Operation and maintenance intelligent scheduling method, device and system for photovoltaic ground power station

Abstract

The invention discloses an operation and maintenance intelligent scheduling method, device and system for a photovoltaic ground power station, and belongs to the field of photovoltaic power station operation and maintenance, and the method comprises the steps: obtaining data, screening and integrating the data, analyzing the fault density based on a photovoltaic power station digital twin map, and selecting a to-be-scheduled task in a mode of selecting a range fault; constructing a scheduling comprehensive cost model; a scheduling comprehensive cost optimal model based on a PSO algorithm is constructed, a to-be-scheduled task and an integration factor are substituted into the model, and a scheduling comprehensive cost optimal scheme is finally found through model iteration optimization; and outputting a scheduling comprehensive cost optimal scheme. According to the method, operation and maintenance personnel can be helped to more clearly determine the sequence and the operation auxiliary content of operation, the operation and maintenance efficiency is improved, the operation and maintenance cost is reduced, and task allocation and route planning are reasonably performed.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic power station operation and maintenance, and more specifically, to an intelligent scheduling method, device and system for operation and maintenance of photovoltaic ground power stations. Background Art

[0002] With the rapid development of renewable energy, the construction scale of ground photovoltaic power stations is expanding. Most ground photovoltaic power stations are located in remote areas, occupy a large area, and have long distances between sub-arrays. At present, the operation and maintenance scheduling of photovoltaic power stations mostly adopts manual scheduling. That is, experienced operation and maintenance engineers obtain the operation and maintenance tasks that need to be performed through the main control alarm, and then observe the weather forecast, personnel schedule, operation and maintenance vehicle information, and spare parts required for operation and maintenance in the next few days, and manually set a series of operation and maintenance scheduling routes, and then issue work orders to operation and maintenance personnel for execution. Therefore, for this multi-objective and multi-constrained scenario, especially for ground power station sub-arrays and spare parts warehouses that are far apart, and there are many faults and inspection tasks, the results of manual scheduling are difficult to achieve the optimal solution, resulting in high comprehensive cost and low efficiency of operation and maintenance scheduling, and the scheduling results lack intelligence and poor results.

[0003] In view of the problems in the background, the inventor of this application believes that: the current ground photovoltaic power stations basically adopt the method of manual scheduling, the distribution of fault locations is not intuitive and clear, the comprehensive cost of operation and maintenance scheduling is high, the efficiency is low, and the scheduling results lack intelligence. In the wind power scenario, the Chinese patent application with publication number CN112132422A discloses a wind farm intelligent operation and maintenance scheduling system, in which statistics and machine learning are combined, tasks are manually adjusted according to the manual scheduling module, and the final scheduling is performed, and the tasks are dispatched to the operation and maintenance personnel through the dispatch module, but the economic factors of operation and maintenance are not considered in its solution; the Chinese patent application with publication number CN113642937A discloses an operation and maintenance scheduling method, device, electronic device and storage medium for a wind turbine cluster, in which the overall capacity loss cost and overall operation and maintenance cost of the wind turbine cluster are comprehensively considered, so that the global cost of the wind turbine cluster is minimized to achieve the purpose of cost saving, but environmental factors and the simultaneous working scenes of multiple operation and maintenance teams are not considered.

[0004] Therefore, for the ground photovoltaic power station scenario, there is still a lack of an optimal comprehensive cost scheduling method for multiple operation and maintenance teams that can clearly display an overview of faults and comprehensively consider weather factors, travel time, personnel location, equipment location, spare parts, personnel models, fault locations, etc., to help operation and maintenance personnel more clearly understand the sequence of operations and auxiliary work content, and solve problems such as inefficiency, poor economy, unreasonable task allocation, and unreasonable route planning in operation and maintenance. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide an intelligent scheduling method, device and system for operation and maintenance of photovoltaic ground power stations, so as to help operation and maintenance personnel to more clearly understand the sequence of operations and auxiliary content of operations, improve operation and maintenance efficiency, reduce operation and maintenance costs, and reasonably perform task allocation and route planning.

[0006] The object of the present invention is achieved through the following solutions: An intelligent operation and maintenance scheduling method for a photovoltaic ground power station includes the following steps: S1, obtains data and screens and integrates it, analyzes the fault density based on the digital twin map of the photovoltaic power station, and selects the tasks to be scheduled by selecting the range of faults; S2, build a comprehensive scheduling cost model; S3, build the optimal model of scheduling comprehensive cost based on PSO algorithm, bring the tasks to be scheduled and the integration factors into the model, and finally find the optimal solution of scheduling comprehensive cost through model iteration; S4, output the optimal scheduling comprehensive cost solution.

[0007] Further, in step S1, selecting the tasks to be scheduled by selecting a fault range specifically includes selecting the tasks to be scheduled by box-selecting a range of faults.

[0008] Furthermore, in step S1, the data acquisition also includes acquiring weather type, temperature, personnel cost, distance cost and fault information.

[0009] Further, in step S2, the scheduling comprehensive cost model specifically includes labor cost, in-transit cost and power loss cost; Human cost = number of personnel * personnel hourly cost * total scheduling time; number of personnel = number of personnel arranged by the operation and maintenance team; total scheduling time = time in transit + time to resolve faults; time to resolve faults = sum of time to resolve fault i; time in transit = power station entrance to fault 1 + fault 1 to fault 2 + ... + fault i-1 to fault i + fault to power station entrance, where for every day of working time exceeding the working time of a day, the time in transit will be increased by (fault x to power station entrance + power station entrance to fault x); Cost in transit = total distance * unit distance cost; Total distance = power station entrance to fault 1 + fault 1 to fault 2 + ... + fault i-1 to fault i + fault to power station entrance, where for every day of working time exceeding the working time of one day, there will be an additional distance (fault x to power station entrance + power station entrance to fault x) in the transit time; Power loss cost = sum (fault x power loss per unit time * (fault x resolution time / comprehensive impact factor K + fault x waiting time)); comprehensive impact factor K = operation and maintenance team efficiency * weather type efficiency coefficient * temperature efficiency coefficient; fault x waiting time = (the sum of the in-transit time before fault x begins to be resolved + the sum of the resolution time of other faults before fault x); Among them, "*" represents multiplication, "+" represents addition, and i represents an integer.

[0010] Further, in step S3, the optimal scheduling comprehensive cost model based on the PSO algorithm is constructed, the tasks to be scheduled and the integration factors are brought into the model, and the optimal scheduling comprehensive cost solution is finally found through model iteration optimization. Specifically for the ground-type photovoltaic power station scenario, the scheduling task execution order is combined with the particles in the PSO to achieve joint operation and maintenance of multiple operation and maintenance teams, and the scheduling comprehensive cost optimal model is used as the objective function of the PSO to construct the scheduling comprehensive cost optimal model based on the PSO algorithm; All the waiting-for-scheduling plans are screened in combination with the current spare parts situation, and the screened waiting-for-scheduling tasks and integration factors are brought into the scheduling comprehensive cost optimal model based on the PSO algorithm, and the following process is run in the model: 1) Initialize the particle swarm algorithm and parameter settings, including the random position and speed in the particle swarm algorithm; initialize a group of particles, the number of particles in the particle swarm = the number of faults * the number of groups, the group size is N, N is the number of scheduling schemes corresponding to a group of cycles; 2) Fault allocation: Allocate faults to corresponding operation and maintenance teams according to each scheduling scheme; suppose there are 3 operation and maintenance teams that need to schedule n faults and the faults of the three teams do not overlap; suppose the PSO particle swarm has 50 groups of particles, the number of elements in each group of particles is 3*n, n is an integer, and one group of particles is as follows: {A1, A2, A3, ..., An, B1, B2, B3, ..., Bn, C1, C2, C3, ..., Cn}; each element is [0,1], representing the possibility of each team handling the corresponding fault; Team A: {A1, A2, A3, ..., An}, representing the possibility of team A handling n faults; Team B: {B1, B2, B3, ..., Bn}, representing the possibility of team B handling n faults; Team C: {C1, C2, C3, ..., Cn}, representing the possibility of team C handling n faults; For the nth fault, take max{An,Bn,Cn} as the execution team of the nth fault, and assign all faults in this way. The execution order of the assigned teams is determined according to the sorting; output the scheduling plan corresponding to the group of particles according to the sorting method; 3) Calculate the comprehensive scheduling cost corresponding to each scheduling scheme; 4) For each scheduling plan, compare its scheduling comprehensive cost with its historical optimal scheduling comprehensive cost. If it is better, use it as the current optimal scheduling comprehensive cost; 5) Determine whether the iteration termination condition is met, if it is met, then the process ends; if it is not met, then update and adjust the particle speed and position, and go to step 2) until the iteration termination condition is met, then the process ends; the iteration termination condition includes reaching the maximum number of iterations Gk.

[0011] Furthermore, the scheduling task execution order is combined with the particles in PSO to realize the joint operation and maintenance of multiple operation and maintenance teams, and the scheduling comprehensive cost optimal model is used as the objective function of PSO to construct the scheduling comprehensive cost optimal model based on the PSO algorithm, which specifically includes the following sub-steps: (1) Consider all the scheduled faults as a set of particles of PSO; (2) The order of the particle values ​​is used as the order of fault scheduling and the scheduling scheme corresponding to the group of particles; (3) Calculate the scheduling comprehensive cost of the scheduling scheme corresponding to the group of particles as the objective function; (4) Use the PSO algorithm to optimize the comprehensive cost of scheduling and find the solution with the best cost.

[0012] Furthermore, in step 1), the position and speed determine how to update the size of the particle value, and the size of the particle value affects the execution order of the scheduled tasks.

[0013] An operation and maintenance intelligent scheduling device for photovoltaic ground power stations includes a processor and a memory. A program is stored in the memory. When the program is loaded by the processor, the operation and maintenance intelligent scheduling method for photovoltaic ground power stations as described in any one of the above items is executed.

[0014] An operation and maintenance intelligent scheduling system for photovoltaic ground power stations includes the operation and maintenance intelligent scheduling device for photovoltaic ground power stations as described above.

[0015] The beneficial effects of the present invention include: The present invention is particularly aimed at the ground-based photovoltaic power station scenario, and constructs a scheduling comprehensive cost optimal model based on PSO. It outputs a multi-operation and maintenance team scheduling plan with the optimal comprehensive cost for the defects and inspection items to be scheduled, helping operation and maintenance personnel to more clearly understand the sequence of operations and auxiliary work content, improve operation and maintenance efficiency, reduce operation and maintenance costs, and reasonably perform task allocation and route planning.

[0016] The PSO-based scheduling comprehensive cost optimal model proposed in the present invention supports a joint operation and maintenance solution for multiple operation and maintenance teams.

[0017] The present invention is based on the digital twin of the photovoltaic power station, analyzes and judges the fault density and severity of the specified area, and can support box-selecting range faults as tasks to be scheduled. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0019] Figure 1 is a flow chart of the overall steps of the method of the present invention; Figure 2 It is a schematic diagram of the scheduling comprehensive cost model of the present invention; Figure 3 The figure is a flowchart of the steps of the method of the preferred embodiment of the present invention. DETAILED DESCRIPTION

[0020] All features disclosed in all embodiments in this specification, or steps in all methods or processes implicitly disclosed, except for mutually exclusive features and / or steps, can be combined and / or expanded or replaced in any manner.

[0021] The specific implementation process of the present invention is as follows: For ground-mounted photovoltaic power station scenarios, the optimal scheduling requirements are essentially based on the current production status and driven by the core indicator system of the power station. Under limited resource conditions, they fully consider the influence of multiple factors, output the optimal scheduling strategy through intelligent dynamic algorithms, and import it into the work order system to effectively guide the operation and maintenance work.

[0022] Therefore, in the conception of the present invention, the focus is on considering various business scenarios of ground-mounted photovoltaic power stations, and utilizing multi-party data, including existing resources, various influencing factors, etc., to realize automated intelligent scheduling of operation and maintenance tasks, and rationally solve personnel arrangements, entry time, materials and tools, spare parts, route planning, and work and rest arrangements, so as to achieve the ultimate goal of shortest operation and maintenance time or lowest operation and maintenance cost.

[0023] In the specific implementation scheme, a scheduling comprehensive cost model is constructed, which can accurately analyze the comprehensive cost consumed by the implementation of the scheduling scheme, and especially for the ground-type photovoltaic power station scenario, the particle swarm optimization algorithm (PSO) is introduced to traverse and optimize the scheduling scheme. In the process of introducing the algorithm, the business scenario of the ground-type photovoltaic power station is considered, and the execution order of the scheduling task is combined with the particles in the PSO. The optimal scheduling comprehensive cost model is used as the objective function of the PSO to construct the optimal scheduling comprehensive cost model based on the PSO, and the scheduling scheme with the lowest comprehensive cost is finally obtained by using this model. Specifically, Figure 1 As shown, the following steps are included: S1, obtain data and filter and integrate it, including fault density analysis and corresponding data acquisition. In the fault density analysis, the fault density and severity of a certain area are analyzed and judged based on the digital twin map of the photovoltaic power station, and it is proposed to use the box selection range fault to select the tasks to be scheduled. In addition, in the data acquisition step, some basic operation and maintenance data will be obtained as an auxiliary reference, such as: weather type, temperature, personnel hourly cost, unit travel cost, fault information (equipment location corresponding to the fault, power loss per unit time of the fault, and fault resolution time), etc.

[0024] S2, construct a comprehensive scheduling cost model. In the photovoltaic power station intelligent scheduling solution proposed in the present invention, constructing a comprehensive scheduling cost model is crucial to optimizing operation and maintenance efficiency and reducing operating costs. The operation and maintenance cost model of the present invention further focuses on the comprehensive cost model of operation and maintenance team scheduling, and subdivides the comprehensive cost model of operation and maintenance team scheduling into three major components: labor cost, in-transit cost, and electricity loss cost, that is, the comprehensive cost model of operation and maintenance team scheduling = labor cost + in-transit cost + electricity loss cost.

[0025] More specifically, if Figure 2 As shown, the labor cost includes the labor cost of the entire scheduling process. Generally speaking, labor cost = number of personnel * personnel hourly cost * total scheduling time; number of personnel = number of people arranged by the operation and maintenance team; total scheduling time = in-transit time + fault resolution time; fault resolution time = sum of fault i resolution time; in-transit time = power station entrance to fault 1 + fault 1 to fault 2 + ... + fault i-1 to fault i + fault to power station entrance, where for every day of working time exceeding the working time of one day, there will be an extra time (fault x to power station entrance + power station entrance to fault x) in the in-transit time.

[0026] Cost in transit = total distance * unit distance cost; total distance = power station entrance to fault 1 + fault 1 to fault 2 + ... + fault i-1 to fault i + fault to power station entrance, where for every day's working time that exceeds the working time of one day, there will be an additional distance (fault x to power station entrance + power station entrance to fault x) in the transit time; power loss cost = sum (power loss per unit time of fault x * (fault x resolution time / comprehensive impact factor K + fault x waiting time)); wherein, the present invention defines the comprehensive impact factor K = operation and maintenance team efficiency (default 1) * weather type efficiency coefficient * temperature efficiency coefficient in combination with the application scenario; fault x waiting time = (sum of the transit time before fault x begins to be resolved + sum of other fault resolution times before fault x).

[0027] S3, construct the optimal scheduling comprehensive cost model based on the PSO algorithm. In order to achieve the shortest operation and maintenance time or the lowest operation and maintenance cost, in the conception of the present invention, the operation and maintenance team scheduling comprehensive cost model in step S2 is further integrated into the particle swarm algorithm, and at the same time, the scheduling task execution order is combined with the particles in the PSO for the ground-type photovoltaic power station scenario to realize the joint operation and maintenance of multiple operation and maintenance teams. The scheduling comprehensive cost optimal model is used as the objective function of the PSO, and the scheduling comprehensive cost optimal model based on the PSO is constructed, which ultimately achieves the optimal comprehensive cost and supports the joint operation and maintenance of multiple operation and maintenance teams.

[0028] All the plans to be scheduled are screened in combination with the current spare parts situation. According to the existing spare parts situation, the tasks to be scheduled are screened, and the tasks with ready spare parts are selected for scheduling (during the photovoltaic operation and maintenance process, there will be a variety of maintenance, overhaul or upgrade plans to be scheduled. These plans are designed to ensure the continuous and efficient operation of the photovoltaic system. However, the implementation of the plan must not only consider technical feasibility and time efficiency, but also be closely combined with the current spare parts situation.

[0029] Spare parts are the material basis of operation and maintenance work. Their inventory, types and status directly affect the selection and execution of the plan. Therefore, when screening the plans to be scheduled, it is necessary to comprehensively evaluate whether the types of existing spare parts match the plan requirements, whether the quantity is sufficient, and whether there are any expired or damaged parts that are not suitable for use.

[0030] Through this screening process, it can be ensured that the selected solution can effectively solve the problem and will not be hindered by spare parts problems. This not only improves the flexibility and response speed of operation and maintenance work, but also avoids the waste of resources caused by blind implementation of the plan. In short, screening the scheduling plan based on the spare parts situation is an important measure to achieve optimal resource allocation and efficient operation and maintenance in photovoltaic operation and maintenance.

[0031] The screened tasks to be scheduled and the integration factors (for example, power loss cost = sum (power loss per unit time of fault x * (fault x resolution time / comprehensive impact factor K + fault x waiting time)); wherein, the present invention defines the comprehensive impact factor K = operation and maintenance team efficiency (default 1) * weather type efficiency coefficient * temperature efficiency coefficient in combination with the application scenario; fault x waiting time = (the sum of the in-transit time before fault x begins to be resolved + the sum of other fault resolution times before fault x, wherein the power loss per unit time of fault x and the comprehensive impact factor K will affect the operation and maintenance efficiency of the team; the comprehensive impact factor K is determined by the operation and maintenance team efficiency (default 1), weather type efficiency coefficient, temperature efficiency coefficient, etc.) into the optimal model of scheduling comprehensive cost based on the PSO algorithm, and finally output the optimal scheduling comprehensive cost solution through model iteration and optimization. Figure 3 As shown, the optimization iteration process is as follows: 1) Initialize the particle swarm algorithm and parameter settings, including the random position and speed in the particle swarm algorithm; initialize a group of particles (the group size is N, N is the number of scheduling schemes corresponding to a group of cycles), the number of particles in the particle swarm = the number of faults * the number of groups; 2) Fault allocation: Allocate faults to corresponding operation and maintenance teams according to each scheduling scheme; specifically, assume that there are 3 operation and maintenance teams that need to schedule n faults and the faults of the three teams do not overlap. Then, in the optimal scheduling comprehensive cost model of the present invention, the PSO particle swarm is assumed to have 50 groups of particles, and the number of elements in each group of particles is 3*n (n is an integer), and one group of particles is as follows: {A1, A2, A3, ..., An, B1, B2, B3, ..., Bn, C1, C2, C3, ..., Cn}; wherein each element is [0,1], representing the possibility of each team handling the corresponding fault.

[0032] Team A: {A1, A2, A3, ..., An}, representing the possibility of team A handling n faults; Team B: {B1, B2, B3, ..., Bn}, representing the possibility of team B handling n faults; Team C: {C1, C2, C3, ..., Cn}, representing the possibility of team C handling n faults; For the nth fault, take max{An,Bn,Cn} as the execution team of the nth fault, and distribute all faults in this way; Assume that team A is assigned to {A2, A3, A6}, and its execution order is further determined by sorting. In the example, the values ​​are sorted from large to small, and those with larger values ​​are executed first, such as A2>A6>A3. Then the fault handling order of team A is {A2, A6, A3}.

[0033] According to the above method, the scheduling scheme corresponding to the group of particles is output, for example: {B1, A2, A3, C4, C5, A6, C7, B8, ..., Bn}, where the scheduling scheme corresponding to each shift is: Team A{A2, A6, A3...}; Team B {B1, B8, ..., Bn}; Team C{C4, C5, C7...}; 3) Calculate the comprehensive scheduling cost corresponding to each scheduling scheme; 4) For each group of plans, compare its scheduling comprehensive cost with its historical optimal scheduling comprehensive cost. If it is better, use it as the current optimal scheduling comprehensive cost; 5) Whether the iteration termination condition is met, if it is met, then the process ends; if not, then the particle velocity and position are updated and adjusted, and then go to step 2); the iteration termination condition is generally selected as the maximum number of iterations Gk according to the specific problem.

[0034] In a specific embodiment, for example, assume that there are three operation and maintenance teams that need to schedule n faults and the faults of the three teams do not overlap. Then, in the scheduling comprehensive cost optimization model of the present invention, it is assumed that the PSO particle swarm has a total of 50 groups of particles, and the number of elements in each group of particles is 3*n (n is an integer), and one group of particles is as follows: {A1, A2, A3, ..., An, B1, B2, B3, ..., Bn, C1, C2, C3, ..., Cn}; wherein each element is [0,1], representing the possibility of each team handling the corresponding fault.

[0035] Team A: {A1, A2, A3, ..., An}, representing the possibility of team A handling n faults; Team B: {B1, B2, B3, ..., Bn}, representing the possibility of team B handling n faults; Team C: {C1, C2, C3, ..., Cn}, representing the possibility of team C handling n faults; For the nth fault, take max{An,Bn,Cn} as the execution team of the nth fault, and distribute all faults in this way; Assume that team A is assigned to {A2, A3, A6}, and its execution order is further determined by sorting. In the example, the values ​​are sorted from large to small, and those with larger values ​​are executed first, such as A2>A6>A3. Then the fault handling order of team A is {A2, A6, A3}.

[0036] According to the above method, the scheduling scheme corresponding to the group of particles is output, for example: {B1, A2, A3, C4, C5, A6, C7, B8, ..., Bn}, where the scheduling scheme corresponding to each shift is: Team A{A2, A6, A3...}; Team B {B1, B8, ..., Bn}; Team C{C4, C5, C7...}; Then the objective function corresponding to this group of particles, that is, the scheduling comprehensive cost corresponding to this scheme = the scheduling comprehensive cost of team A + the scheduling comprehensive cost of team B + the scheduling comprehensive cost of team C. The scheduling comprehensive cost of each team is calculated separately according to step (3). Note: The number of teams depends on the actual situation.

[0037] S4, output scheduling plans for multiple operation and maintenance teams.

[0038] In summary, the present invention proposes a digital and intelligent scheduling method for power plant operation and maintenance based on the power plant digital twin, with the power plant core data indicator system as the improvement target, which has the following advantages: (1) Based on the digital twin of the photovoltaic power station, the fault density and severity of the specified area are analyzed and judged, and the fault range can be selected as a task to be scheduled.

[0039] (2) The weather, travel time, personnel location, equipment location, spare parts location, personnel model, and fault location are uniformly managed, and the fault loss factor and personnel skill factor are integrated (for example, power loss cost = sum (power loss per unit time of fault x * (fault x resolution time / comprehensive impact factor K + fault x waiting time)); wherein, the present invention defines the comprehensive impact factor K = operation and maintenance team efficiency (default 1) * weather type efficiency coefficient * temperature efficiency coefficient in combination with the application scenario; fault x waiting time = (the sum of the in-transit time before fault x begins to be resolved + the sum of other fault resolution times before fault x, wherein the power loss per unit time of fault x and the comprehensive impact factor K will affect the operation and maintenance efficiency of the team; the fault loss factor is the power loss per unit time of fault x; the comprehensive impact factor K is determined by the operation and maintenance team efficiency (default 1), weather type efficiency coefficient, temperature efficiency coefficient, etc.) to construct a scheduling comprehensive cost model.

[0040] (4) Combined with PSO, a PSO-based scheduling comprehensive cost optimal model is constructed, and the optimal scheduling plan with comprehensive cost is output based on the faults, defects, and inspection tasks to be scheduled.

[0041] (5) The optimal scheduling comprehensive cost model based on PSO can support the joint operation and maintenance plan of multiple operation and maintenance teams.

[0042] The units involved in the embodiments of the present invention may be implemented by software or hardware, and the units described may also be arranged in a processor. The names of these units do not, in some cases, limit the units themselves.

[0043] According to one aspect of an embodiment of the present invention, a computer program product or a computer program is provided, the computer program product or the computer program includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in the above various optional implementations.

[0044] As another aspect, an embodiment of the present invention further provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the above embodiment.

Claims

1. An intelligent operation and maintenance scheduling method for photovoltaic ground power stations, characterized in that: The following steps are involved: S1, obtains data and screens and integrates it, analyzes the fault density based on the digital twin map of the photovoltaic power station, and selects the tasks to be scheduled by selecting the range of faults; S2, build a comprehensive scheduling cost model; S3, build the optimal model of scheduling comprehensive cost based on PSO algorithm, bring the tasks to be scheduled and the integration factors into the model, and finally find the optimal solution of scheduling comprehensive cost through model iteration; S4, output the optimal scheduling comprehensive cost solution.

2. The intelligent operation and maintenance scheduling method for photovoltaic ground power stations according to claim 1 is characterized in that: In step S1, selecting the tasks to be scheduled by selecting a fault range specifically includes selecting the tasks to be scheduled by selecting a fault range by box selection.

3. The intelligent operation and maintenance scheduling method for photovoltaic ground power stations according to claim 1 is characterized in that: In step S1, the data acquisition also includes acquiring weather type, temperature, personnel cost, distance cost and fault information.

4. The intelligent operation and maintenance scheduling method for photovoltaic ground power stations according to claim 1 is characterized in that: In step S2, the scheduling comprehensive cost model specifically includes labor cost, in-transit cost and power loss cost; Human cost = number of personnel * personnel hourly cost * total scheduling time; number of personnel = number of personnel arranged by the operation and maintenance team; total scheduling time = time in transit + time to resolve faults; time to resolve faults = total time to resolve fault i; Time in transit = power station entrance to fault 1 + fault 1 to fault 2 + ... + fault i-1 to fault i + fault to power station entrance, where for every day of working time exceeding the working time of the other day, the time in transit will be increased by (fault x to power station entrance + power station entrance to fault x); Cost in transit = total distance * unit distance cost; Total distance = power station entrance to fault 1 + fault 1 to fault 2 + ... + fault i-1 to fault i + fault to power station entrance, where for every day of working time exceeding the working time of the day, there will be an additional distance (fault x to power station entrance + power station entrance to fault x) in the travel time; Power loss cost = sum (fault x power loss per unit time * (fault x resolution time / comprehensive impact factor K + fault x waiting time)); comprehensive impact factor K = operation and maintenance team efficiency * weather type efficiency coefficient * temperature efficiency coefficient; fault x waiting time = (the sum of the in-transit time before fault x begins to be resolved + the sum of the resolution time of other faults before fault x); Among them, "*" represents multiplication, "+" represents addition, and i represents an integer.

5. The intelligent operation and maintenance scheduling method for photovoltaic ground power stations according to claim 1 is characterized in that: In step S3, the optimal scheduling comprehensive cost model based on the PSO algorithm is constructed, the tasks to be scheduled and the integration factors are brought into the model, and the optimal scheduling comprehensive cost solution is finally found through model iteration optimization. Specifically for the ground-type photovoltaic power station scenario, the scheduling task execution order is combined with the particles in the PSO to achieve joint operation and maintenance of multiple operation and maintenance teams. The optimal scheduling comprehensive cost model is used as the objective function of the PSO to construct the optimal scheduling comprehensive cost model based on the PSO algorithm; All the waiting-for-scheduling plans are screened in combination with the current spare parts situation, and the screened waiting-for-scheduling tasks and integration factors are brought into the scheduling comprehensive cost optimal model based on the PSO algorithm, and the following process is run in the model: 1) Initialize the particle swarm algorithm and parameter settings, including the random position and speed in the particle swarm algorithm; initialize a group of particles, the number of particles in the particle swarm = the number of faults * the number of groups, the group size is N, and N is the number of scheduling schemes corresponding to a group of cycles; 2) Fault allocation: Allocate faults to corresponding operation and maintenance teams according to each scheduling scheme; suppose there are 3 operation and maintenance teams that need to schedule n faults and the faults of the three teams do not overlap; suppose the PSO particle swarm has 50 groups of particles, the number of elements in each group of particles is 3*n, n is an integer, and one group of particles is as follows: {A1, A2, A3, ..., An, B1, B2, B3, ..., Bn, C1, C2, C3, ..., Cn}; each element is [0,1], representing the possibility of each team handling the corresponding fault; Team A: {A1, A2, A3, ..., An}, representing the possibility of team A handling n faults; Team B: {B1, B2, B3, ..., Bn}, representing the possibility of team B handling n faults; Team C: {C1, C2, C3, ..., Cn}, representing the possibility of team C handling n faults; For the nth fault, take max{An,Bn,Cn} as the execution team of the nth fault, and assign all faults in this way. The execution order of the assigned teams is determined according to the sorting; output the scheduling plan corresponding to the group of particles according to the sorting method; 3) Calculate the comprehensive scheduling cost corresponding to each scheduling scheme; 4) For each scheduling plan, compare its scheduling comprehensive cost with its historical optimal scheduling comprehensive cost. If it is better, use it as the current optimal scheduling comprehensive cost; 5) Determine whether the iteration termination condition is met, if it is met, then end; if not, update and adjust the particle speed and position, and go to step 2) until the iteration termination condition is met, then end; the iteration termination condition includes reaching the maximum number of iterations Gk.

6. The intelligent operation and maintenance scheduling method for photovoltaic ground power stations according to claim 5 is characterized in that: The scheduling task execution order is combined with the particles in PSO to realize the joint operation and maintenance of multiple operation and maintenance teams. The scheduling comprehensive cost optimal model is used as the objective function of PSO to construct the scheduling comprehensive cost optimal model based on the PSO algorithm, which specifically includes the following sub-steps: (1) Consider all the scheduled faults as a set of particles of PSO; (2) The order of the particle values ​​is used as the order of fault scheduling and the scheduling scheme corresponding to the group of particles; (3) Calculate the scheduling comprehensive cost of the scheduling scheme corresponding to the group of particles as the objective function; (4) Use the PSO algorithm to optimize the comprehensive cost of scheduling and find the solution with the best cost.

7. The intelligent operation and maintenance scheduling method for photovoltaic ground power stations according to claim 5 is characterized in that: In step 1), the position and speed determine how to update the size of the particle value, and the size of the particle value affects the execution order of the scheduled tasks.

8. An intelligent operation and maintenance scheduling device for photovoltaic ground power stations, characterized in that: The method comprises a processor and a memory, wherein a program is stored in the memory, and when the program is loaded by the processor, the method for intelligent operation and maintenance scheduling for a photovoltaic ground power station as described in any one of claims 1 to 7 is executed.

9. An intelligent operation and maintenance scheduling system for photovoltaic ground power stations, characterized in that: Including the intelligent operation and maintenance scheduling device for photovoltaic ground power stations as described in claim 8.

Images