A Two-Stage Planning Method for Substation Inspection Path Based on Improved Gold-Extraction Algorithm

Abstract

The present invention discloses a two-stage planning method for a substation inspection path based on an improved gold panning algorithm, which belongs to the technical field of substation operation and maintenance, including: constructing an improved gold panning algorithm based on the free exploration behavior of gold panners and a convergence factor based on a cosine form; constructing a multi-team travel model and a single-team travel model based on multiple patrol scenarios; optimizing the multi-team travel model and the single-team travel model respectively based on the improved gold panning algorithm to screen out a benchmark travel model; optimizing the travel mode of each route of the benchmark travel model to screen out the optimal inspection path of the substation. After the present invention introduces free exploration behavior and a convergence factor based on a cosine form, the early search efficiency of the gold panning algorithm is improved, and the global optimization capability is improved to a certain extent.

Description

Technical Field

[0001] The present invention belongs to the technical field of substation operation and maintenance, and in particular relates to a two-stage planning method for a substation inspection path based on an improved gold panning algorithm. Background Art

[0002] In recent years, in order to serve the high-quality development of the social economy, the number of new and expanded substations built by State Grid companies in various provinces and cities is increasing year by year. Some municipal power supply companies are facing a tense situation of insufficient substation operation and inspection personnel and heavy daily inspection tasks. In view of the high investment and maintenance costs of digital inspection equipment such as intelligent inspection robots and drones, and their limited adaptability to complex environments inside and outside the station, they cannot completely replace manual labor and achieve widespread promotion. Therefore, this embodiment proposes a two-stage planning method for substation inspection paths based on an improved gold panning algorithm to reduce the burden and increase the efficiency of inspection and maintenance work for grassroots front-line employees.

[0003] At present, domestic and foreign research on substation inspection path optimization is mainly focused on the intra-station level. By establishing a substation environmental model, an improved optimization algorithm is used to search for the inspection route with the shortest path length. There are few planning for inter-station inspection routes, and most of the research focuses on the optimization of a single inspection type. The impact of multiple inspection types on the optimal inspection path is not considered. At the same time, it does not support inspectors to customize differentiated inspection routes based on actual working conditions. In addition, the quality of the algorithm improvement effect is closely related to the robustness of the original algorithm. As a new heuristic optimization algorithm proposed in 2023, the gold panning algorithm has stronger robustness and higher convergence speed than today's mainstream path optimization algorithms. However, based on path optimization in specific engineering scenarios, how to adjust the search strategy of the gold panning algorithm to ensure global optimization capabilities is a difficult problem that needs to be solved urgently. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a two-stage planning method for substation inspection paths based on an improved gold panning algorithm to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above object, the present invention provides a two-stage planning method for substation inspection path based on an improved gold panning algorithm, comprising:

[0006] An improved gold mining algorithm is constructed based on the free exploration behavior of gold miners and the convergence factor based on the cosine form;

[0007] Construct a multi-team travel model and a single-team travel model based on multiple patrol scenarios;

[0008] Based on the improved gold panning algorithm, the multi-team travel model and the single-team travel model are optimized respectively to select a benchmark travel model;

[0009] Optimize the travel modes of each route of the reference travel model, and screen out the optimal inspection path of the substation.

[0010] Preferably, the expression of the behavior process of the gold miner's free exploration behavior is:

[0011]

[0012] In the formula, ω1, ω2, and ω3 are the position weights of gold miners g1, g2, and i respectively; are the Euclidean distances of gold miners g1, g2, and i respectively; g1, g2, and i represent three gold miners respectively.

[0013] Preferably, the expression of the convergence factor based on the cosine form is:

[0014]

[0015] Among them, μ is the attenuation factor, 0 < μ ≤ 1, and here μ = 0.95; ζ is the perturbation coefficient, taking 0.02; r3 is a random quantity between [0, 1]; t max is the maximum number of iterations of the algorithm; π is the pi; cos(·) is the cosine function.

[0016] Preferably, the expression of the multi-team travel model is:

[0017]

[0018] Among them, f1(·) is the time function of the multi-team travel model; n is the number of substations; m is the substation number; t m,m+1 t m,m+2 are the average times required for the vehicle to travel from substation m to substations m + 1 and m + 2 respectively; Δt m is the time compensation amount of substation m.

[0019] Preferably, the expression of the single-team travel model is:

[0020]

[0021] In the formula, f2(·) is the time function of the single-team travel model; t n,0 is the average time required for the vehicle to travel from substation n to the origin; n is the number of substations; is the inspection duration of substation m; l is the inspection type, where l = 1, 2, 3 correspond to comprehensive inspection, routine inspection, and light-off inspection respectively; d is the importance level of the substation, where d = 1, 2, 3, 4 represent class 1 station, class 2 station, class 3 station, and class 4 station respectively; m is the substation number.

[0022] Preferably, the expression for screening out the reference travel model is:

[0023]

[0024] min F base = [F1, F2];

[0025] where F1 and F2 are the passing time functions under the "multiple teams traveling together" model and the "single team traveling" model respectively; C is the number of engineering vehicles; n c is the total number of substations on the route traveled by vehicle c; c is the engineering vehicle number; m and k are both substation numbers; is a logical variable with a value of 0 or 1. For vehicle c, when the vehicle travels from substation m to substation k, takes 1, otherwise takes 0; Δt m is the time compensation amount of substation m; l is the inspection type; L is the total number of types, which is taken as 3 here; note that when m = 0, t m,k is the average time required for the vehicle to travel from substation m to substation k; is the inspection duration of substation m; min(·) is the minimum value function; F base is the reference travel model; l is the inspection type, where l = 1, 2, 3 correspond to comprehensive inspection, routine inspection, and lights-out inspection respectively.

[0026] Preferably, the expression for optimizing the travel mode of each route of the reference travel model is:

[0027]

[0028] where n c is the total number of substations on the route traveled by vehicle c; m is the substation number; is the inspection duration of substation m; l is the inspection type; d is the importance level of the substation, where d = 1, 2, 3, 4 represent type 1 station, type 2 station, type 3 station, and type 4 station respectively; Δt m is the time compensation amount of substation m; t 1,2 is the average time required for the vehicle to travel from substation 1 to substation 2 on the inspection path; t m,m+2 is the average time required for the vehicle to travel from substation m to substation m + 2 on this path; is the average time required for the vehicle to travel from substation n c - 1 to substation n c required on this path.

[0029] Preferably, the expression of the objective function for optimizing the travel mode of each route of the reference travel model is:

[0030]

[0031] Among them, F new (·) is the travel time function of the new route; κ is the route number in the baseline model; R base is the total number of routes in the baseline model; λ is the new route division plan number; is the total number of substations passed by vehicle c under plan λ; is the optimization function. When the baseline model is the "single-team travel" model, otherwise, C(·) is the combination function; is the total travel time of the remaining routes after removing the routes in plan λ from the baseline model; is the new baseline model; F best is the optimal inspection path model for substations; C is the number of engineering vehicles; c is the engineering vehicle number; m and k are both substation numbers; ι is the number of routes to be selected in the combination function.

[0032] Compared with the prior art, the present invention has the following advantages and technical effects:

[0033] 1) After the present invention introduces the free exploration behavior and the convergence factor based on the cosine form, the early search efficiency of the gold panning algorithm is improved, and the global optimization ability is enhanced to a certain extent.

[0034] 2) The planning of the present invention can take into account various inspection types, and integrate the respective advantages of the "multi-team travel" model and the "single-team travel" model to optimize the optimal inspection path with the least total time consumption and the most moderate total number of route travels within the cycle.

[0035] 3) The present invention supports the inspection personnel to customize the passing time limit, inspection types and the number of vehicles according to the actual situation, and reasonably plan a differentiated inspection plan.

[0036] 4) The planning method of the present invention has strong scalability and is not limited to the substation field. For the distribution and transmission specialties, by appropriately adjusting the inspection model and presetting relevant parameters, the present invention is still applicable. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings forming a part of this application are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0038] Figure 1 is a schematic diagram of the free exploration behavior of the embodiment of the present invention;

[0039] Figure 2 is a comparison chart of the convergence factors of the embodiment of the present invention;

[0040] Figure 3 is a schematic diagram of the "multi-team travel" model of the embodiment of the present invention;

[0041] Figure 4 Schematic diagram of the "single-team travel" model according to an embodiment of the present invention;

[0042] Figure 5 Two-stage planning flowchart according to an embodiment of the present invention;

[0043] Figure 6 Geographical distribution map of stations under the substation operation area according to an embodiment of the present invention;

[0044] Figure 7 Algorithm convergence curve diagram according to an embodiment of the present invention;

[0045] Figure 8 Algorithm convergence curve diagram under different convergence factors according to an embodiment of the present invention;

[0046] Figure 9 Geographical distribution map of stations under the centralized control station according to an embodiment of the present invention;

[0047] Figure 10 Algorithm convergence curve diagrams of different algorithms according to an embodiment of the present invention;

[0048] Figure 11 Optimal inspection model route map according to an embodiment of the present invention, where (a) is the inspection route 1 map, (b) is the inspection route 2 map, (c) is the inspection route 3 map, (d) is the inspection route 4 map, (e) is the inspection route 5 map, and (f) is the inspection route 6 map. Detailed implementation manners

[0049] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0050] It should be noted that the steps shown in the flowchart of the drawings may be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than here.

[0051] Embodiment 1

[0052] In this embodiment, a two-stage planning method for substation inspection paths based on an improved gold panning algorithm is provided, including:

[0053] Step 1: For the emerging gold panning algorithm, introduce the free exploration behavior of gold panners and a convergence factor based on the cosine form to improve the solution accuracy of the algorithm in the substation inspection model.

[0054] Step 2: Plan for multiple inspection types, construct two substation inspection models of "multiple teams traveling together" and "single team traveling", preset inspection parameters such as the longest passing time limit for the route, inspection type, and number of vehicles, and use the improved gold panning algorithm to optimize and screen out the benchmark travel model.

[0055] Step 3: Based on the benchmark travel model, combine the travel advantages of the two types of models of "multiple teams traveling together" and "single team traveling", perform secondary optimization on each route, and determine the optimal inspection path with the least total time consumption within the cycle and a moderate number of route trips.

[0056] The following is a detailed description of the preferred examples in conjunction with the accompanying drawings:

[0057] 1. The solution process of the gold panning algorithm is divided into three stages: migration, gold panning, and collaboration. Among them, the migration process of the gold panners can be described by equations (1) to (2):

[0058]

[0059] In the formula, is the migration vector; is the position of the best gold mine; is the current position of gold panner i; t is the number of iterations; is the new position of gold panner i; A1 and C1 are variable coefficients, and their values are shown in equations (3) and (4) respectively:

[0060]

[0061] C1 = 2·r2 (4)

[0062] In the formula, both A1 and C1 are variable coefficients; r1 and r2 are random quantities between [0,1]; l1 is the value of the convergence factor l e when e takes 1, and its calculation method is shown in equation (5):

[0063]

[0064] In the formula, t max is the maximum number of iterations; l e is the convergence factor; e is the serial number; t is the number of iterations.

[0065] The gold panning process of the gold panners is shown in equations (6) to (7):

[0066]

[0067] In the formula, is the gold panning vector; is the position of randomly selected gold panner g1; is the current position of gold panner i; is the new position of gold miner i; A2 is a variable coefficient, and its value is shown in Equation (8):

[0068] A2 = l2(2r1 - 1) (8)

[0069] In the formula, A2 is a variable coefficient; r1 is a random quantity between [0, 1]; l2 is the value of l e when e takes the value of 2.

[0070] To obtain more gold, gold miners usually cooperate, and the mathematical model of their cooperation process is shown in Equations (9) - (10):

[0071]

[0072] In the formula, is the cooperation vector; is the position of randomly selected gold miner g1; is the position of randomly selected gold miner g2; is the new position of gold miner i; r1 is a random quantity between [0, 1]; is the current position of gold miner i.

[0073] If gold miner i can obtain more gold at the new position, then update the current position of i.

[0074] 2. The free exploration behavior of gold miners exists in the cooperation process. During this period, gold miner i searches for gold in the direction specified by two randomly selected gold miners. To further find a better gold mine, the position update formula of the improved grey wolf optimization algorithm is improved, and the gold mining direction of gold miners is dynamically changed near the centroid of the triangle formed by three gold miners as vertices. The schematic diagram of its behavior is as Figure 1 shown, and the exploration process is shown in Equations (11) - (14):

[0075]

[0076] In the formula, ω1, ω2, and ω3 are the position weights of gold miners g1, g2, and i respectively; are the Euclidean distances of gold miners g1, g2, and i respectively; is the new position of gold miner i.

[0077] 3. The expression of the convergence factor based on the cosine form is shown in Equation (15), and the function graph is as Figure 2 shown. Its function is to replace the original convergence factor l1 to broaden the exploration space of gold miners and ensure the optimization accuracy of the algorithm.

[0078]

[0079] Wherein, l1′ is the convergence factor; μ is the attenuation factor, 0 < μ ≤ 1, and in this embodiment, μ is taken as 0.95; ζ is the perturbation coefficient, taken as 0.02; r3 is a random quantity between [0, 1]; t max is the maximum number of iterations; t is the number of iterations; cos(·) is the cosine function; π is the ratio of the circumference of a circle to its diameter.

[0080] 4. For the three inspection types of comprehensive inspection, routine inspection, and blackout inspection, two substation inspection models of "multiple teams traveling together" and "single team traveling" are constructed, where: "multiple teams traveling together" means that multiple inspection teams share vehicles for travel, and the schematic diagram of its model is as Figure 3 shown; "single team traveling" means that one inspection team travels by vehicle to complete the inspection work of the substation, and the schematic diagram is as Figure 4 shown.

[0081] Figure 3 Among them, the engineering vehicle cooperates with 2 inspection teams for inspection, and its process is: the vehicle first sends inspection team 1 to Station A, then sends inspection team 2 to Station B, and then the vehicle returns to Station A to pick up inspection team 1 and sends it to Station C, and then goes to Station B to carry inspection team 2 to Station D...;

[0082] The mathematical model of "multiple teams traveling together" is shown in Equation (16):

[0083]

[0084] Wherein, f1(·) is the time function of the "multiple teams traveling together" model; m is the substation number; n is the number of substations; t m,m+1 、t m,m+2 are the average time required for the vehicle to travel from substation m to substations m + 1 and m + 2 respectively; t n-1,n is the average time required for the vehicle to travel from substation n - 1 to substation n; t n,0 is the average time required for the vehicle to travel from substation n to the origin; Δt m is the time compensation amount of substation m, and its value is defined by Equation (17):

[0085]

[0086] Wherein, Δt m is the time compensation amount of substation m; t m,m′ is the time experienced by the vehicle from leaving substation m to returning to substation m again; is the inspection duration of substation m; l is the inspection type, where l = 1, 2, 3 correspond to comprehensive inspection, routine inspection, and blackout inspection respectively; d is the importance degree of the substation, where d = 1, 2, 3, 4 represent Class 1 station, Class 2 station, Class 3 station, and Class 4 station respectively; m is the substation number.

[0087] The mathematical model of "single-team travel" is shown in Equation (18):

[0088]

[0089] where: f2(·) is the time function of the "single-team travel" model; t n,0 is the average time for the vehicle to travel from substation n to the origin; m is the substation number; n is the number of substations; t m,m+1 is the average time required for the vehicle to travel from substation m to substation m + 1; is the inspection duration of substation m; l is the inspection type, where l = 1, 2, 3 correspond to comprehensive inspection, routine inspection, and blackout inspection respectively; d is the importance level of the substation, where d = 1, 2, 3, 4 represent Class 1 station, Class 2 station, Class 3 station, and Class 4 station respectively.

[0090] 5. The maximum travel time limit for the route corresponds to the driving time of the route, and its value is set by the inspection personnel according to the actual situation. By optimizing the "single-team travel" and "multi-team travel" models respectively, the travel model with the minimum total travel time selected is used as the benchmark travel model, and the optimization process is shown in Equations (19) to (21):

[0091]

[0092] min F base =[F1,F2] (21)

[0093] where F1 and F2 are the travel time functions under the "multi-team travel" model and the "single-team travel" model respectively; C is the number of engineering vehicles; n c is the total number of substations on the line traveled by vehicle c; c is the engineering vehicle number; m and k are both substation numbers; is a logical variable, whose value is 0 or 1. For vehicle c, when the vehicle travels from substation m to substation k, takes 1, otherwise takes 0; Δt m is the time compensation amount of substation m; l is the inspection type; L is the total number of types, which is 3 in this embodiment; note that when m = 0, t m,k is the average time required for the vehicle to travel from substation m to substation k; is the inspection duration of substation m; min(·) is the function to take the minimum value; F base is the benchmark travel model; l is the inspection type, where l = 1, 2, 3 correspond to comprehensive inspection, routine inspection, and blackout inspection respectively.

[0094] 6. Taking the total time taken by the model to patrol a certain route as the consideration, if the travel time of "multiple teams traveling together" is shorter than that of "single-team travel", then travel in the way of "multiple teams traveling together", otherwise choose the "single-team travel" way. The selection criteria of the model are as follows:

[0095]

[0096] In the formula, n c is the total number of substations on the line traveled by vehicle c; m is the substation number; is the inspection duration of substation m; l is the inspection type; d is the importance level of the substation, where d = 1, 2, 3, 4 represent class 1 station, class 2 station, class 3 station, and class 4 station respectively; Δt m is the time compensation amount of substation m; t 1,2 is the average time required for the vehicle to travel from substation 1 to substation 2 on the inspection path; t m,m+2 is the average time required for the vehicle to travel from substation m to substation m + 2 on this path; is the vehicle on this path from substation n c -1 to substation n c required average time.

[0097] 7. The secondary optimization is to optimize the travel mode of each route in the benchmark travel model. Based on the steps shown in formulas (23) to (25), the optimal inspection path of the substation is selected. The two-stage planning process of the substation inspection path is as Figure 5 shown.

[0098]

[0099] In the formula, F new (·) is the travel time function of the new route; κ is the route number in the benchmark model; R base is the total number of routes in the benchmark model; λ is the new route division plan number; is the total number of substations passed by vehicle c under plan λ; is the optimization function. When the benchmark model is a "single-team travel" model, otherwise, C(·) is the combination number function; is the total travel time of the remaining routes after removing the route in plan λ from the benchmark model; is the new benchmark model; F best is the optimal inspection path model of the substation; C is the number of engineering vehicles; c is the engineering vehicle number; m and k are both substation numbers; ι is the number of routes to be selected in the combination number function.

[0100] The technical effects of the present invention will be further illustrated by examples below:

[0101] Taking a substation area and a centralized control station in a certain city as examples, the present invention conducts simulation analysis respectively to verify the effectiveness and universality of the proposed method. In view of the fact that the examples in this embodiment only involve Class 3 stations and Class 4 stations, the inspection parameters of the substations are set as shown in Table 1. In addition, the maximum passing time limit T of the route is set lim = 240 min; General parameters of the gold panning algorithm: The total number of gold panners is 30; t max = 100.

[0102] Table 1

[0103]

[0104] (1) Example of stations under the substation area

[0105] Figure 6 The geographical distribution overview of the stations under this area is shown. According to the geographical coordinates of each station, the passing time between each station is calculated using a map navigation software, and a substation passing time matrix is generated.

[0106] 1) Effect of algorithm improvement

[0107] To reflect the accuracy advantage of the algorithm after introducing the convergence factor based on the cosine form and the free exploration behavior, the algorithms are now compared respectively, Figure 7 The convergence curve diagrams of the two algorithms are given, Figure 8 and the convergence curves of the algorithms based on two convergence factors are compared.

[0108] From Figure 7 it can be seen that: The improved gold panning algorithm saves 3.79 min in the total route time compared with the gold panning algorithm, which shows that the improvement strategy proposed in this embodiment can improve the optimization accuracy of the algorithm to a certain extent and enhance the global optimization ability of the algorithm.

[0109] Figure 8 Among them, the iteration curve of the convergence factor based on the cosine form has higher convergence accuracy, because the improved convergence factor improves the search efficiency of the algorithm in the early stage of iteration and the optimization accuracy in the later stage, making it easier to search for smaller values and maintain them for a longer time, and the global search and local search capabilities of the algorithm are balanced. It should be emphasized that since the optimization time is not the main concern of the engineering planning problem, it is not considered in this embodiment.

[0110] 2) Comparison of inspection plans for the area example

[0111] In this embodiment, the planning period is taken as 1 month, and a total of 72 inspections of the stations under its jurisdiction are required within this period. To highlight the advantages of the two-stage planning method in this embodiment, the optimal inspection path model of the substation is now compared with the "multiple teams traveling together" model and the "single team traveling" model, and the comparison results are shown in Table 2. Tables 3-5 give the optimization plans of each model.

[0112] Table 2

[0113]

[0114] Table 3

[0115]

[0116]

[0117] Table 4

[0118]

[0119]

[0120] Table 5

[0121]

[0122]

[0123] As can be seen from Tables 2 to 5: The optimal inspection path model for the substation is based on the "multi-team travel" model with a total time of 7565.57 minutes screened in the first-stage planning. Combining the travel advantages of the "single-team travel" model, the best inspection path with a total time of 7163.63 minutes is finally optimized. In addition, in terms of the total number of route trips, the optimal inspection model has fewer total trips than the benchmark model, reducing the frequency of out-of-station inspections by the inspection teams within the cycle and improving the work efficiency to a certain extent.

[0124] (2) Case study of the substations under the centralized control station

[0125] Figure 9 The centralized control station shown is responsible for the centralized monitoring and comprehensive management of the substations in the western region of a certain city, including 11 220 kV substations, 39 110 kV substations, and 2 35 kV substations, involving 16 substations of 3 types and 36 substations of 4 types. Each inspection team gathers at the centralized control station and is uniformly allocated by it.

[0126] 1) Comparison of optimization algorithms

[0127] In the field of path optimization, the sparrow search algorithm (SSA), greedy algorithm (GA), and GWO are the current mainstream algorithms. Compared with some classical algorithms, they have higher search efficiency and are not easily trapped in local optima. To demonstrate the competitiveness of the GRO algorithm, a horizontal comparison is now carried out, and the comparison results are as Figure 10 shown.

[0128] From Figure 10It can be seen that none of the algorithms got stuck in local optima during the iteration process, and all found the optimal solutions within the global scope. In terms of finding competitive solutions, the GRO algorithm is superior to other algorithms and has stronger global optimization ability. Its convergence curve tends to be stable during the 50th iteration process, showing a relatively fast convergence speed. Thus, it can be seen that the GRO algorithm is a better choice for solving the path optimization problem.

[0129] 2) Comparison of Patrol Inspection Plans for Centralized Control Station Examples

[0130] To further prove the universality of the planning method in this embodiment, a large-scale centralized control station system is now simulated and verified. In the first stage of the example planning, the "single-team travel" model with a smaller total travel time is selected as the benchmark model. Based on this, in the second stage of planning, a travel plan with 99 total trips is optimized, and its route map is as Figure 11 shown, and the optimization results of the quadratic programming are shown in Table 6.

[0131] Figure 11 In the figure: the thin line represents the patrol inspection route of the "single-team travel" model, and the thick line represents the patrol inspection route of the "multi-team travel" model. The advantage of adopting the "multi-team travel" model for some routes is that if the time taken for the vehicle to travel from the first arrival at a substation to its return to the same station is less than the corresponding inspection duration of the station, then compared with the benchmark model, this model can save twice the travel time from this station to the next station, significantly reducing the vehicle's travel time.

[0132] Table 6

[0133]

[0134] As can be seen from Table 6, the total cycle time of the optimal patrol inspection model is saved by 640.7 minutes, approximately 11 hours, compared with the benchmark model, and the total number of route trips is 7 less than that of the benchmark model, greatly improving the patrol inspection efficiency of the patrol inspection team and enhancing the management efficiency of the centralized control station.

[0135] The above is only a preferred specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present application should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A two-stage planning method for substation inspection paths based on an improved gold panning algorithm, characterized in that, The following steps are involved: An improved gold mining algorithm is constructed based on the free exploration behavior of gold miners and the convergence factor based on the cosine form; Construct a multi-team travel model and a single-team travel model based on multiple patrol scenarios; Based on the improved gold panning algorithm, the multi-team travel model and the single-team travel model are optimized respectively to select a benchmark travel model; Optimizing the travel mode of each route of the benchmark travel model to select the optimal inspection path of the substation; The expression of the multi-team peer model is: Among them, f1(·) is the time function of the multi-team parallel model; n is the number of substations; m is the substation number; t m,m+1 , t m,m+2 are respectively the average time required for the vehicle to travel from substation m to substations m+1 and m+2; Δt m is the time compensation amount of substation m; The expression of the single-team travel model is: In the formula, f2(·) is the time function of the single-team travel model; t n,0 is the average time for the vehicle to travel from substation n to the origin; n is the number of substations; is the inspection duration of substation m; l is the inspection type, where l = 1, 2, 3 correspond to comprehensive inspection, routine inspection, and blackout inspection respectively; d is the importance level of the substation, where d = 1, 2, 3, 4 represent class 1 station, class 2 station, class 3 station, and class 4 station respectively; m is the substation number.

2. The two-stage planning method for the substation inspection path based on the improved gold panning algorithm according to claim 1, wherein The expression of the behavior process of the gold digger's free exploration behavior is: where ω1, ω2, and ω3 are the position weights of gold miners g1, g2, and i, respectively; are the Euclidean distances of gold miners g1, g2, and i, respectively; g1, g2, and i represent three gold miners, respectively.

3. The two-stage planning method for the substation inspection path based on the improved gold panning algorithm according to claim 1, characterized in that, The expression of the convergence factor based on the cosine form is: Among them, μ is the attenuation factor, where 0 < μ ≤ 1, and 0.95 is taken here; ζ is the perturbation coefficient, and 0.02 is taken; r3 is a random quantity between [0, 1]; t max is the maximum number of iterations of the algorithm; π is the pi; cos(·) is the cosine function.

4. The two-stage planning method for the substation inspection path based on the improved gold panning algorithm according to claim 1, characterized in that The expression for filtering out the benchmark travel model is: minF base = [F1,F2]; where F1 and F2 are the travel time functions under the "multiple teams traveling together" model and the "single team traveling" model respectively; C is the number of engineering vehicles; n c is the total number of substations on the route traveled by vehicle c; c is the engineering vehicle number; m and k are both substation numbers; is a logical variable with a value of 0 or 1. For vehicle c, when the vehicle travels from substation m to substation k, takes 1, otherwise takes 0; Δt m is the time compensation amount of substation m; l is the inspection type; L is the total number of types, which is taken as 3 here; note that when m = 0, t m,k is the average time required for the vehicle to travel from substation m to substation k; is the inspection duration of substation m; min(·) is the minimum value function; F base is the reference travel model; l is the inspection type, where l = 1, 2, 3 correspond to comprehensive inspection, routine inspection, and lights-out inspection respectively.

5. The two-stage planning method for the substation inspection path based on the improved gold panning algorithm according to claim 1, characterized in that, The expression for optimizing the travel mode for each route of the benchmark travel model is: Among them, n c is the total number of substations on the line traveled by vehicle c; m is the substation number; is the inspection duration of substation m; l is the inspection type; d is the importance level of the substation, where d = 1, 2, 3, 4 represent Class 1 station, Class 2 station, Class 3 station, and Class 4 station respectively; Δt m is the time compensation amount of substation m; t 1,2 is the average time required for the vehicle to travel from substation 1 to substation 2 on the inspection path; t m,m+2 is the average time required for the vehicle to travel from substation m to substation m + 2 on this path; is the average time required for the vehicle to travel from substation n c -1 to substation n c on this path.

6. The two-stage planning method for the substation inspection path based on the improved gold panning algorithm according to claim 1, characterized in that The expression of the objective function for optimizing the travel mode for each route of the benchmark travel model is: where F new (·) is the travel time function of the new route; κ is the route number in the base model; R base is the total number of routes in the baseline model; λ is the new route division plan number; is the total number of substations passed by vehicle c under plan λ; is the optimization function. When the baseline model is the "single-team travel" model, otherwise, C(·) is the combination number function; is the total travel time of the remaining routes after removing the routes in plan λ from the baseline model; is the new baseline model; F best is the optimal inspection path model of the substation; C is the number of engineering vehicles; c is the engineering vehicle number; m and k are both substation numbers; ι is the number of routes to be selected in the combination number function.

Images