition, those skilled in the art will understand that, although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features of different embodiments also means being within the scope of protection of the present invention and forming different embodiments. For example, in the above embodiments, those skilled in the art can use them in combination according to the known technical solutions and the technical problems to be solved by this application. The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any technician familiar with this patent can make some changes or modify the equivalent embodiments of equivalent changes by using the technical content suggested above without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still belongs to the scope of the solution of the present invention.

Claims

1. A substation intelligent inspection robot monitoring method based on MABC algorithm, characterized by: The steps include: S1. Substation channel scene virtualization; S2, substation channel sub-area division and coverage; S3, Substation channel area coverage; Step S1 includes the following sub-steps: S11, obstacle feature extraction; S12, virtual map drawing; S13, topological map drawing; In step S2, the steps of dividing the substation channel sub-areas are as follows: S21. According to the shape of the target area, a point cloud of the virtual map is obtained using a particle swarm and Poisson disk sampling algorithm; S22. Each point cloud in the figure represents a target area point, and each circle with a point as the center represents the maximum area that can be radiated by this point; S23, a circle with the point as the center, each circle will have a cutting edge with other circles or the boundary of the coverage area, and then connect each cutting point to preliminarily divide the shape of each sub-interval; In step S3, the divided sub-areas are virtually colored using the four-color algorithm, and then modeled according to the problem: the sub-areas in the virtual area are covered with different colors, and every three different colors have an area boundary intersection, and the surplus robots are placed at the cut point positions in the figure.

2. A substation intelligent inspection robot monitoring method based on the MABC algorithm as claimed in claim 1, characterized in that: In step S11, a virtual coordinate system is established according to the substation channel to quantify the position information of the robot; the range of the obstacle area is calculated according to the quantified data; and the boundary information of the obstacle area is generalized.

3. A substation intelligent inspection robot monitoring method based on the MABC algorithm as claimed in claim 2, characterized in that: Generalize the obstacle area boundary and use the following formula: Among them, S1 is the area of ​​the region formed by the obstacle and the area tangent point, S AOB is the area of ​​the robot position and the area intersection point, P is the area of ​​S1 and S AOB The ratio between them, area D is an obstacle, S(x i ,y i ,d1,d2) represents the robot position state, x i ,y i The unique information represents the position of the robot in the channel within the station. d1 and d2 represent the length of the tangent point from point S to the boundary of the area measured by the robot sensor. α1 and α2 represent the angles of the two tangent lines to the x-axis respectively.

4. The substation intelligent inspection robot monitoring method based on the MABC algorithm as claimed in claim 1 is characterized by: In step S12, combined with fog computing, the block map model explored by the robot is integrated and summarized, and mapped to the mobile robot that will perform the task in the next step.

5. The substation intelligent inspection robot monitoring method based on the MABC algorithm as claimed in claim 1 is characterized by: In step S13, a topological map is drawn based on the inherent obstacles and newly added obstacle information quantified from the geometric map.

6. The substation intelligent inspection robot monitoring method based on the MABC algorithm as claimed in claim 1 is characterized by: In step S2, the problem is modeled. The shape and area of ​​region D are known. Assume that deployment n = D / D c The point clouds cover the area, and the divided areas are independent of each other. Let the position of the center of gravity i of the point cloud in the global coordinate system be p i (x i ,y i ), then the particle swarm algorithm is used to solve the point cloud E = {v i |i=1,2,...,n}, where v i The position is p i (x i ,y i ), so that each robot group covers an area of The target area can be covered to the maximum extent, that is, the coverage rate R ci =D i / D c The overall coverage is 7. A substation intelligent inspection robot monitoring method based on the MABC algorithm as claimed in claim 6, characterized in that: To solve the above problem using particle swarm algorithm, we need to: (a) Position quantization Construct a vector based on the position of the point cloud in the global coordinate system; (b) Constraints The location of the point cloud should not exceed the area to be covered; (c) Constraint function For the target area, a secant circle covering the entire target area is drawn with the point cloud as the circle point.

8. The substation intelligent inspection robot monitoring method based on the MABC algorithm as claimed in claim 1 is characterized by: In step S2, the first nectar source placement algorithm for sub-area coverage adopts the traditional ABC algorithm, and the robot perception range threshold is set to r. The second and subsequent nectar source placement algorithms adopt the following formula: x ij =x minj +r·(x maxj -x minj ) Among them, x ij Represents the i-th nectar source x i The j-th dimension value of , i is taken from {1,2…,SN}, j is taken from {1,2.}, x minj ,x maxj They represent the minimum and maximum values ​​in the j dimension respectively. Initializing the nectar source means randomly generating SN nectar sources in a certain range in all dimensions of each nectar source.

9. A substation intelligent inspection robot monitoring method based on the MABC algorithm as claimed in claim 8, characterized in that: Select a parameter M to change the exploration dimension information of the exploration peak. M is a probability value, which means whether it is necessary to search the adjacent area. The exploration peak makes a judgment for each dimension, that is, first generate a random number P ij , then P ij With M ij For comparison, if M ij >P ij , then the corresponding dimension will perform neighborhood search, otherwise it will not be performed. The judgment expression of M is as follows: Among them, α is the evaluation coefficient, F is the number of evaluations that tend to search the neighborhood, and Max is the number of global evaluations.

10. The substation intelligent inspection robot monitoring method based on the MABC algorithm as claimed in claim 1, characterized in that: In step S3, the principle of placement follows the following method: The placement of the redundant robot group can be defined as a triplet<E,R,G> , the environment E of the redundant robots in the group represents the environmental conditions of the three-color area around each cut point, and whether each area can operate normally. Let P i is the probability that region i needs this robot to cooperate, i∈{1,2,3}, and each region is required to have only one redundant robot for monitoring. R represents a group of N redundant robots, and each robot R i Represents the input, behavior and collaboration of the robot; input mainly refers to the perception of external information, behavior mainly refers to the interaction between internal and external information, and collaboration mainly refers to the interaction between redundant robots and between redundant robots and robots that are performing tasks; G refers to the target area, indicating whether the robot takes action to perform tasks in the target area. (a) Position quantization According to the position of the cut point in the global coordinate system, the vector P(x j ,y j ){j|1,2,...,n}; (b) Constraints The position of the surplus robot is placed at the cutting point, that is: (c) Constraint function Set the Boolean variable N, which indicates whether the robot can be placed on this node. Where P(x j ,y j ) represents the location of the cut point, A k Represents three regions around the cut point, Visite (A k ) indicates whether the three surrounding areas are covered, where k can only take three possible values. If it is True, the robot can be placed; if it is False, the robot cannot be placed.

Images