Photovoltaic power generation scheduling method based on AI intelligent optimization
By constructing AI photovoltaic digital model, evaluating the production capacity and power consumption trends of photovoltaic power stations, generating scheduling nodes and power consumption scheduling solutions, the problem of lack of flexibility and accuracy in scheduling decision-making in traditional photovoltaic power generation scheduling methods is solved, and the efficient use of photovoltaic power generation resources and grid stability are achieved.
Patent Information
- Application Number
- CN202510580478.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-07
- Publication Date
- 2025-06-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional photovoltaic power generation scheduling methods are difficult to accurately predict real-time changes in photovoltaic power generation, resulting in a lack of flexibility and accuracy in scheduling decisions, and often there is over or insufficient power generation, affecting the stability of the power grid and resource utilization efficiency.
Using the photovoltaic power generation scheduling method based on AI intelligent optimization, by constructing AI photovoltaic digital models, the operation data and meteorological data of each photovoltaic power station are obtained, the photovoltaic production capacity is evaluated, the power consumption trend is analyzed, the dispatch nodes and power consumption scheduling schemes are generated, and the dynamic matching between the photovoltaic power station and the power consumption needs is achieved.
It improves the efficiency of photovoltaic power generation resources, realizes the reasonable allocation of photovoltaic power station production capacity, avoids over- or insufficient power generation, and enhances the stability of the power grid.
Smart Images
Figure CN120109928A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of photovoltaic power generation, and in particular to a photovoltaic power generation scheduling method based on AI intelligent optimization. Background Art
[0002] As the global demand for clean energy continues to grow, photovoltaic power generation, as a sustainable energy solution, is gradually becoming an important part of power supply. However, photovoltaic power generation is intermittent and volatile, and its power generation is affected by many factors such as light intensity, temperature, and weather, which brings great challenges to the stable operation and dispatch of the power system.
[0003] Traditional photovoltaic power generation scheduling methods are mainly based on historical data and empirical models. It is difficult to accurately predict the real-time changes in photovoltaic power generation, resulting in a lack of flexibility and accuracy in scheduling decisions. In practical applications, there is often a surplus or shortage of power generation, which not only causes energy waste, but also may affect the stability of the power grid. How to perform adaptive power scheduling based on the production capacity and power consumption of photovoltaic power stations, thereby improving the efficiency of resource utilization, is a problem we need to solve. To this end, a photovoltaic power generation scheduling method based on AI intelligent optimization is now provided. Summary of the invention
[0004] The purpose of the present invention is to provide a photovoltaic power generation scheduling method based on AI intelligent optimization.
[0005] The purpose of the present invention can be achieved through the following technical solutions: A photovoltaic power generation scheduling method based on AI intelligent optimization, comprising: By building an AI photovoltaic digital model, obtaining the operating data of each photovoltaic power station, and importing the operating data of the photovoltaic power station into the AI photovoltaic digital model; Combined with the meteorological data of the areas where each photovoltaic power station is located, the photovoltaic capacity of each photovoltaic power station is evaluated to obtain the photovoltaic capacity coefficient of the photovoltaic power station; Obtain the power consumption data of the service end of each photovoltaic power station, analyze the power consumption trend of each service end, and determine whether the photovoltaic power station needs power scheduling based on the power consumption trend and photovoltaic capacity coefficient. If power scheduling is required, generate the corresponding scheduling node; The generated scheduling tasks are generated according to the generated scheduling nodes, and the generated scheduling tasks are integrated to obtain the final power scheduling plan.
[0006] Furthermore, the region where the photovoltaic power station is located and the service end of the photovoltaic power station are obtained; The photovoltaic power station group includes photovoltaic modules and a server. The operation data of the photovoltaic power station includes the light intensity at the location of the photovoltaic module and the current generated by the photovoltaic module.
[0007] Furthermore, a corresponding virtual photovoltaic point is constructed based on the photovoltaic power station, and then a corresponding virtual electricity output point is constructed based on the service end of the photovoltaic power station; Linking the virtual power output point with the virtual photovoltaic point, and generating data points corresponding to the virtual power output point and the virtual photovoltaic point respectively; Construct corresponding virtual meteorological points based on the area where the photovoltaic power station is located, and associate the virtual meteorological points with the virtual photovoltaic points; The virtual photovoltaic points, virtual electricity output points, data points and virtual meteorological points corresponding to the same photovoltaic power station are aggregated to obtain an AI regional sub-model corresponding to the photovoltaic power station; According to the dispatchability between photovoltaic power stations, the corresponding scheduling links are generated, and the AI regional sub-models are linked through the scheduling links; The AI regional sub-models of each PV power station are integrated to obtain the AI PV digital model.
[0008] Furthermore, the photovoltaic capacity of the photovoltaic power station is evaluated, and the process of obtaining the photovoltaic capacity coefficient of the photovoltaic power station is as follows: According to the obtained meteorological data, the corresponding time series is constructed, and the theoretical light intensity corresponding to the time series is generated, so as to obtain the theoretical total light amount; The obtained light intensity at the location of the photovoltaic module is recorded as the actual light intensity; Obtain the light intensity difference between the theoretical light intensity and the actual light intensity at the same moment in the time series, and then obtain the corresponding light intensity difference change curve; The photovoltaic power generation coefficient at the current moment is obtained according to the obtained light intensity difference variation curve.
[0009] Further, according to the input current of the server, the input current is integrated to obtain the power consumption of the server from the initial time t0 to the current time t, and a corresponding power consumption change curve is generated; Divide the initial time to the current time into n equal-length time segments, and obtain the power consumption trend coefficient of each time segment ; when > 0, then the time segment is recorded as the uplink segment. <0, then the time segment is recorded as a downlink segment. =0, then the time segment is marked as a balanced segment; If the time segment at the current moment is an uplink segment, obtain whether the previous time segment is an uplink segment. If it is an uplink segment, continue to obtain whether the previous time segment is an uplink segment, and so on, until the time segment is not an uplink segment; Get the time segment corresponding to the maximum value of the power consumption trend coefficient in all upstream segments, and use the time segment as the central axis to obtain the symmetrical time segment of the time segment at the current moment, and use the power consumption trend coefficient corresponding to the previous time segment of the symmetrical time segment as the predicted power consumption trend coefficient of the next time segment. According to the obtained predicted power consumption trend coefficient, judge the power consumption trend of the server.
[0010] Furthermore, based on the meteorological data, a predicted light intensity change curve from the current moment to the end of the evaluation period is generated, and any moment between the current moment and the end of the evaluation period is recorded as t 预 , then get t 预 Predicted photovoltaic capacity coefficient corresponding to the time ; According to the predicted power consumption trend coefficient and the power consumption change curve of the current time segment, the power demand of the next time segment is obtained. ; Then determine whether the photovoltaic power station can meet the power demand of the service end. If it can meet the power demand of the service end, obtain the power redundancy of the photovoltaic power station. If the photovoltaic power station cannot meet the power demand of the service end, then t 预 The moment is recorded as a scheduling node, and the PV power station is marked as the target power station, and a scheduling instruction is generated at the same time.
[0011] Furthermore, the process of generating a scheduling task according to the generated scheduling node and integrating the generated scheduling tasks to obtain a final power consumption scheduling solution includes: Get the photovoltaic power stations with redundant power, record them as pre-dispatching power stations, and mark the t corresponding to the redundant power of each pre-dispatching power station 预 time; Sort the pre-dispatched power stations in chronological order; Obtain the time interval between the dispatching nodes of the pre-dispatching power station and the target power station, as well as the shortest path from the pre-dispatching power station to the target power station, and obtain the dispatching links passed by the shortest path, and integrate the dispatching loss coefficients of each dispatching link to obtain the corresponding overall dispatching loss coefficient; Then the dispatching priority of each pre-dispatched power station is obtained; According to the obtained dispatching priorities, the pre-dispatching power stations are re-sorted from high to low. According to the sorting results, the power redundancy of each pre-dispatching power station is obtained, and the corresponding number of pre-dispatching power stations are selected.
[0012] Compared with the prior art, the present invention has the following beneficial effects: By analyzing the photovoltaic capacity of the photovoltaic power station, the photovoltaic capacity of the photovoltaic power station is obtained, and combined with the power consumption data of the service end corresponding to the photovoltaic power station, the power supply and demand of the photovoltaic power station is evaluated. When the photovoltaic capacity of the photovoltaic power station is insufficient, it is dispatched from other power stations with excess power, so as to achieve reasonable allocation of resources. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0014] Figure 1 is a flow chart of the method of the present invention; Figure 2 It is a schematic diagram of the AI photovoltaic digital model of the present invention; Figure 3 This is a flow chart of photovoltaic capacity assessment of the present invention. DETAILED DESCRIPTION
[0015] The solution described in the embodiments of the present invention is to solve the technical problem that the photovoltaic power generation capacity in different regions is inconsistent in the prior art, which leads to unreasonable resource allocation, low resource utilization efficiency, and thus waste of resources. The overall idea adopted is as follows: First, based on the operating data of each photovoltaic power station and the meteorological data of the area where the photovoltaic power station is located, the photovoltaic production capacity of the photovoltaic power station and the electricity consumption of the service end are evaluated to determine whether the photovoltaic production capacity can meet the electricity demand of the service end. According to the judgment result, it is determined whether the production capacity of the photovoltaic power station is excessive or insufficient. If it is excessive, it will be used as a pre-dispatch power station to dispatch those power stations with insufficient production capacity. If it is insufficient, it will be dispatched from other power stations that generate excess production capacity, so as to achieve reasonable allocation of resources.
[0016] like Figure 1 As shown, the photovoltaic power generation scheduling method based on AI intelligent optimization includes the following steps: By building an AI photovoltaic digital model, obtaining the operating data of each photovoltaic power station, and importing the operating data of the photovoltaic power station into the AI photovoltaic digital model; Combined with the meteorological data of the areas where each photovoltaic power station is located, the photovoltaic capacity of each photovoltaic power station is evaluated to obtain the photovoltaic capacity coefficient of the photovoltaic power station; Obtain the power consumption data of the service end of each photovoltaic power station, analyze the power consumption trend of each service end, and determine whether the photovoltaic power station needs power scheduling based on the power consumption trend and photovoltaic capacity coefficient. If power scheduling is required, generate the corresponding scheduling node; The generated scheduling tasks are generated according to the generated scheduling nodes, and the generated scheduling tasks are integrated to obtain the final power scheduling plan.
[0017] like Figure 2 As shown, in another embodiment of the present invention, the region where the photovoltaic power station is located and the service end of the photovoltaic power station are obtained; Build a corresponding virtual photovoltaic point based on the photovoltaic power station, and then build a corresponding virtual electricity output point based on the service end of the photovoltaic power station; Linking the virtual power output point with the virtual photovoltaic point, and generating data points corresponding to the virtual power output point and the virtual photovoltaic point respectively; Construct corresponding virtual meteorological points based on the area where the photovoltaic power station is located, and associate the virtual meteorological points with the virtual photovoltaic points; The virtual photovoltaic points, virtual electricity output points, data points and virtual meteorological points corresponding to the same photovoltaic power station are aggregated to obtain an AI regional sub-model corresponding to the photovoltaic power station; According to the dispatchability between photovoltaic power stations, the corresponding scheduling links are generated, and the AI regional sub-models are linked through the scheduling links; The AI regional sub-models of each PV power station are integrated to obtain the AI PV digital model.
[0018] In another embodiment of the present invention, the photovoltaic power station group includes a photovoltaic module and a service end, and the operation data of the photovoltaic power station includes the light intensity at the location of the photovoltaic module and the current generated by the photovoltaic module; It is necessary to set up corresponding data collection terminals in each component of each photovoltaic power station. The data collection terminals have different functions, including light collection terminals and current collection terminals set up in photovoltaic modules, and current collection terminals set up at the service end, which are used to obtain the operation data of the photovoltaic power station in each evaluation cycle; The light intensity at the location of the photovoltaic module is obtained in real time through a light collection terminal set in the photovoltaic module, and the current generated by the photovoltaic module is obtained through a current collection terminal set in the photovoltaic module; The input current of the server end is obtained through a current acquisition terminal arranged at the server end; Import the obtained operating data into the corresponding virtual node in the AI photovoltaic digital model, for example: The light intensity at the location of the photovoltaic module and the current generated by the photovoltaic module are introduced into the virtual photovoltaic point; the input current of the service end is introduced into the data point.
[0019] like Figure 3 As shown, in another embodiment of the present invention, the meteorological data of the location of the photovoltaic power station is imported into the virtual meteorological point, and the photovoltaic capacity of the photovoltaic power station is evaluated by combining the meteorological data imported into the virtual meteorological point and the operation data of the photovoltaic power station in the AI photovoltaic digital model to obtain the photovoltaic capacity coefficient of the photovoltaic power station. The specific process is: According to the obtained meteorological data, the corresponding time series is constructed, and the theoretical light intensity corresponding to the time series is generated, and then the theoretical total light intensity is obtained, which is recorded as ; The obtained light intensity at the location of the photovoltaic module is recorded as the actual light intensity, and the actual light intensity is mapped to the position corresponding to the time series according to the acquisition time; Obtain the light intensity difference between the theoretical light intensity and the actual light intensity at the same moment in the time series, and then obtain the corresponding light intensity difference change curve; The photovoltaic production capacity coefficient at the current moment is obtained according to the obtained light intensity difference change curve, for example: The current moment is recorded as t, the initial moment of the evaluation period is recorded as t0, and the real-time theoretical total amount of light intensity corresponding to the current moment is obtained according to the theoretical light intensity of the time period from the initial moment t0 to the current moment t, recorded as ; The duration of the evaluation cycle is set to T, and when the time interval from the initial time t0 to the current time t reaches T, the evaluation of the next evaluation cycle is carried out; The light intensity difference is recorded as Gc, and the light intensity difference from the initial time t0 to the current time t is integrated over time to obtain the light intensity difference, which is recorded as ; According to the current generated by the photovoltaic module, a corresponding current change curve is generated, and the current change curve is integrated to obtain the corresponding power value, which is recorded as ; Then the photovoltaic capacity coefficient of the photovoltaic power station at the current moment is obtained, which is recorded as ,in: .
[0020] In another embodiment of the present invention, based on the photovoltaic capacity coefficient of the photovoltaic power station at the current moment, the power consumption data of the server is analyzed, and the power consumption trend of the server is obtained according to the analysis result, so as to judge whether the photovoltaic power station needs to be dispatched in combination with the photovoltaic capacity coefficient of the photovoltaic power station at the current moment, specifically: Retrieve the input current of the server in the data point in the AI photovoltaic digital model, integrate the input current, and obtain the power consumption of the server from the initial time t0 to the current time t, which is recorded as , and generate the corresponding power consumption change curve; Divide the time from the initial moment to the current moment into n equal-length time segments, mark each time segment as i, and record the start time of each time segment as , the end time is recorded as ; Then the electricity consumption trend coefficient of the time segment labeled i is obtained, which is recorded as ,in: ; in, Indicates the power consumption corresponding to the end of the time segment labeled i, represents the power consumption corresponding to the start time of the time segment labeled i, and tn is the duration of each time segment; when >0, the time segment is recorded as an upward segment, indicating that the power consumption trend of the server is rising. <0, the time segment is recorded as a downlink segment, indicating that the power consumption trend of the server is in a downward state. =0, then the time segment is marked as a balanced segment; In the specific implementation process, the time segment where the current moment is located is the time segment labeled i=n, and the current moment is the end moment of the time segment; If the time segment at the current moment is an uplink segment, then obtain whether the time segment i=n-1 is an uplink segment. If it is an uplink segment, then continue to obtain whether the time segment i=n-2 is an uplink segment, and so on, until the time segment is not an uplink segment; if the time segment at the current moment is a downlink segment, the same method is used, which is not repeated here; Obtain the time segment corresponding to the maximum value of the power consumption trend coefficient in all the upstream segments, and use the time segment as the central axis to obtain the symmetrical time segment of the time segment at the current moment, and use the power consumption trend coefficient corresponding to the previous time segment of the symmetrical time segment as the predicted power consumption trend coefficient of the next time segment, and judge the power consumption trend of the server according to the obtained predicted power consumption trend coefficient, that is; Example: The uplink segments are set to k1, k2, k3, k4, k5, and k6 in sequence, where the maximum value of the power consumption trend coefficient is k5, and the symmetrical time segment corresponding to k6 is k4, then k3 is used as the predicted power consumption trend coefficient of the next time segment.
[0021] In another embodiment of the present invention, after obtaining the power consumption trend of the service end, the power supply and demand relationship of the photovoltaic power station is determined in combination with the photovoltaic capacity coefficient of the photovoltaic power station, so as to determine whether the photovoltaic power station needs to be dispatched for power consumption. If power dispatching is required, a corresponding dispatching node is generated. The specific process is as follows: According to the meteorological data, the predicted light intensity change curve from the current moment to the end of the evaluation period is generated, and any time between the current moment and the end of the evaluation period is recorded as t 预 , then t 预 The predicted photovoltaic capacity coefficient at the corresponding time is recorded as ,in: ; in, Indicates the current time to t 预 The theoretical total amount of light at the moment; According to the predicted power consumption trend coefficient and the power consumption change curve of the current time segment, the power demand of the next time segment is obtained, which is recorded as ; when , it means that the photovoltaic power station can meet the power demand of the service end, then further obtain the power redundancy of the photovoltaic power station, the power redundancy is ; when < When , it means that the photovoltaic power station cannot meet the power demand of the service end, then t 预 The moment is recorded as a scheduling node, and the photovoltaic power station is marked as a target power station, and a scheduling instruction is generated at the same time; in is the power loss coefficient.
[0022] In another embodiment of the present invention, the process of generating a scheduling task according to the generated scheduling node and integrating the generated scheduling tasks to obtain a final power scheduling solution includes: Get the photovoltaic power stations with redundant power, record them as pre-dispatching power stations, and mark the t corresponding to the redundant power of each pre-dispatching power station 预 time; Sort each pre-dispatched power station in chronological order, and label each pre-dispatched power station according to the sorting result, denoted as j, where j=1, 2, ..., m; Then the time interval between the pre-dispatched power station labeled j and the target power station dispatching node is recorded as ; Get the shortest path from the pre-scheduled power station labeled j to the target power station, and get the scheduling link that the shortest path passes through, and integrate the scheduling loss coefficients of each scheduling link to obtain the corresponding overall scheduling loss coefficient, which is recorded as ; Then the dispatch priority of each pre-dispatched power station is obtained, which is recorded as ,in: {Dy}_{j}=[{Ds}_{j}\times {Ry}_{j}-({Y}_{q}-\beta \times ({Yc}_{pre+}{Yx}_{i}))]{}^{T{}_{jg}} ; in, is the power redundancy of the pre-dispatched power station with label j; According to the obtained dispatching priorities, the pre-dispatching power stations are re-sorted from high to low, and according to the sorting results, the power redundancy of each pre-dispatching power station is obtained, a corresponding number of pre-dispatching power stations are selected, and the dispatching instructions are sent to the corresponding pre-dispatching power stations; For example, the pre-dispatch power station is selected according to the order of the pre-dispatch power stations, and the power redundancy of each pre-dispatch power station is added in turn until it exceeds .
[0023] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments with equivalent changes without departing from the scope of the technical solution of the present invention. However, any modification or equivalent replacement of the above embodiments made according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of the technical solution of the present invention.
Claims
1. Photovoltaic power generation scheduling method based on AI intelligent optimization, characterized in that: include: By building an AI photovoltaic digital model, obtaining the operating data of each photovoltaic power station, and importing the operating data of the photovoltaic power station into the AI photovoltaic digital model; Combined with the meteorological data of the areas where each photovoltaic power station is located, the photovoltaic capacity of each photovoltaic power station is evaluated to obtain the photovoltaic capacity coefficient of the photovoltaic power station; Obtain the power consumption data of the service end of each photovoltaic power station, analyze the power consumption trend of each service end, and determine whether the photovoltaic power station needs power scheduling based on the power consumption trend and photovoltaic capacity coefficient. If power scheduling is required, generate the corresponding scheduling node; Generate scheduling tasks according to the generated scheduling nodes, and integrate the generated scheduling tasks to obtain the final power scheduling plan; Get the area where the photovoltaic power station is located and the service end of the photovoltaic power station; The photovoltaic power station group includes photovoltaic modules and a server. The operation data of the photovoltaic power station includes the light intensity at the location of the photovoltaic module, the current generated by the photovoltaic module, and the input current of the server. Build a corresponding virtual photovoltaic point based on the photovoltaic power station, and then build a corresponding virtual electricity output point based on the service end of the photovoltaic power station; Linking the virtual power output point with the virtual photovoltaic point, and generating data points corresponding to the virtual power output point and the virtual photovoltaic point respectively; Construct corresponding virtual meteorological points based on the area where the photovoltaic power station is located, and associate the virtual meteorological points with the virtual photovoltaic points; The virtual photovoltaic points, virtual electricity output points, data points and virtual meteorological points corresponding to the same photovoltaic power station are aggregated to obtain an AI regional sub-model corresponding to the photovoltaic power station; According to the dispatchability between photovoltaic power stations, the corresponding scheduling links are generated, and the AI regional sub-models are linked through the scheduling links; The AI regional sub-models of each PV power station are integrated to obtain the AI PV digital model.
2. The photovoltaic power generation scheduling method based on AI intelligent optimization according to claim 1 is characterized in that: The process of evaluating the photovoltaic capacity of a photovoltaic power station and obtaining the photovoltaic capacity coefficient of the photovoltaic power station is as follows: According to the obtained meteorological data, the corresponding time series is constructed, and the theoretical light intensity corresponding to the time series is generated, so as to obtain the theoretical total light amount; The obtained light intensity at the location of the photovoltaic module is recorded as the actual light intensity; Obtain the light intensity difference between the theoretical light intensity and the actual light intensity at the same moment in the time series, and then obtain the corresponding light intensity difference change curve; The photovoltaic power generation coefficient at the current moment is obtained according to the obtained light intensity difference variation curve.
3. The photovoltaic power generation scheduling method based on AI intelligent optimization according to claim 2 is characterized in that: According to the input current of the server, the input current is integrated to obtain the power consumption of the server from the initial time t0 to the current time t, and a corresponding power consumption change curve is generated; Divide the initial time to the current time into n equal-length time segments, and obtain the power consumption trend coefficient of each time segment ; when > 0, then the time segment is recorded as the uplink segment. <0, then the time segment is recorded as a downlink segment. =0, then the time segment is marked as a balanced segment; If the time segment at the current moment is an uplink segment, obtain whether the previous time segment is an uplink segment. If it is an uplink segment, continue to obtain whether the previous time segment is an uplink segment, and so on, until the time segment is not an uplink segment; Get the time segment corresponding to the maximum value of the power consumption trend coefficient in all upstream segments, and use the time segment as the central axis to obtain the symmetrical time segment of the time segment at the current moment, and use the power consumption trend coefficient corresponding to the previous time segment of the symmetrical time segment as the predicted power consumption trend coefficient of the next time segment. According to the obtained predicted power consumption trend coefficient, judge the power consumption trend of the server.
4. The photovoltaic power generation scheduling method based on AI intelligent optimization according to claim 3 is characterized in that: According to the meteorological data, the predicted light intensity change curve from the current moment to the end of the evaluation period is generated, and any time between the current moment and the end of the evaluation period is recorded as t 预 , then get t 预 Predicted photovoltaic capacity coefficient corresponding to the time ; According to the predicted power consumption trend coefficient and the power consumption change curve of the current time segment, the power demand of the next time segment is obtained. ; Then determine whether the photovoltaic power station can meet the power demand of the service end. If it can meet the power demand of the service end, obtain the power redundancy of the photovoltaic power station. If the photovoltaic power station cannot meet the power demand of the service end, then t 预 The moment is recorded as a scheduling node, and the PV power station is marked as the target power station, and a scheduling instruction is generated at the same time.
5. The photovoltaic power generation scheduling method based on AI intelligent optimization according to claim 4 is characterized in that: The process of generating scheduling tasks according to the generated scheduling nodes and integrating the generated scheduling tasks to obtain the final power scheduling plan includes: Get the photovoltaic power stations with redundant power, record them as pre-dispatching power stations, and mark the t corresponding to the redundant power of each pre-dispatching power station 预 time; Sort the pre-dispatched power stations in chronological order; Obtain the time interval between the dispatching nodes of the pre-dispatching power station and the target power station, as well as the shortest path from the pre-dispatching power station to the target power station, and obtain the dispatching links that the shortest path passes through, and integrate the dispatching loss coefficients of each dispatching link to obtain the corresponding overall dispatching loss coefficient; Then the dispatching priority of each pre-dispatched power station is obtained; According to the obtained dispatching priorities, the pre-dispatching power stations are re-sorted from high to low. According to the sorting results, the power redundancy of each pre-dispatching power station is obtained, and the corresponding number of pre-dispatching power stations are selected.