Power Intelligent Operation and Maintenance Path Planning Method

Abstract

The present invention discloses a method for power intelligent operation and maintenance path planning, which relates to the field of power operation and maintenance. At present, when implementing intelligent operation and maintenance, only the shortest path is aimed to be achieved. The present invention includes the following steps: obtaining work order data; processing work order information; judging whether the total workload exceeds the maintenance capacity of all workers during their working hours; if so, discarding a batch of faults considered impossible to complete; re-establishing a path planning model and adding time limit and capacity limit; solving the path planning model to obtain an optimal dispatching plan, outputting dispatching work orders, and obtaining the power intelligent operation and maintenance path. According to the present technical solution, considering the fault urgency degree and distance factors comprehensively, the faults with high urgency degree are preferentially solved; appropriate workloads are arranged for each maintenance worker; according to the working hours, location and distance of the staff, the dispatching of embarrassing-time-consuming or excessive-total maintenance work is avoided, so that the work tasks of each worker are scientific, reasonable and feasible.

Description

Technical Field

[0001] The present invention relates to the field of power operation and maintenance, and particularly to a method for planning intelligent power operation and maintenance paths. Background Art

[0002] There are a large number of modern smart grid devices with complex distributions. Existing path planning methods are only simple mathematical solving methods, aiming only to achieve the shortest path during intelligent operation and maintenance. However, in practical applications, there may be more restrictive conditions, such as the working hours of individual workers, the priorities and durations of operation and maintenance tasks at different locations, etc. These real conditions are difficult to be taken into account by traditional methods, and often result in path plans that cannot be realized. Summary of the Invention

[0003] The technical problem to be solved and the technical task proposed by the present invention are to improve and refine the existing technical solutions, and provide a method for planning intelligent power operation and maintenance paths, so as to achieve the purpose of taking into account the length of the repair journey, the utilization rate of human resources, and the priority of faults during fault repair. To this end, the present invention adopts the following technical solutions.

[0004] A method for planning intelligent power operation and maintenance paths includes the following steps:

[0005] 1) Obtain work order data, including the location coordinates of the fault point, the urgency level, the maintenance time, as well as the working hours, moving speed, the starting location, and the ending location of each maintenance worker.

[0006] 2) Process the work order information, establish a path planning model, and preliminarily estimate the total time required for the work order.

[0007] 3) Judge whether the total workload exceeds the maintenance capacity of all workers during their working hours, that is, the over-saturation phenomenon; if it exists, discard a batch of faults considered impossible to complete the repair through a task screening algorithm.

[0008] 4) Re-establish a path planning model based on the tasks retained after screening, and add time limits and capacity limits.

[0009] 5) Solve the path planning model through a solver to obtain an optimal work assignment plan, output the work assignment work order, and obtain the intelligent power operation and maintenance path.

[0010] As a preferred technical measure: in step 2), when preliminarily estimating the total time required for the work order, the priority order of the urgency level and the working time limit of the workers are not considered, only the total amount of the journey and the maintenance time are considered, and it is judged whether the current work assignment exceeds the processing capacity of all workers within the remaining working time by estimating the total amount of work for the current work assignment.

[0011] As a preferred technical means: in step 3), the task screening algorithm discards a batch of maintenance tasks that are considered impossible to complete within the current working cycle, including the tasks at the end of the work order list that exceed the remaining working hours of the workers and single long-term maintenance tasks, in the case of exceeding the remaining working time of the workers.

[0012] As a preferred technical means: the task screening algorithm includes the steps:

[0013] 301) Set the retention ratio variable C to 1, that is, 100%;

[0014] 302) Conduct the first round of urgency screening and retain the maintenance tasks ranked in the top C% in terms of urgency;

[0015] 303) Conduct re-screening and retain the maintenance tasks whose journey and maintenance time are less than the inspection time; the inspection time is the basis for judging whether the maintenance work can be completed. In the case where there has been no over-saturation of dispatching tasks previously, the inspection time is set to 80% of the working time w of the workers. If there is an over-saturation situation, the inspection time is reduced to 80% of w and then divided by the average number of tasks per worker;

[0016] 304) Judge whether the retained maintenance tasks after screening are less than the set maintenance tasks; if so, supplement a certain amount of sub-urgent maintenance tasks;

[0017] 305) Use a path planning model without capacity limit and time window constraint to estimate the total time T required for the current maintenance tasks to estimate the time required to complete all tasks;

[0018] 305) If the total time T is greater than the working time of all maintenance personnel , then take the new retention ratio variable C as ; where: M is the current number of workers, and T is the total time required for the previous maintenance tasks;

[0019] 306) Return to step 303) and conduct a new round of iteration with the new retention ratio variable C until the dispatching duration is less than or equal to the remaining working hours of all operation and maintenance personnel.

[0020] Under the condition that the task screening algorithm takes into account the priority of operation and maintenance tasks, path duration, and the number of operation and maintenance personnel, the expected working duration of each operation and maintenance personnel's work order is as close as possible to the current remaining working limit; ensure that the tasks assigned to each person currently do not exceed the off-duty time and can fill the remaining time as much as possible.

[0021] The inspection time is set to 80% of the working hours w of the workers, leaving 20% of the time as "travel time to the maintenance point", which is related to the area managed by each maintenance personnel on average. As a preferred technical means: in step 4), the time limit exists in the form of a time window. Each fault corresponds to a time window, and it is stipulated that each worker must complete the fault repair within the time window; the time window is set according to the urgency of the fault point, only the latest time is set, and the earliest time is not set. The latest time is determined by the urgency of the fault point. The higher the urgency, the earlier the specified latest repair time. In this way, the priority of this fault point is higher when planning the dispatching path; if there is a situation where the maintenance time of a certain node is close to or exceeds the cut-off time of the corresponding time window, the time window corresponding to the maintenance task of this node will be widened accordingly;

[0022] The capacity limit is used to limit the workload of a single dispatching of workers, so that the dispatching task can be completed within the working hours.

[0023] In step 5), the solver adopts a guided local search method, which is a local search algorithm based on penalty. It keeps the solution structure and neighborhood structure unchanged during the search process, while dynamically modifying the objective function, so that the current local extreme value no longer has local optimality. A penalty term is generated during the search process. When the given local search algorithm stabilizes at the local optimum, the guided local search method modifies the objective function according to the penalty term, and the modified objective function can enable the search to go beyond the local optimum; input the coordinates of the maintenance station and operation and maintenance nodes, and the numbers of operation and maintenance personnel into the solver; the solver outputs: the order of work orders for each operation and maintenance personnel and the planned path trajectory.

[0024] Beneficial effects: The fault repair of this technical solution takes into account the length of the journey, the utilization rate of human resources, and the fault priority. It comprehensively considers the urgency of the fault and the distance factor, and gives priority to solving the fault repair with a high degree of urgency; and it can arrange appropriate workloads for each repair worker, and assign appropriate workloads that can be completed during working hours; in addition, it can avoid dispatching embarrassing or excessive total repair work according to the working hours, location and distance of the staff, so that the work tasks of each worker are scientific, reasonable and feasible. The fault repair takes into account the length of the journey, the utilization rate of human resources, and the fault priority. Brief Description of the Drawings

[0025] Figure 1 is a flowchart of the present invention. Detailed Embodiments

[0026] The technical solution of the present invention will be further described in detail below with reference to the drawings in the specification.

[0027] Such as Figure 1As shown in the figure, 1. A method for power intelligent operation and maintenance path planning, characterized by including the following steps:

[0028] 1) Obtain work order data, including the location coordinates of the fault point, the urgency level, the maintenance time, as well as the working hours of the maintenance workers, the moving speed, the starting location, and the ending location of each maintenance worker.

[0029] 2) Process the work order information, establish an unrestricted path planning model, and initially estimate the total time used for the work order. Here, the priority order of the urgency level and the working hours of the workers are not considered, only the total amount of the journey and the maintenance time are considered. Since the influence of the restrictions of the urgency level and the working hours on the total working time is not significant, it is used to estimate the total workload of the current dispatch to determine whether it exceeds the processing capacity of all workers in the current dispatch.

[0030] 3) Judge whether the total workload exceeds the maintenance capacity of all workers during their working hours, that is, the over-saturation phenomenon; if it exists, discard a batch of faults considered impossible to complete the repair through the task screening algorithm.

[0031] If the total working time estimated in step 2) is greater than the sum of the working hours of all workers, or higher than a certain proportion of the sum, it is considered that the current work order cannot be completed during the workers' working hours, and a part needs to be discarded through the task screening algorithm.

[0032] The task screening algorithm discards a batch of faults considered impossible to complete the repair. Here, the repairs considered impossible to complete not only include a part of the tasks that exceed the working hours of the workers, but also include the situation where the repair time required for a single fault is too long and exceeds the working hours of the workers.

[0033] Although when the total workload does not exceed the maintenance capacity of all workers during their working hours, that is, when there is no over-saturation phenomenon, it is also necessary to screen tasks with too long time consumption. To avoid the situation where the remaining working time of a single worker is too short and cannot be completed by an individual. After screening tasks with too long time consumption, appropriately supplement the screened tasks.

[0034] The steps of the screening algorithm include:

[0035] Set the retention ratio variable C to 1 (100%);

[0036] Conduct a round of screening, and retain the maintenance tasks with the urgency level ranked in the top C%.

[0037] Conduct a second-round screening to retain maintenance tasks with travel distance and repair time less than the inspection time. The inspection time is the basis for judging whether the maintenance work can be completed. In the case where oversaturation has not been considered before, the inspection time is set to 80% of the worker's working time w. If there is an oversaturated situation, the inspection time is reduced to 80% of w divided by the average number of tasks assigned to each worker (since it is assumed that each worker has a task assignment at this time);

[0038] If too few maintenance tasks are retained due to screening, supplement a certain number of sub-urgent maintenance tasks;

[0039] Use the VRP algorithm without capacity limit and time window constraint to estimate the total time T required for the current maintenance tasks to estimate the time required to complete all tasks;

[0040] If the total time T is greater than the working time of all maintenance personnel (or a certain proportion of it), then take the new retention ratio variable C as ;

[0041] Perform a new round of iteration with the new retention ratio variable C until C meets the above conditions;

[0042] 4) According to the tasks retained after screening, re-establish the path planning model and add time limit and capacity limit;

[0043] The time limit exists in the form of a time window. Each fault corresponds to a time window, stipulating that each worker must complete the fault repair within the time window. The time window is set according to the urgency of the fault point, only setting the latest time without setting the earliest time. The latest time is determined by the urgency of the fault point. The higher the urgency, the earlier the stipulated latest repair time, so that the priority of this fault point is higher in the dispatching path planning. If there is a situation where the maintenance time of a certain node is close to or exceeds the cut-off time of the corresponding time window, then the time window corresponding to the maintenance task of this node is widened accordingly.

[0044] The capacity limit originally refers to the vehicle path planning problem where a vehicle has a limited capacity, that is, only a limited amount of goods can be loaded in one journey. In the dispatching problem, a similar constraint lies in the working hours of workers. The capacity limit aims to limit the workload of a worker for a single dispatch so that the dispatch task can be completed within the working hours.

[0045] 5) Solve the path planning model through the guided local search method to obtain the optimal dispatching plan;

[0046] Guided Local Search (GLS) is a kind of heuristic algorithm and a local search algorithm based on penalty, with the characteristics of good generality and compact structure. Its basic principle is to keep the solution structure and neighborhood structure unchanged during the search process, while dynamically modifying the objective function, so that the current local extreme value no longer has local optimality. It will generate penalty terms during the search process. When the given local search algorithm stabilizes at the local optimum, the Guided Local Search modifies the objective function according to the penalty terms in a specific way, and the modified objective function can enable the search to go beyond the local optimum.

[0047] 6) Output the dispatching plan to obtain the intelligent power operation and maintenance path.

[0048] When not all workers are dispatched in the dispatching plan, the nearest non-dispatched worker to the fault point is sent to complete the maintenance tasks that have been screened out but can still be completed under certain conditions.

[0049] The above Figure 1 The above-described intelligent power operation and maintenance path planning method is a specific embodiment of the present invention, which has already reflected the substantial characteristics and progress of the present invention. According to actual usage needs, under the inspiration of the present invention, equivalent modifications can be made to its shape, structure, etc., and all are within the protection scope of this solution.

Claims

1. Power intelligent operation and maintenance path planning method, characterized in that It includes the following steps: 1) Obtain work order data, including the location coordinates of the fault point, the urgency level, the maintenance time, as well as the working hours of the maintenance workers, the moving speed, the working location and the off-duty location of each maintenance worker; 2) Process the work order information, establish an unrestricted path planning model, and preliminarily estimate the total time used for the work order; 3) Judge whether the total workload exceeds the maintenance capacity of all workers during their working hours, that is, the over-saturation phenomenon; If it exists, discard a batch of faults considered impossible to complete the maintenance through the task screening algorithm; 4) According to the tasks retained after screening, re-establish the path planning model and add time limit and capacity limit; 5) Solve the path planning model through a solver to obtain the optimal work assignment plan; output the work assignment work order to obtain the intelligent power operation and maintenance path; In step 2), when preliminarily estimating the total time used for the work order, the priority order of the urgency level and the working time limit of the workers are not considered, only the total amount of the journey and the maintenance time are considered, and it is judged whether the current work assignment exceeds the processing capacity of all workers within the remaining working time by estimating the total workload of the current work assignment; In step 3, the task screening algorithm discards a batch of maintenance tasks considered impossible to complete within the current working cycle, including the tasks at the end of the work order list that exceed the remaining working hours of the workers and individual long-term maintenance tasks, in the case of exceeding the remaining working time of the workers; The task screening algorithm includes the steps: 301) Set the retention ratio variable C to 1, that is, 100%; 302) Conduct the first round of urgency screening, and retain the maintenance tasks ranked in the top C% in terms of urgency; 303) Conduct another screening, and retain the maintenance tasks with the journey and maintenance time less than the inspection time; the inspection time is the basis for judging whether the maintenance work can be completed. In the case where it was not previously considered that there is an over-saturation phenomenon in the work assignment tasks, the inspection time is set to 80% of the working time w of the workers. If there is an over-saturation situation, the inspection time is reduced to 80% of w and then divided by the average number of jobs of each worker; 304) Judge whether the maintenance tasks retained after screening are less than the set maintenance tasks; if so, supplement a certain amount of sub-urgent maintenance tasks; 305) Use a path planning model without capacity restrictions and time window constraints to estimate the total time required for the maintenance tasks at this time to estimate the time required to complete all tasks; 306) If the total time is greater than the working time of all maintenance personnel ( ), then take the new retention ratio variable C as ; where: M is the current number of workers, and T is the total time required for the previous maintenance task; 307) Return to step 303) to conduct a new round of iteration with the new retention ratio variable C until the work assignment duration is less than or equal to the remaining working hours of all operation and maintenance personnel.

2. The power intelligent operation and maintenance path planning method according to claim 1, wherein: In step 4), the time limit exists in the form of a time window. Each fault corresponds to a time window, and it is stipulated that each worker must complete the fault maintenance within the time window; the time window is set according to the urgency level of the fault point, only the latest time is set, and the earliest time is not set. The latest time is determined by the urgency level of the fault point. The higher the urgency level, the earlier the specified latest completion time is, so that the priority of this fault point is higher when planning the work assignment path; if there is a situation where the maintenance time of a certain node is close to or exceeds the cut-off time of the corresponding time window, the time window corresponding to the maintenance task of this node will be widened accordingly; The capacity limit is used to limit the workload of the workers for a single work assignment so that the work assignment tasks can be completed during the working hours.

3. The power intelligent operation and maintenance path planning method according to claim 1, characterized in that: In step 5), the solver adopts a guided local search method, which is a penalty-based local search algorithm. It keeps the solution structure and neighborhood structure unchanged during the search process while dynamically modifying the objective function, so that the current local extreme value no longer has local optimality. A penalty term is generated during the search process. When the given local search algorithm stabilizes at the local optimum, the guided local search method modifies the objective function according to the penalty term, and the modified objective function can enable the search to go beyond the local optimum; the coordinates of the maintenance station and operation and maintenance nodes and the numbers of operation and maintenance personnel are input into the solver; the solver outputs: the order of work orders for each operation and maintenance personnel and the planned path trajectory.

Images