New energy power dispatching optimization method and system based on smart grid, and medium

Abstract

This invention discloses a method, system, and medium for optimizing renewable energy power dispatch based on intelligent grids, relating to the field of power dispatching technology. The method includes: collecting multi-dimensional data of the area to be dispatched to obtain multi-dimensional information; traversing the multi-dimensional information based on predetermined partitioning factors to obtain a partitioning factor parameter set; analyzing the parameter set according to a gridding mechanism to generate gridded decisions and obtain a regional grid set; performing predictive analysis on the target grid to obtain predicted load data and generation data; simulating the optimal dispatching scheme using a dispatching fitness evaluation strategy as a benchmark and the predicted data as constraints; and dispatching renewable energy power according to the optimal scheme. This addresses the technical problems in renewable energy power dispatching that fail to fully consider the characteristics of the dispatching area and the correlation between multi-dimensional data, leading to inaccurate load and generation predictions and unbalanced resource allocation, thereby improving the efficiency of renewable energy consumption and the real-time performance of grid dispatching decisions.

Description

Technical Field

[0001] This invention relates to the field of power dispatching technology, and in particular to a method, system, and medium for optimizing the dispatching of new energy power based on smart grids. Background Technology

[0002] As the global energy structure transitions towards a low-carbon model, the proportion of new energy sources (such as wind power and photovoltaics) in the power system continues to increase. However, their inherent intermittent and fluctuating characteristics place higher demands on power dispatch. Traditional power dispatch methods rely heavily on experience or static regional divisions, making it difficult to accurately match the dynamic changes in new energy generation and load demand, easily leading to problems such as wind and solar curtailment or insufficient power supply. On the one hand, planning changes brought about by rapid urban development (such as the construction of new functional areas and adjustments to land use types) and differences in geographical environment (such as the impact of terrain and climate on power generation efficiency) have not been fully incorporated into dispatch considerations, making static grid divisions difficult to adapt to dynamic changes and limiting prediction accuracy. On the other hand, the correlation between multi-dimensional data (historical load, power generation records, urban planning, geographic information, etc.) has not been deeply explored, resulting in poor adaptability of load and power generation prediction models and difficulty in balancing supply and demand in dispatch schemes. Summary of the Invention

[0003] This invention provides a smart grid-based method, system, and medium for optimizing new energy power dispatch, which addresses the technical problems in new energy power dispatch that fail to fully consider the characteristics of the dispatch area and the correlation between multi-dimensional data, leading to inaccurate load and power generation forecasts and unbalanced resource allocation. It achieves the technical effect of dynamically and accurately matching regional new energy power load and power generation capacity through multi-dimensional data-driven grid partitioning and forecast optimization, thereby improving the efficiency of renewable energy consumption and the real-time performance of grid dispatch decisions.

[0004] In a first aspect, the present invention provides a new energy power dispatch optimization method based on intelligent grids, wherein the new energy power dispatch optimization method based on intelligent grids includes:

[0005] Multi-dimensional data is collected from the area to be dispatched to obtain multi-dimensional information; the multi-dimensional information is traversed based on predetermined partitioning factors to obtain a partitioning factor parameter set; the partitioning factor parameter set is analyzed according to a gridding mechanism to generate gridding decisions, and a regional grid set is obtained based on the gridding decisions; the target grid in the regional grid set is predicted and analyzed to obtain target prediction information, wherein the target prediction information includes target predicted load data and target predicted power generation data; the optimal dispatch scheme is simulated using the dispatch fitness evaluation strategy as the dispatch optimization benchmark and the target predicted load data and target predicted power generation data as dispatch optimization constraints; and new energy power is dispatched to the area to be dispatched according to the optimal dispatch scheme.

[0006] Secondly, the present invention also provides a new energy power dispatch optimization system based on intelligent grids, wherein the new energy power dispatch optimization system based on intelligent grids includes:

[0007] The system comprises the following modules: a data collection module (collecting multi-dimensional data of the area to be dispatched to obtain multi-dimensional information), a partitioning factor traversal module (traversing the multi-dimensional information based on predetermined partitioning factors to obtain a partitioning factor parameter set), a grid partitioning module (analyzing the partitioning factor parameter set according to a gridding mechanism to generate gridded decisions and obtain a regional grid set based on the gridded decisions), a prediction analysis module (performing prediction analysis on the target grid in the regional grid set to obtain target prediction information, including target predicted load data and target predicted power generation data), a dispatch optimization module (simulating the optimal dispatch scheme using the dispatch fitness evaluation strategy as the dispatch optimization benchmark and the target predicted load data and target predicted power generation data as dispatch optimization constraints), and a power dispatch module (dispatching new energy power to the area to be dispatched according to the optimal dispatch scheme).

[0008] Thirdly, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the smart grid-based new energy power dispatch optimization method provided by the present invention.

[0009] This invention discloses a new energy power dispatch optimization method, system, and medium based on intelligent grids, comprising: collecting multi-dimensional data of the area to be dispatched to obtain multi-dimensional information; traversing the multi-dimensional information based on predetermined partitioning factors to obtain a partitioning factor parameter set; analyzing the partitioning factor parameter set according to a gridding mechanism to generate gridded decisions, and obtaining a regional grid set based on the gridded decisions; performing predictive analysis on the target grids in the regional grid set to obtain target prediction information, wherein the target prediction information includes target predicted load data and target predicted power generation data; using a dispatch fitness evaluation strategy as the dispatch optimization benchmark, and using the target predicted load data as the basis for optimization. Using the predicted power generation data of the target as scheduling optimization constraints, the optimal scheduling scheme is simulated and obtained. Based on the optimal scheduling scheme, new energy power is dispatched to the area to be dispatched. The new energy power dispatch optimization method, system and medium based on smart grid disclosed in this invention solves the technical problem that the characteristics of the dispatch area and the correlation between multi-dimensional data are not fully considered in the new energy power dispatch, resulting in inaccurate load and power generation forecasts and unbalanced resource allocation. It achieves the technical effect of realizing dynamic and accurate matching of regional new energy power load and power generation capacity through grid division and prediction optimization driven by multi-dimensional data, and improving the efficiency of renewable energy consumption and the real-time performance of grid dispatch decision-making. Attached Figure Description

[0010] Figure 1This is a flowchart illustrating the new energy power dispatch optimization method based on smart grids according to the present invention.

[0011] Figure 2 This is a schematic diagram of the structure of the smart grid-based new energy power dispatch optimization system of the present invention.

[0012] Figure labeling: Data collection module 11, factor traversal module 12, grid partitioning module 13, predictive analysis module 14, scheduling optimization module 15, power dispatching module 16. Detailed Implementation

[0013] The above technical solutions will now be described in detail with reference to the accompanying drawings and specific embodiments to provide a better understanding of them. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments used only to explain the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. Furthermore, it should be noted that, for ease of description, only the parts related to the present invention are shown in the drawings, not all of them.

[0014] Example 1, as Figure 1 This is a flowchart illustrating the intelligent grid-based new energy power dispatch optimization method of the present invention, wherein the intelligent grid-based new energy power dispatch optimization method includes:

[0015] Multi-dimensional data collection is performed on the area to be scheduled to obtain multi-dimensional information.

[0016] Specifically, in the process of new energy power dispatch, in order to achieve accurate and efficient dispatch optimization, multi-dimensional information of the area to be dispatched is collected. This information covers all factors that may affect the stability, efficiency and sustainability of the power system during the power dispatch process, including but not limited to load data, new energy power generation data, geographic information and urban planning data. By summarizing these collected data, a multi-dimensional information set is formed. This multi-dimensional information not only reflects the historical operating status of the area to be dispatched, but also provides rich basic data support for subsequent dispatch optimization, helping to gain a deeper understanding of the various operating characteristics of the power system.

[0017] In some embodiments, including:

[0018] Obtain historical load records for the area to be dispatched; obtain historical renewable energy power generation records for the area to be dispatched; obtain real-time urban status for the area to be dispatched, wherein the real-time urban status includes real-time urban planning and real-time geographic information; construct the multi-dimensional information based on the historical load records, the historical renewable energy power generation records, the real-time urban planning, and the real-time geographic information.

[0019] Specifically, historical load records for the area to be dispatched are obtained through power monitoring systems or load records provided by power companies. These historical load records help understand the changing trends of electricity demand in the area, including peak hours, off-peak hours, and periodic fluctuations. Historical renewable energy generation records for the area to be dispatched are obtained through renewable energy generation monitoring logs. These historical renewable energy generation records reflect the generation of renewable energy (such as solar and wind power) in the area, helping to determine the patterns and volatility of renewable energy generation. Real-time urban status data for the area to be dispatched is obtained through reserved real-time data interfaces and Geographic Information Systems (GIS). This real-time urban status data includes real-time urban planning and real-time geographic information, reflecting the current status of the area and external factors that may affect changes in electricity load. Real-time urban planning includes, but is not limited to, urban development direction, construction projects, and the construction status of industrial parks and commercial areas, which directly affect the region's electricity demand. Real-time geographic information includes, but is not limited to, geographical location, topography, and climate, which affect the effectiveness of renewable energy generation. Finally, the obtained historical load records, historical renewable energy power generation records, real-time urban planning, and real-time geographic information are integrated. For example, according to the geographical location of the region, the load and power generation data of the same region are associated with the urban planning and geographic information of that region. Then, the associated data is processed by time standardization, unit unification, and formatting to ensure that all data items meet the unified analysis standards, thereby obtaining a multi-dimensional information set to support subsequent precise renewable energy power dispatch optimization.

[0020] The multidimensional information is traversed based on predetermined dividing factors to obtain a set of dividing factor parameters.

[0021] Specifically, after obtaining multidimensional information, it is traversed. At each traversal, the traversed data is compared with pre-defined keys in the predetermined partitioning factors. These predetermined partitioning factors include several important variables affecting regional power dispatch, such as elevation, slope, temperature, wind speed, solar radiation intensity, average load per unit area for different land use types, wind energy density, photovoltaic available hours, and average solar radiation intensity. By comparing these preset keys with the corresponding keys in the traversed data, the partitioning factor parameters corresponding to all predetermined partitioning factors are extracted, forming a partitioning factor parameter set. This parameter set will be further used in subsequent gridded decision-making and power dispatch optimization, providing accurate reference data for the power demand and generation capacity of each region. This ensures that the dispatching strategy accurately reflects the resource and demand characteristics of different regions, thereby improving the overall optimization level of the power grid.

[0022] The set of partitioning factor parameters is analyzed according to the gridding mechanism to generate gridded decisions, and a regional grid set is obtained based on the gridded decisions.

[0023] Specifically, after obtaining the set of partitioning factor parameters, the area to be dispatched is divided into multiple blocks according to the range partitioning strategy in the gridding mechanism. For each block, it is further subdivided according to the grid splitting strategy in the gridding mechanism to generate gridding decisions. Then, based on the determined gridding decisions, each block of the area to be dispatched is divided to form a regional grid set. This regional grid set provides accurate data support and decision-making basis for subsequent power dispatching, enabling power dispatching to make optimal adjustments according to the needs and resource distribution of different grids, thereby achieving more efficient power resource allocation and more stable power supply.

[0024] In some embodiments, the set of partitioning factor parameters is analyzed according to a gridding mechanism to generate a gridded decision, including:

[0025] According to the range partitioning strategy in the gridding mechanism, the first range in the area to be scheduled is obtained; according to the grid segmentation strategy in the gridding mechanism, the first range is segmented and analyzed to form the gridding decision; wherein, the process includes: uniformly gridding the area to be scheduled to obtain a set of grid blocks; sequentially extracting the first block and the second block from the set of grid blocks, and matching the first state of the first block and the second state of the second block in the real-time city state respectively; comparing the state similarity between the first state and the second state; if the state similarity reaches a predetermined limit, the first range is formed based on the first block and the second block.

[0026] Specifically, before gridding, the area to be dispatched is first determined according to the range partitioning strategy in the gridding mechanism. During this process, a preset grid size is obtained from the range partitioning strategy, and the area to be dispatched is uniformly divided starting from the upper left corner based on this grid size, resulting in a set of grid blocks. This set of grid blocks includes multiple grid blocks of equal size, such as the first block, the second block, the third block, etc. Subsequently, the first block and the second block are extracted sequentially from the grid block set and matched against real-time urban status data to obtain the first state of the first block and the second state of the second block. By weighting the urban planning deviation and geographic information deviation of the first and second states, the state similarity between the first and second states can be obtained. When the state similarity between the first and second states is greater than or equal to a predetermined limit, it indicates that the first and second blocks have similar electricity demand characteristics and resource allocation. In this case, the first and second blocks are merged to form the first area. Next, the third block is extracted sequentially, and the state similarity between the third state of the third block and the average first state of the current first block is calculated. If this state similarity also meets the preset limit, the third block is merged into the first block. This process is repeated until every grid block has been analyzed. For grid blocks not assigned to the first block, the same analysis and merging process is repeated sequentially to obtain the second block. This process continues until all grid blocks are assigned to a single block. Then, historical data for the first block is extracted from historical load records and historical renewable energy generation records. Based on predetermined grid load thresholds and predetermined grid power generation thresholds, the corresponding historical data is analyzed to generate a gridded decision. This gridded decision specifies the grid segmentation size, used to divide the first block into multiple smaller grid cells. The same operation is performed for other block ranges to obtain corresponding gridded decisions. These gridded decisions make power dispatch more flexible and adaptable, better able to cope with the uncertainty of renewable energy generation and load fluctuations, and improve the overall dispatch efficiency of the power grid.

[0027] In some embodiments, comparing the state similarity between the first state and the second state includes:

[0028] The system matches the first load density corresponding to the first planned land use type in the first state with the second load density corresponding to the second planned land use type in the second state; compares the first load density with the second load density to obtain the urban planning deviation; compares the first geographical feature parameter in the first state with the second geographical feature parameter in the second state to obtain the geographical information deviation; reads the predetermined allocation weights, and performs a weighted calculation on the urban planning deviation and the geographical information deviation based on the predetermined allocation weights to obtain the state similarity; wherein, the predetermined allocation weights include a first weight of the urban planning deviation and a second weight of the geographical information deviation, and the first weight is greater than the second weight.

[0029] Specifically, when calculating the state similarity between the first and second states, the first load density corresponding to the first planned land use type in the first state and the second load density corresponding to the second planned land use type in the second state are first obtained from the set of dividing factor parameters. Then, the first and second load densities are compared, and the absolute difference between the load densities is calculated and normalized using the maximum-minimum method to obtain the urban planning deviation. Next, the first geographical feature parameters in the first state and the second geographical feature parameters in the second state are compared, such as regional elevation, slope, temperature, wind speed, and solar radiation intensity. The absolute deviation of these geographical feature parameters is calculated and then normalized to obtain the geographical information deviation. Finally, the urban planning deviation and the geographical information deviation are weighted according to predetermined weights. These predetermined weights are determined by expert decision-making based on business needs; typically, the first weight for urban planning deviation is higher, and the second weight for geographical information deviation is lower. Finally, by subtracting 1 from the weighted sum, we can obtain the state similarity between the first and second states. This state similarity will be used to determine the degree of similarity between grid blocks, thereby helping to decide whether to merge the two regions to form a new block range.

[0030] Table 1: Examples of Planned Land Use Types and Load Densities

[0031]

[0032] Table 1 above is an example table of planned land use type and load density. The table shows different planned land use types and their corresponding load density values, which are used to reflect the load characteristics of various types of land use in electricity demand.

[0033] In some embodiments, the mesh segmentation strategy includes a predetermined mesh load threshold. Based on the mesh segmentation strategy in the meshing mechanism, the first block is segmented and analyzed to form the meshing decision, including:

[0034] Match the first historical total load corresponding to the first block range in the historical load record; match the first historical total power generation corresponding to the first block range in the historical new energy power generation record; based on any grid load in the predetermined grid load threshold, coordinate with the first historical total load to form a first initial splitting scheme, and use the first initial splitting scheme as the gridding decision; wherein, the first initial splitting scheme splits the first block range into grids of the first initial splitting grid number.

[0035] Specifically, firstly, multiple power loads corresponding to multiple locations within the first grid area are retrieved from historical load records. These loads are then accumulated to obtain the first historical total load data for the first grid area. Simultaneously, the first historical total power generation matching the first grid area is extracted from historical renewable energy generation records. Next, based on predetermined grid load thresholds stored in the grid partitioning strategy, an arbitrary grid load requirement is extracted (e.g., a maximum load of 200MW for a single grid). This extracted grid load requirement is then divided by the first historical total load to obtain the number of grids needed to partition the first grid area. Finally, based on the calculated number of grids, a first initial partitioning scheme is generated and used as part of the gridding decision-making process, providing foundational data for subsequent scheduling decisions.

[0036] In some embodiments, the grid partitioning strategy further includes a predetermined grid power generation threshold, and after using the first initial partitioning scheme as the gridding decision, it further includes:

[0037] The first grid set of the first block range is obtained by dividing the grid according to the gridding decision; the grid power generation is obtained by combining the first initial grid division number and the first historical total power generation; when the grid power generation is not at the predetermined grid power generation threshold, the first grid set is marked as a standby priority scheduling grid; wherein, the first grid set is marked as a standby priority in-grid grid when the grid power generation is less than the predetermined grid power generation threshold; and the first grid set is marked as a standby priority out-grid grid when the grid power generation is greater than the predetermined grid power generation threshold.

[0038] Specifically, after obtaining the gridded decision, it is applied to the first region. The first region is divided equally according to the initial grid number in the gridded decision, determining the first grid set for the first region. Then, the first historical total power generation is divided by the initial grid number to obtain the grid power generation of each grid in the first grid set. If this grid power generation is below the predetermined grid power generation threshold, the first grid set is marked as a reserve priority dispatch grid. That is, if the grid power generation is less than the minimum of the predetermined grid power generation threshold, the first grid set is marked as a reserve priority receiving grid, indicating that the grid set's power generation capacity is insufficient to meet demand, and power needs to be prioritized from grids with sufficient power generation capacity. Conversely, if the grid power generation is greater than the maximum of the predetermined grid power generation threshold, the first grid set is marked as a reserve priority dispatching grid, indicating that the grid's power generation capacity is excessive, and the excess power can be dispatched to other underloaded areas. By comparing with the predetermined power generation threshold, power dispatch can be optimized, enabling the grid to adjust resources according to actual conditions, improving the efficiency and stability of power dispatch.

[0039] Predictive analysis is performed on the target grid in the regional grid set to obtain target prediction information, which includes target predicted load data and target predicted power generation data.

[0040] Specifically, after obtaining the regional grid set, the grid to be predicted is selected as the target grid based on the current business needs. The future time points to be predicted for the target grid are then input into the corresponding fitting function to calculate the target predicted load data and target predicted power generation data for the target grid. By storing the target predicted load data and target predicted power generation data, the target prediction information of the target grid is obtained, providing data support for subsequent power dispatch, helping to optimize the operation of the power system and improve the allocation efficiency of power resources.

[0041] In some embodiments, predictive analysis is performed on the target grid in the regional grid set to obtain target prediction information, wherein the target prediction information includes target predicted load data and target predicted power generation data, including:

[0042] The process involves acquiring historical renewable energy power records for the target grid, including historical load records and historical power generation records; sequentially analyzing the historical load records to obtain the target predicted load data and analyzing the historical power generation records to obtain the target predicted power generation data; and including: performing regression fitting on the historical load time series of the historical load records to obtain a historical load fitting formula; reading the target time and calculating the target predicted load data of the target grid at the target time using the historical load fitting formula.

[0043] Specifically, before conducting predictive analysis on the target grid, historical renewable energy power records for the target grid are collected. These historical renewable energy power records include historical load records and historical power generation records for the region over a certain period of time. The historical load records provide information on the region's electricity demand, while the historical power generation records provide data on renewable energy generation. Subsequently, regression fitting is performed on the historical load records and historical power generation records respectively to construct historical load fitting equations and historical power generation fitting equations. By inputting the target time to be predicted into the historical load fitting equations and historical power generation fitting equations, the target predicted load data and target predicted power generation data for the target grid at the target time can be calculated. Taking historical load records as an example, the historical load time series data is first represented as a sample set (…). , ),in, This represents the i-th time point. This represents the corresponding actual load value. To fit the curve, the appropriate polynomial order can be selected by plotting a scatter plot, or by using cross-validation or an error metric (such as mean squared error, MSE) to select the polynomial order, such as a quadratic polynomial, thus constructing a prediction model. ,in, b and c are the regression coefficients to be solved. They can also be represented in vector form: Xβ; where X is the design matrix constructed from each time point, and β=[a,b,c Let n be the regression coefficient vector. In the case of a quadratic polynomial, if there are a total of n data points, then... , Then, to minimize the sum of squared errors. To achieve the objective, the least squares method is used to solve for the regression coefficient vector: L, thus, we can calculate the coefficient 'a' for the first component, the coefficient 'b' for the second component, and the constant term 'c' for the third component in the β vector, thereby deriving the polynomial regression equation, i.e., the historical load fitting equation, which represents the overall trend of the load. In summary, by predicting load and generation data, we can understand future electricity demand and generation capacity, providing key inputs for power dispatch optimization, helping the system dynamically adjust power distribution, and ensuring the stable and efficient operation of the power grid.

[0044] Using the scheduling fitness evaluation strategy as the scheduling optimization benchmark and the target predicted load data and the target predicted power generation data as scheduling optimization constraints, the optimal scheduling scheme is simulated and obtained.

[0045] Specifically, in the optimization of renewable energy power dispatch, a dispatch fitness evaluation strategy is first adopted as the benchmark for dispatch optimization. This strategy measures the merits of dispatch schemes based on multiple factors, including operating costs, wind and solar curtailment, and load-renewable energy matching, using a weighted approach. Then, target predicted load data and target predicted power generation data are used as constraints for dispatch optimization to set dispatch schemes. Subsequently, the generated dispatch schemes are simulated, and based on the simulation results, the dispatch fitness of each scheme is calculated using the fitness evaluation strategy. This same simulation and evaluation is then performed on other dispatch schemes to obtain the dispatch fitness of all schemes. Finally, the scheme with the highest dispatch fitness is selected as the optimal dispatch scheme. This optimal dispatch scheme satisfies the predicted load demand and power generation capacity while maximizing resource utilization efficiency and ensuring the stable operation of the power grid.

[0046] In some embodiments, using the scheduling fitness evaluation strategy as the scheduling optimization benchmark and the target predicted load data and the target predicted power generation data as scheduling optimization constraints, the optimal scheduling scheme is simulated and obtained, including:

[0047] Randomly generate any scheduling scheme, wherein the arbitrary scheduling scheme refers to transferring the renewable energy generation of any grid out of the reserve priority grid to any grid in the reserve priority grid; obtain arbitrary scheduling simulation records of the arbitrary scheduling scheme, and obtain the arbitrary scheduling fitness according to the scheduling optimization benchmark; determine the optimal scheduling scheme with the goal of maximizing the arbitrary scheduling fitness; wherein the renewable energy generation of the arbitrary grid out is greater than the demand scheduling power of the arbitrary grid in, and the demand scheduling power is obtained by comparing the predicted load data and the predicted power generation data of the arbitrary grid in.

[0048] Specifically, in the power dispatch optimization process, an arbitrary number of dispatch schemes are first generated. For any given scheme, a grid is randomly selected from the reserve priority dispatch grid, and its power generation is dispatched out. Then, a grid is randomly selected from the reserve priority dispatch grid, and the dispatched power generation is dispatched in. It is important to note that the renewable energy generation of the randomly selected reserve priority dispatch grid must be greater than the demand dispatch power of the randomly selected reserve priority dispatch grid. This demand dispatch power is obtained by subtracting the predicted load data from the predicted power generation data. Subsequently, any dispatch scheme is input into a pre-built simulation environment, and the dispatch process is simulated according to the dispatch scheme. The data generated during the simulation process are recorded, such as dispatch cost, energy storage loss cost, wind and solar curtailment (unutilized dispatchable power), and source-load matching degree (the ratio of actual dispatched power to demand dispatch power), forming a dispatch simulation record. Next, based on the operating cost weight, wind and solar curtailment weight, and source-load matching degree stored in the scheduling optimization benchmark, the normalized scheduling simulation records are weighted and summed to obtain the scheduling fitness of the scheduling scheme. Specifically, the operating cost weight is multiplied by the difference between 1 and the normalized value of the operating cost, and the wind and solar curtailment weight is multiplied by the difference between 1 and the normalized value of the wind and solar curtailment. Then, the scheduling fitness of all scheduling schemes is sorted, the maximum scheduling fitness is extracted, and this scheduling fitness is compared with a scheduling fitness threshold. If it is greater than or equal to the scheduling fitness threshold, the scheduling scheme corresponding to that scheduling fitness is taken as the optimal scheduling scheme; otherwise, any number of scheduling schemes are regenerated, and the above steps are repeated. Through this process, power dispatch can be ensured to be both economical and efficient, maximizing the utilization of new energy sources and reducing wind and solar curtailment.

[0049] The optimal scheduling scheme is used to schedule new energy power in the area to be scheduled.

[0050] Specifically, after scheduling optimization calculations and determining the optimal scheduling scheme, this scheme is applied to the area to be dispatched for renewable energy power dispatch. During this process, actual operations are performed on the corresponding locations within the area to be dispatched, according to the dispatch-out and dispatch-in grids set in the optimal scheduling scheme. By implementing the optimal scheduling scheme, the allocation of power resources within the area to be dispatched can be dynamically adjusted based on load demand, generation capacity, and dispatch constraints, ensuring the stable operation of the power grid and the efficient utilization of renewable energy resources.

[0051] In summary, the new energy power dispatch optimization method based on smart grid provided by this invention has the following technical effects:

[0052] Multi-dimensional data is collected from the area to be dispatched to obtain multi-dimensional information. Based on predetermined partitioning factors, the multi-dimensional information is traversed to obtain a partitioning factor parameter set. The partitioning factor parameter set is analyzed according to a gridding mechanism to generate gridded decisions, and a regional grid set is obtained based on the gridded decisions. Predictive analysis is performed on the target grids in the regional grid set to obtain target prediction information, which includes target predicted load data and target predicted power generation data. Using a dispatch fitness evaluation strategy as the dispatch optimization benchmark and the target predicted load data and target predicted power generation data as dispatch optimization constraints, an optimal dispatch scheme is simulated to obtain the optimal dispatch scheme. New energy power is dispatched to the area to be dispatched according to the optimal dispatch scheme. This achieves the technical effect of dynamically and accurately matching regional new energy power load and power generation capacity through multi-dimensional data-driven grid partitioning and predictive optimization, thereby improving the efficiency of renewable energy consumption and the real-time performance of grid dispatch decisions.

[0053] Example 2, as Figure 2 This is a schematic diagram of the structure of the new energy power dispatch optimization system based on intelligent grid according to the present invention. For example, Figure 1 The flowchart of the new energy power dispatch optimization method based on smart grid of the present invention can be illustrated as follows: Figure 2 The structure shown is implemented.

[0054] Based on the same concept as the smart grid-based new energy power dispatch optimization method in the above embodiments, the present invention also provides a smart grid-based new energy power dispatch optimization system comprising:

[0055] Data collection module 11: Collects multi-dimensional data of the area to be dispatched to obtain multi-dimensional information; Partitioning factor traversal module 12: Traverses the multi-dimensional information based on predetermined partitioning factors to obtain a partitioning factor parameter set; Grid partitioning module 13: Analyzes the partitioning factor parameter set according to the gridding mechanism, generates gridding decisions, and obtains a regional grid set based on the gridding decisions; Prediction analysis module 14: Performs prediction analysis on the target grid in the regional grid set to obtain target prediction information, wherein the target prediction information includes target predicted load data and target predicted power generation data; Dispatch optimization module 15: Simulates and obtains the optimal dispatching scheme using the dispatching fitness evaluation strategy as the dispatching optimization benchmark and the target predicted load data and the target predicted power generation data as dispatching optimization constraints; Power dispatching module 16: Performs new energy power dispatching on the area to be dispatched according to the optimal dispatching scheme.

[0056] In some embodiments, the data collection module 11 includes:

[0057] Obtain historical load records for the area to be dispatched; obtain historical renewable energy power generation records for the area to be dispatched; obtain real-time urban status for the area to be dispatched, wherein the real-time urban status includes real-time urban planning and real-time geographic information; construct the multi-dimensional information based on the historical load records, the historical renewable energy power generation records, the real-time urban planning, and the real-time geographic information.

[0058] In some embodiments, the mesh division module 13 includes:

[0059] According to the range partitioning strategy in the gridding mechanism, the first range in the area to be scheduled is obtained; according to the grid segmentation strategy in the gridding mechanism, the first range is segmented and analyzed to form the gridding decision; wherein, the process includes: uniformly gridding the area to be scheduled to obtain a set of grid blocks; sequentially extracting the first block and the second block from the set of grid blocks, and matching the first state of the first block and the second state of the second block in the real-time city state respectively; comparing the state similarity between the first state and the second state; if the state similarity reaches a predetermined limit, the first range is formed based on the first block and the second block.

[0060] In some embodiments, the mesh division module 13 further includes:

[0061] The system matches the first load density corresponding to the first planned land use type in the first state with the second load density corresponding to the second planned land use type in the second state; compares the first load density with the second load density to obtain the urban planning deviation; compares the first geographical feature parameter in the first state with the second geographical feature parameter in the second state to obtain the geographical information deviation; reads the predetermined allocation weights, and performs a weighted calculation on the urban planning deviation and the geographical information deviation based on the predetermined allocation weights to obtain the state similarity; wherein, the predetermined allocation weights include a first weight of the urban planning deviation and a second weight of the geographical information deviation, and the first weight is greater than the second weight.

[0062] In some embodiments, the mesh division module 13 includes:

[0063] Match the first historical total load corresponding to the first block range in the historical load record; match the first historical total power generation corresponding to the first block range in the historical new energy power generation record; based on any grid load in the predetermined grid load threshold, coordinate with the first historical total load to form a first initial splitting scheme, and use the first initial splitting scheme as the gridding decision; wherein, the first initial splitting scheme splits the first block range into grids of the first initial splitting grid number.

[0064] In some embodiments, the mesh division module 13 includes:

[0065] The first grid set of the first block range is obtained by dividing the grid according to the gridding decision; the grid power generation is obtained by combining the first initial grid division number and the first historical total power generation; when the grid power generation is not at the predetermined grid power generation threshold, the first grid set is marked as a standby priority scheduling grid; wherein, the first grid set is marked as a standby priority in-grid grid when the grid power generation is less than the predetermined grid power generation threshold; and the first grid set is marked as a standby priority out-grid grid when the grid power generation is greater than the predetermined grid power generation threshold.

[0066] In some embodiments, the predictive analysis module 14 includes:

[0067] The process involves acquiring historical renewable energy power records for the target grid, including historical load records and historical power generation records; sequentially analyzing the historical load records to obtain the target predicted load data and analyzing the historical power generation records to obtain the target predicted power generation data; and including: performing regression fitting on the historical load time series of the historical load records to obtain a historical load fitting formula; reading the target time and calculating the target predicted load data of the target grid at the target time using the historical load fitting formula.

[0068] In some embodiments, the scheduling optimization module 15 includes:

[0069] Randomly generate any scheduling scheme, wherein the arbitrary scheduling scheme refers to transferring the renewable energy generation of any grid out of the reserve priority grid to any grid in the reserve priority grid; obtain arbitrary scheduling simulation records of the arbitrary scheduling scheme, and obtain the arbitrary scheduling fitness according to the scheduling optimization benchmark; determine the optimal scheduling scheme with the goal of maximizing the arbitrary scheduling fitness; wherein the renewable energy generation of the arbitrary grid out is greater than the demand scheduling power of the arbitrary grid in, and the demand scheduling power is obtained by comparing the predicted load data and the predicted power generation data of the arbitrary grid in.

[0070] In embodiment three, the present invention also provides a computer-readable storage medium that can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the smart grid-based new energy power dispatch optimization method in the embodiments of the present invention, thereby realizing the above-mentioned smart grid-based new energy power dispatch optimization method.

[0071] It should be understood that the embodiments disclosed in this invention and the above description enable those skilled in the art to implement this invention. However, this invention is not limited to the embodiments mentioned above. It should be understood that those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this invention, and should all be included within the protection scope of this invention.

Claims

1. A new energy power dispatch optimization method based on smart grid, characterized in that, include: Multi-dimensional data collection is performed on the area to be scheduled to obtain multi-dimensional information; The multidimensional information is traversed based on predetermined dividing factors to obtain a set of dividing factor parameters. The set of partitioning factor parameters is analyzed according to the gridding mechanism to generate gridded decisions, and a regional grid set is obtained based on the gridded decisions. Predictive analysis is performed on the target grid in the regional grid set to obtain target prediction information, wherein the target prediction information includes target predicted load data and target predicted power generation data; Using the scheduling fitness evaluation strategy as the scheduling optimization benchmark and the target predicted load data and the target predicted power generation data as scheduling optimization constraints, the optimal scheduling scheme is simulated and obtained. The optimal scheduling scheme is used to schedule new energy power in the area to be scheduled. This includes conducting multi-dimensional data collection, which includes: Obtain the historical load records of the area to be scheduled; Obtain the historical renewable energy power generation records of the area to be dispatched; Obtain the real-time urban status of the area to be scheduled, wherein the real-time urban status includes real-time urban planning and real-time geographic information; The multidimensional information is constructed based on the historical load records, the historical renewable energy power generation records, the real-time urban planning, and the real-time geographic information. The analysis of the partitioning factor parameter set based on the gridding mechanism includes: According to the range partitioning strategy in the gridding mechanism, the first range in the region to be scheduled is obtained; Based on the grid segmentation strategy in the gridding mechanism, the first block is segmented and analyzed to form the gridding decision. This includes: The region to be scheduled is uniformly rasterized to obtain a set of raster blocks; The first block and the second block in the grid block set are extracted sequentially, and the first state of the first block and the second state of the second block are obtained by matching them in the real-time city state respectively; The similarity between the first state and the second state is obtained by comparison. If the state similarity reaches a predetermined limit, then the range of the first block is formed based on the first block and the second block; The comparison of the state similarity between the first state and the second state includes: The first load density corresponding to the first planned land use type in the first state and the second load density corresponding to the second planned land use type in the second state are respectively matched; The urban planning deviation is obtained by comparing the first load density with the second load density; By comparing the first geographic feature parameter in the first state with the second geographic feature parameter in the second state, the geographic information deviation is obtained; Read the predetermined allocation weights, and calculate the weighted average of the urban planning deviation and the geographic information deviation based on the predetermined allocation weights to obtain the state similarity; The predetermined allocation weights include a first weight for the urban planning deviation and a second weight for the geographic information deviation, wherein the first weight is greater than the second weight.

2. The new energy power dispatch optimization method based on smart grid as described in claim 1, characterized in that, The grid segmentation strategy includes a predetermined grid load threshold. Based on the grid segmentation strategy in the gridding mechanism, the first block is segmented and analyzed to form the gridding decision, including: Match the first historical total load corresponding to the first block range in the historical load record; Match the first historical total power generation corresponding to the first block range in the historical new energy power generation records; Based on any grid load in the predetermined grid load threshold, a first initial splitting scheme is formed in conjunction with the first historical total load, and the first initial splitting scheme is used as the gridding decision. The first initial segmentation scheme divides the first block range into a grid of the first initial segmentation grid number.

3. The new energy power dispatch optimization method based on smart grid as described in claim 2, characterized in that, The grid partitioning strategy also includes a predetermined grid power generation threshold, and after using the first initial partitioning scheme as the gridding decision, it further includes: The first grid set of the first block range is obtained by dividing it according to the gridding decision; The grid power generation is obtained by combining the first initial grid number with the first historical total power generation; When the grid power generation is not at the predetermined grid power generation threshold, the first grid set is marked as a standby priority scheduling grid; This includes: When the grid power generation is less than the predetermined grid power generation threshold, the first grid set is marked as a backup priority grid to be loaded. When the power generation of the grid exceeds the predetermined power generation threshold, the first grid set is marked as a standby priority grid to be called out.

4. The new energy power dispatch optimization method based on smart grid as described in claim 1, characterized in that, Predictive analysis is performed on the target grid within the aforementioned regional grid set to obtain target prediction information, wherein the target prediction information includes target predicted load data and target predicted power generation data, including: Obtain historical renewable energy power records for the target grid, wherein the historical renewable energy power records include historical load records and historical power generation records; The target predicted load data is obtained by sequentially analyzing the historical load records, and the target predicted power generation data is obtained by analyzing the historical power generation records. This includes: The historical load time series of the historical load records are subjected to regression fitting to obtain the historical load fitting formula; The target time is read, and the target predicted load data of the target grid at the target time is calculated by the historical load fitting formula.

5. The new energy power dispatch optimization method based on smart grid as described in claim 3, characterized in that, Using the scheduling fitness evaluation strategy as the scheduling optimization benchmark and the target predicted load data and the target predicted power generation data as scheduling optimization constraints, the optimal scheduling scheme is simulated and obtained, including: Randomly generate any scheduling scheme, wherein the arbitrary scheduling scheme refers to transferring the renewable energy generation from any grid in the backup priority grid to any grid in the backup priority grid; Obtain arbitrary scheduling simulation records of the arbitrary scheduling scheme, and obtain the fitness of arbitrary scheduling based on the scheduling optimization benchmark; The optimal scheduling scheme is determined with the goal of maximizing the fitness of any given schedule. Wherein, the renewable energy generation of any grid being transferred out is greater than the demand dispatch power of any grid being transferred in, and the demand dispatch power is obtained by comparing the predicted load data and the predicted power generation data of any grid being transferred in.

6. A new energy power dispatching and optimization system based on intelligent grids, characterized in that, The method for optimizing new energy power dispatch based on smart grids as described in any one of claims 1-5 includes: Data collection module: Collects multi-dimensional data from the area to be scheduled to obtain multi-dimensional information; Dividing factor traversal module: Based on predetermined dividing factors, the multidimensional information is traversed to obtain a set of dividing factor parameters; Grid partitioning module: Analyzes the set of partitioning factor parameters according to the gridding mechanism, generates gridding decisions, and obtains a regional grid set based on the gridding decisions; Predictive analysis module: performs predictive analysis on the target grid in the regional grid set to obtain target prediction information, wherein the target prediction information includes target predicted load data and target predicted power generation data; The scheduling optimization module uses the scheduling fitness evaluation strategy as the scheduling optimization benchmark and the target predicted load data and the target predicted power generation data as scheduling optimization constraints to simulate and obtain the optimal scheduling scheme. Power dispatching module: Dispatches new energy power to the area to be dispatched according to the optimal dispatching scheme.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the smart grid-based new energy power dispatch optimization method as described in any one of claims 1 to 5.

Images