Main-distribution-microgrid cooperative scheduling method and system based on seasonal characteristics

Abstract

The invention discloses a main-distribution-micro power grid cooperative scheduling method and system based on seasonal characteristics, and belongs to the technical field of power grid cooperative scheduling. Power grid operation data are preliminarily screened to obtain a typical daily operation curve, and the typical daily operation curve is used as an optimization object; and performing seasonal characteristic analysis to obtain a seasonal characteristic quantitative index, thereby enabling the constraint condition to dynamically adapt to the season, improving the feasibility of strategy implementation, constructing a target function by constructing the highest comprehensive benefit of the power grid as the target, and solving the target function, thereby not only improving the overall economic benefit of the power grid, but also improving the economic benefit of the power grid. And the objective function is guided to preferentially optimize the high-potential new energy through a seasonal characteristic quantitative index, so that the wind and light abandoning rate is greatly reduced, and the development demand of high-proportion new energy permeation of a novel power system is met.

Description

Technical Field

[0001] This invention relates to the field of power grid coordinated dispatch technology, specifically to a method and system for coordinated dispatch of primary, distribution and microgrids based on seasonal characteristics. Background Technology

[0002] To address the increasingly prominent issues of energy security, environmental pollution, and climate change, the transformation of the power system towards low-carbon, clean, and sustainable development is being actively promoted. Building a new power system with a high proportion of renewable energy penetration is an effective way to achieve these goals. However, with the large-scale integration of distributed generation, electric vehicles, energy storage, and other diverse entities, traditional power systems face numerous new opportunities and challenges, including the intermittency and volatility of high-proportion renewable energy, the complexity of massive resource regulation, and insufficient coordination across multiple levels of the main grid, distribution network, and microgrid. This significantly increases the complexity, uncertainty, and vulnerability of system operation. On the one hand, the abundance of distributed generation resources effectively improves power supply flexibility and economy, reducing dependence on centralized power sources in the main grid, and the distribution network's operation mode is gradually shifting from passive participation to active support. On the other hand, as the proportion of distributed renewable energy generation capacity increases, the inherent randomness of renewable energy leads to increased volatility in the net load of the distribution network and microgrid. This places an additional burden on the boundary power matching between the main grid, distribution network, and microgrid. If resources within the multi-level grid cannot be effectively coordinated, system safety risks such as power imbalances and voltage exceeding limits may occur in local grids.

[0003] The main grid is the backbone of the power system, undertaking the critical task of long-distance, high-capacity power transmission. The distribution network, an extension of the main grid, directly serves end users, responsible for distributing and supplying power to households and various industrial and commercial users. While significantly different in voltage levels, functional positioning, network structure, and operating modes, the two are closely coupled, together forming the "aorta" and "capillaries" of the power flow network. Microgrids, as the "cellular units" of the new power system, can improve local self-balancing capabilities through integrated aggregation of source, grid, load, and storage; however, their coordinated dispatch mechanism with the main grid and distribution network is still imperfect.

[0004] Chinese Patent, Publication No. CN119419804A, Publication Date: February 11, 2025, discloses a multi-level collaborative optimization control method and system for main-distribution-microgrids adapted to non-smooth characteristics. It constructs an equivalent model of the converter and the grid topology based on grid measurement data; obtains the piecewise function of the three-phase reactive power of distributed energy sources based on the equivalent model and the droop control principle; approximates the droop control function using a fitting function; decouples the microgrid using node replication and microgrid adjacency relationships; constructs a multi-level collaborative optimization scheduling model using the droop control function, microgrid decoupling, and optimization objective function; and outputs the main-distribution-microgrid control strategy using the interior-point method based on real-time grid data and the multi-level collaborative optimization scheduling model combined with a time-series consistency strategy. However, it does not consider the impact of different seasons on the energy output of the grid, making it difficult to dynamically adjust the scheduling strategy. Summary of the Invention

[0005] This invention addresses the problem of poor grid operation efficiency caused by the increasing complexity of resource regulation due to the surge in the number of multi-source energy storage entities connected to the main-distribution-microgrid system. It provides a coordinated scheduling method and system for the main-distribution-microgrid system based on seasonal characteristics. By initially screening grid operation data to obtain typical daily operation curves, and using these curves as optimization targets, seasonal characteristic analysis is performed to obtain quantitative indicators of seasonal characteristics. This allows constraints to dynamically adapt to the seasons, improving the feasibility of strategy implementation. Furthermore, by constructing an objective function with the goal of maximizing the overall grid efficiency and solving it, not only is the overall economic efficiency of the grid improved, but the seasonal characteristic quantitative indicators also guide the objective function to prioritize the optimization of high-potential renewable energy sources, significantly reducing wind and solar curtailment rates and meeting the development needs of high-proportion renewable energy penetration in new power systems.

[0006] In a first aspect, one technical solution provided in this embodiment of the invention is: a main-distribution-microgrid coordinated scheduling method based on seasonal characteristics, comprising the following steps: S1. Based on the typical day screening principle, the power grid operation data is processed to obtain the typical daily operation curves of the main-distribution-microgrid; S2. Extract seasonal characteristics from typical daily operating curves to obtain seasonal characteristic quantitative indicators of the main-distribution-microgrid, and determine grid operation constraints based on seasonal characteristic quantitative indicators; S3. Under the constraints of power grid operation and preset resource constraints, construct an objective function with the goal of maximizing the comprehensive benefits of coordinated dispatch of main-distribution-microgrid; solve the objective function to obtain the coordinated dispatch strategy under different seasonal scenarios.

[0007] In this scheme, by selecting typical days, the 8760 hours of the whole year are condensed into typical day curves for each season. This eliminates the need to solve the massive amount of data throughout the year hourly, optimizing only for typical day scenarios. This significantly reduces the computational load of solving the objective function and avoids solution delays caused by data overload, allowing the scheduling scheme to respond quickly to seasonal changes. By extracting seasonal features and transforming them into grid operation constraints, the constraints can be dynamically adapted to seasonal changes. This also allows for the correlation of multi-level grids, strengthening the coordination of each level of the grid. As a result, the final coordinated scheduling strategy optimizes the operation of each level while ensuring overall coordination. By solving for seasonal scheduling strategies under constraints with the goal of maximizing comprehensive benefits, this scheme not only maximizes resource utilization efficiency in different seasons and reduces ineffective costs, but also prioritizes the optimization of high-potential new energy sources, significantly reducing wind and solar curtailment rates. Furthermore, it can respond to different scheduling strategies according to different seasons, making coordinated scheduling more flexible and adaptable.

[0008] Preferably, in S1, the power grid operation data is processed based on the typical day screening principle to obtain the typical daily operation curves of the main-distribution-microgrid, including the following steps: The annual renewable energy output data and load power data of the main-distribution-microgrid are collected and normalized to obtain the operation dataset; The number of clusters is determined to be k based on the seasonal quantity. A data point is randomly selected from the running dataset as the initial cluster center. The shortest distance between all other data points and the initial cluster center is calculated to obtain the probability distribution of all other data points being selected as cluster centers. The top k-1 data points with the highest probability are taken as target cluster centers. Clustering is performed on the data in the running dataset based on the initial cluster centers and target cluster centers to obtain clustered data. The clustered data is then divided into seasonal scenarios and recorded based on time series to obtain the typical daily operating curves of the main-distribution-microgrid.

[0009] In this scheme, data normalization eliminates the magnitude difference between renewable energy output and load power, ensuring the objectivity of clustering results and avoiding data bias from affecting subsequent analysis. By selecting cluster centers based on probability distribution, the initial center distribution is made more uniform, and the clustering results are more in line with the actual operating characteristics of summer, winter, and spring and autumn, accurately extracting the core output and load curves of each level of the main-distribution-microgrid. By condensing massive amounts of data throughout the year into typical daily curves, the computational redundancy of the subsequent objective function is greatly reduced, improving the solution efficiency, while retaining seasonal differences. This lays a precise data foundation for the subsequent extraction of seasonal characteristic quantitative indicators and constraint setting, avoiding the blindness of traditional scheduling.

[0010] Preferably, the clustered data is divided into seasonal scenarios and recorded based on time series to obtain typical daily operating curves of the main-distribution-microgrid system, including the following steps: Clustered data with photovoltaic output greater than or equal to the output threshold and the proportion of output during midday hours greater than or equal to the proportion threshold are defined as summer scenario data; Clustered data with wind power output greater than or equal to the output threshold and the proportion of output during nighttime hours greater than or equal to the proportion threshold are defined as winter scenario data; The remaining clustering data will be classified as spring and autumn scene data; The summer, winter, and spring / autumn scenario data were recorded based on time series to obtain the typical daily operation curves of the main-distribution-microgrid in summer, winter, and spring / autumn respectively.

[0011] This scheme categorizes scenarios based on the core characteristics of photovoltaic power output at midday and wind power output at night, perfectly aligning with the energy output patterns of summer and winter, allowing typical daily curves to accurately reflect the essence of the seasons. It clearly distinguishes and records the data of the three scenarios in time sequence, precisely preserving the core seasonal curves of each level of the power grid, providing an accurate data source for the subsequent extraction of quantitative indicators of seasonal characteristics. This avoids errors caused by mixed data and reduces subsequent computational redundancy, laying a solid foundation of accurate data for the formulation of seasonal scheduling strategies.

[0012] As a preferred embodiment, in S2, seasonal characteristic quantitative indicators of the main-distribution-microgrid are obtained by extracting seasonal features from typical daily operating curves, including the following steps: The correction coefficient for new energy output is determined based on the average daily output of new energy in different seasons and the average daily output of new energy throughout the year in the typical daily operation curve. The load peak-valley difference threshold is determined based on the load peak-valley difference and peak-valley ratio in different seasons in the typical daily operation curve; The output-load matching degree is determined based on the spatiotemporal matching degree of new energy output and load in different seasons in typical daily operation curves; The new energy output correction coefficient, the load peak-valley difference threshold, and the output-load matching degree are used as quantitative indicators of the seasonal characteristics of the main-distribution-microgrid.

[0013] In this scheme, the difference in power generation capacity in different seasons can be accurately quantified by the new energy output correction coefficient, providing quantitative support for the dynamic adjustment of power output constraints of wind and solar units and avoiding the impact of one-size-fits-all constraints on the final control effect. The load peak-valley difference threshold can clarify the load fluctuation bottom line for stable grid operation, transforming abstract stability requirements into quantitative standards, thereby reducing the risk when implementing control strategies. The spatiotemporal adaptability of the output-load matching degree energy quantification strategy provides trigger thresholds for hierarchical scheduling strategies. The quantitative index system composed of these three elements makes the subsequent constraints more in line with seasonal characteristics, thereby making the solution results more accurate. At the same time, it provides a clear decision-making basis for dynamic scheduling strategies, avoiding the subjectivity and blindness of traditional scheduling and consolidating the quantitative foundation of collaborative scheduling.

[0014] Preferably, the output-load matching degree includes single-level matching degree, multi-level matching degree, and seasonal comprehensive matching degree; The single-level matching degree is used to quantify the renewable energy absorption capacity and load matching effect of a certain level in the main-distribution-microgrid in different seasons; The multi-level matching degree is used to quantify the cross-level energy utilization rate between various levels in the main-distribution-microgrid; The seasonal comprehensive matching degree is used to quantify the annual renewable energy absorption capacity and load matching effect in the main-distribution-microgrid.

[0015] In this scheme, the single-level matching degree focuses on the single-level power grid, providing a clear trigger basis for dispatching actions at each level and avoiding the blindness of hierarchical control; the multi-level matching degree quantifies the cross-level energy utilization rate, guiding the synergistic complementarity of the main grid, distribution grid, and microgrid, thereby avoiding local optima but global imbalance; the seasonal comprehensive matching degree takes into account the whole-year scenario and is used to balance the optimization priorities of different seasons. The three factors, from local to global and from single season to whole year, comprehensively quantify the effect of new energy consumption and load matching, making subsequent constraint setting and strategy formulation more accurate, thus building a quantitative foundation for coordinated dispatch.

[0016] Preferably, in S3, the objective function aims to minimize the sum of the main grid loss cost and the wind and solar curtailment penalty cost of the main-distribution-microgrid. The main grid loss cost is the product of the unit grid loss cost coefficient and the main grid loss; the wind and solar curtailment penalty cost is the product of the sum of the wind and solar curtailment of the main-distribution-microgrid and the unit wind and solar curtailment penalty cost coefficient.

[0017] This solution directly focuses on two key costs: grid loss and wind / solar curtailment. The objectives are clear and easily quantifiable. It also aligns with seasonal scheduling needs, guiding scheduling strategies to prioritize the consumption of new energy sources, thereby reducing the penalty for curtailment and making scheduling both economical and efficient.

[0018] Preferably, in S3, the process of solving the objective function specifically includes: An improved particle swarm optimization algorithm is adopted, and several output curves and load curves are randomly generated as scheduling schemes. The scheduling schemes are set as particles, and the main grid loss cost and the wind and solar curtailment penalty cost of the main-distribution-microgrid are calculated as the fitness value for each particle. The number of iterations, the initial learning factor and the initial velocity of the particles are set. Set the minimum fitness value as the population optimum, adjust the learning factor based on the number of iterations, update the generation rate of each particle based on the learning factor and the population optimum, and update the particles. The fitness value of each particle is calculated until the change is less than the change threshold or the iteration ends when the population reaches its optimal value after n consecutive iterations, thus obtaining the optimal output-load curve of the main-distribution-microgrid.

[0019] In this scheme, an improved particle swarm optimization algorithm is used to solve the objective function, covering all possible resource combinations of the main grid, distribution grid, and microgrid. By combining seasonal characteristics, the optimal solution adapted to the seasonal scenario is avoided, thus avoiding the pitfall of local optima in traditional algorithms. The design of dynamically adjusting the learning factor balances the exploratory nature of the initial search with the convergence of the later search. Combined with the iteration termination condition, it not only improves the solution efficiency but also ensures the stability of the results. Particle updates are based on the population optimum, thereby guiding all scheduling schemes to move towards the global optimum. This ensures that the final output optimal output-load curve satisfies both single-level optimization and cross-level coordination of the main grid, distribution grid, and microgrid. As a result, the scheduling strategy is both economical and in line with seasonal characteristics and hierarchical requirements, improving the feasibility of the scheduling strategy.

[0020] Preferably, in S3, the process of obtaining the collaborative scheduling strategy for different seasonal scenarios includes: The optimal output-load curve of the main-distribution-microgrid is divided into seasonal scenarios to obtain the output-load curves for different seasons; The optimal output-load matching degree for different seasons is determined based on the spatiotemporal matching degree of the output-load curves for different seasons. The corresponding scheduling strategy is queried from the preset scheduling strategy comparison table based on the optimal output-load matching degree, which serves as the collaborative scheduling strategy for different seasonal scenarios.

[0021] In this scheme, the optimal output-load curve is divided seasonally to ensure that the strategy conforms to the energy and load patterns of different seasons, avoiding the rigidity of using the same strategy throughout the year. The optimal matching degree is determined by the spatiotemporal matching degree, which is directly linked to a multi-dimensional matching degree index system, so that the strategy triggering has a clear quantitative basis rather than subjective judgment, thus improving the scheduling accuracy. The overall scheme combines the optimal solution of the objective function with seasonal characteristics, so that the final strategy not only meets the core goal of cost reduction and absorption, but also has strong engineering feasibility, perfectly adapting to the scheduling needs of the new power system.

[0022] Preferably, the resource constraints include constraints on thermal power units, wind power / photovoltaic units, energy storage systems, and grid operation safety constraints. The power grid operation constraints include load peak-valley difference constraints, minimum matching degree, and output boundary.

[0023] In this scheme, constraints on thermal power units and energy storage are used to avoid equipment overload and overcharging / over-discharging, while grid security constraints are used to prevent load fluctuations and matching imbalance risks. Constraints such as load peak-valley difference and minimum matching degree take into account seasonal characteristics and matching degree indicators, making the constraints more in line with the needs of seasonal and multi-level coordination, thereby ensuring that the dispatch strategy does not break through the safety boundaries of equipment and grid while reducing costs and absorbing the load, thus improving the reliability of the scheme implementation.

[0024] Secondly, one technical solution provided in this embodiment of the invention is: a main-distribution-microgrid coordinated dispatching system based on seasonal characteristics, including a data acquisition module, a data processing module, a feature extraction module, a calculation module, and a coordinated dispatching module; The data acquisition module is used to collect operational data of the main-distribution-microgrid. The data processing module processes the power grid's operating data based on the typical day screening principle to obtain the typical daily operating curves of the main-distribution-microgrid. The feature extraction module extracts seasonal features from typical daily operating curves to obtain seasonal characteristic quantitative indicators of the main-distribution-microgrid, and generates grid operation constraints based on the seasonal characteristic quantitative indicators. Under the constraints of power grid operation, the calculation module constructs an objective function with the goal of maximizing the comprehensive benefits of coordinated scheduling of the main grid, distribution grid, and microgrid, and solves it to obtain the coordinated scheduling strategies under different seasonal scenarios. The coordinated scheduling module adjusts the power output allocation and energy storage operation status of each level of the power grid in the main-distribution-microgrid based on the coordinated scheduling strategy.

[0025] In this solution, a corresponding system is built to integrate the collaborative scheduling method, thereby enabling human-computer interaction and improving the user experience.

[0026] The beneficial effects of this invention are as follows: This invention obtains typical daily operating curves by initially screening power grid operation data, uses these typical daily operating curves as optimization objects, and performs seasonal characteristic analysis and quantification of seasonal characteristics, thereby dynamically adapting the constraints to the seasons and improving the feasibility of strategy implementation. Furthermore, by constructing an objective function with the goal of maximizing the overall benefits of the power grid and solving it, this invention not only improves the overall economic benefits of the power grid, but also guides the objective function to prioritize the optimization of high-potential new energy sources through the quantification of seasonal characteristics, significantly reducing the wind and solar curtailment rate, which meets the development needs of high-proportion new energy penetration in the new power system.

[0027] The above description of the invention is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0028] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings. The drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings.

[0029] Figure 1 This is a flowchart of the main-distribution-microgrid coordinated scheduling method based on seasonal characteristics according to the present invention; Figure 2 Typical intraday curves for wind and solar power and load power in the main grid; Figure 3 Typical intraday curves for wind and solar power and load power in the distribution network; Figure 4 Typical intraday curves for microgrid wind and solar power load; Figure 5 This is a schematic diagram of the main-distribution-microgrid coordinated dispatching system based on seasonal characteristics according to the present invention; Figure 6 This is a power grid topology diagram for a specific scenario in this embodiment; Figure 7 This is a schematic diagram illustrating the multi-level matching degree in different seasons under a certain scenario in this embodiment; Figure 8 This is a schematic diagram illustrating the scheduling costs for different seasons in a specific scenario of this embodiment; Figure 9 This is a schematic diagram illustrating the matching degree distribution of each single level in a certain scenario of this embodiment; Figure 10 This is a diagram comparing the total cost of the scheduling scheme of this application with that of a conventional scheduling scheme in this embodiment; Figure 11 This is a schematic diagram comparing network loss using the scheduling scheme of this application with that of a common scheduling scheme in this embodiment; Figure 12 This is a schematic diagram comparing the wind and solar curtailment rates of the scheduling scheme of this application with those of a normal scheduling scheme in this embodiment. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only one preferred embodiment of this invention and are only used to explain this invention. They do not limit the scope of protection of this invention. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0031] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the operations (or steps) as sequential processes, many of the operations (or steps) can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but it may also have additional steps not included in the figures; the process may correspond to a method, function, procedure, subroutine, subroutine, etc.

[0032] Example 1: As Figure 1 As shown, to address the problem of poor grid operation efficiency caused by the increasing complexity of resource regulation due to the surge in the number of multi-source energy storage entities connected to the main-distribution-microgrid system, this embodiment provides a main-distribution-microgrid collaborative scheduling method based on seasonal characteristics, including the following steps: S1: Based on the typical day screening principle, the power grid operation data is processed to obtain the typical daily operation curves of the main-distribution-microgrid.

[0033] In this embodiment, the typical daily operation curves of the main-distribution-microgrid are obtained by processing the power grid operation data based on the typical daily screening principle, including the following steps: The annual renewable energy output data and load power data of the main-distribution-microgrid are collected and normalized to obtain the operation dataset; The number of clusters is determined to be k based on the seasonal quantity. A data point is randomly selected from the running dataset as the initial cluster center. The shortest distance between all other data points and the initial cluster center is calculated to obtain the probability distribution of all other data points being selected as cluster centers. The top k-1 data points with the highest probability are taken as target cluster centers. Clustering is performed on the data in the running dataset based on the initial cluster centers and target cluster centers to obtain clustered data. The clustered data is then divided into seasonal scenarios and recorded based on time series to obtain the typical daily operating curves of the main-distribution-microgrid.

[0034] Specifically, the clustering algorithm used can be k-means++, and the specific clustering process is as follows: First, a point is randomly selected from the running dataset as the first cluster center, i.e., the initial cluster center; then, for each point in the running dataset... Calculate its shortest distance to the initial cluster center, denoted as . Then calculate the probability that each point will be selected as the next cluster center. The farther a point is from the existing center, the higher its probability of being selected, as expressed by the formula below: Then, the roulette wheel selection method is used to select the next cluster center based on the probability distribution. Specifically, a random number in the interval [0,1] is generated, and then the probability interval of which point it falls within is determined. The above steps are repeated until k cluster centers are selected. Since this embodiment divides the seasons into summer, winter, and spring / autumn, k is chosen as 3. Once k initial centers are selected, the subsequent process is exactly the same as the standard k-means algorithm: assigning points to the nearest center, updating the center positions, and iterating until convergence is obtained to obtain 3 types of cluster data.

[0035] This embodiment eliminates the magnitude difference between renewable energy output and load power through data normalization, ensuring the objectivity of clustering results and avoiding data bias from affecting subsequent analysis. By selecting cluster centers based on probability distribution, the initial center distribution is made more uniform, and the clustering results are more in line with the actual operating characteristics of summer, winter, and spring and autumn, accurately extracting the core output and load curves of each level of the main-distribution-microgrid. By condensing massive amounts of data throughout the year into typical daily curves, the computational redundancy of the subsequent objective function is greatly reduced, improving the solution efficiency, while retaining seasonal differences. This lays a precise data foundation for the subsequent extraction of seasonal characteristic quantitative indicators and constraint setting, avoiding the blindness of traditional scheduling.

[0036] In this embodiment, the clustered data is divided into seasonal scenarios and recorded based on time series to obtain the typical daily operation curves of the main-distribution-microgrid, including the following steps: Clustered data with photovoltaic output greater than or equal to the output threshold and the proportion of output during midday hours greater than or equal to the proportion threshold are defined as summer scenario data; Clustered data with wind power output greater than or equal to the output threshold and the proportion of output during nighttime hours greater than or equal to the proportion threshold are defined as winter scenario data; The remaining clustering data will be classified as spring and autumn scene data; The summer, winter, and spring / autumn scenario data were recorded based on time series to obtain the typical daily operation curves of the main-distribution-microgrid in summer, winter, and spring / autumn respectively.

[0037] like Figure 2 , Figure 3 and Figure 4 The figures shown are typical intraday curves of wind and solar power and load power in the main grid, distribution grid, and microgrid, respectively.

[0038] This embodiment divides scenarios according to the core characteristics of photovoltaic midday power output and wind power nighttime power output, perfectly matching the energy output patterns of summer and winter, allowing typical daily curves to truly reflect the essence of the seasons; it clearly distinguishes and records the data of the three types of scenarios in time sequence, accurately retaining the core seasonal curves of each level of the power grid, providing an accurate data source for the subsequent extraction of seasonal characteristic quantitative indicators, thereby avoiding errors caused by data mixing, reducing subsequent computational redundancy, and laying a solid accurate data foundation for the formulation of seasonal scheduling strategies.

[0039] S2: Extract seasonal characteristics from typical daily operating curves to obtain seasonal characteristic quantitative indicators of the main-distribution-microgrid, and determine grid operation constraints based on seasonal characteristic quantitative indicators.

[0040] In this embodiment, seasonal characteristic quantitative indicators of the main-distribution-microgrid are obtained by extracting seasonal features from typical daily operating curves, including the following steps: The correction coefficient for new energy output is determined based on the average daily output of new energy in different seasons and the average daily output of new energy throughout the year in the typical daily operation curve. The load peak-valley difference threshold is determined based on the load peak-valley difference and peak-valley ratio in different seasons in the typical daily operation curve; The output-load matching degree is determined based on the spatiotemporal matching degree of new energy output and load in different seasons in typical daily operation curves; The new energy output correction coefficient, the load peak-valley difference threshold, and the output-load matching degree are used as quantitative indicators of the seasonal characteristics of the main-distribution-microgrid.

[0041] Specifically, the new energy output correction factor is a multiplier that scales the predicted wind and solar power output based on seasonal characteristics. It is used to accurately reflect the inherent new energy power generation capacity in different seasons. In summer, when photovoltaic output is strong and wind power output is weak, the comprehensive new energy output correction factor is 1.1. In winter, when photovoltaic output is weak and wind power output is strong, the comprehensive new energy output correction factor is 1.05. In spring and autumn, when photovoltaic and wind power output are slightly lower than the annual average, the comprehensive new energy output correction factor is 0.95.

[0042] The load peak-valley difference threshold is an upper limit for the allowable fluctuation between peak and off-peak electricity consumption in different seasons to ensure grid stability. Therefore, the load peak-valley difference is an important indicator for measuring the stability and economy of grid operation. In different seasons, users' electricity consumption behavior patterns are different, resulting in different natural peak-valley differences. The load demand peak-valley difference threshold ensures the safety and stability of the grid and directly transforms the grid operation stability into a specific quantitative target. The specific acquisition process is as follows: calculate the load peak-valley difference (i.e., the daily maximum load minus the daily minimum load) and the peak-valley ratio (i.e., the ratio of the daily maximum load to the daily minimum load) for typical days in three seasons. Then, according to the requirements for stable grid operation, take the maximum value of the peak-valley ratio of the three seasons as a unified threshold. In this embodiment, the value is taken as 1.5, that is, the peak-valley ratio in the load peak-valley difference constraint is less than or equal to 1.5.

[0043] This embodiment uses a new energy output correction coefficient to accurately quantify the differences in power generation capacity across different seasons, providing quantitative support for the dynamic adjustment of power output constraints for subsequent wind and solar turbines, and avoiding the impact of a one-size-fits-all approach on the final control effect. The load peak-valley difference threshold clarifies the load fluctuation baseline for stable grid operation, transforming abstract stability requirements into quantitative standards, thereby reducing the risk during the execution of control strategies. The spatiotemporal adaptability of the output-load matching degree energy quantification strategy provides trigger thresholds for hierarchical scheduling strategies. The quantitative index system composed of these three elements allows subsequent constraints to better align with seasonal characteristics, resulting in more accurate solutions. Simultaneously, it provides a clear decision-making basis for dynamic scheduling strategies, avoiding the subjectivity and blindness of traditional scheduling, and solidifying the quantitative foundation for collaborative scheduling.

[0044] In this embodiment, the output-load matching degree includes single-level matching degree, multi-level matching degree, and seasonal comprehensive matching degree; The single-level matching degree is used to quantify the renewable energy absorption capacity and load matching effect of a certain level in the main-distribution-microgrid in different seasons; The multi-level matching degree is used to quantify the cross-level energy utilization rate between various levels in the main-distribution-microgrid; The seasonal comprehensive matching degree is used to quantify the annual renewable energy absorption capacity and load matching effect in the main-distribution-microgrid.

[0045] Specifically, the formula for calculating the matching degree of a single level is as follows: Where s represents the season type (1 for summer / 2 for winter / 3 for spring and autumn), P res L(t) represents the output of new energy sources (MW) during time period t, L(t) represents the load power (MW) during time period t, and α(s) season represents the seasonal correction factor (1.1 for summer, 1.05 for winter, and 0.95 for spring and autumn).

[0046] Multi-level matching degree The calculation is based on a single-level matching degree, using the following formula: in, , and These are the main network weight, distribution network weight, and microgrid weight (specifically, the values ​​are 0.5 for the main network, 0.3 for the distribution network, and 0.2 for the microgrid). The single-layer matching degree of the mainnet, P trans The actual transmitted power of the tie line must meet power balance constraints. Contribute to new energy sources For load power, For the single-layer matching degree of the distribution network, This represents the single-layer matching degree of the microgrid.

[0047] The seasonal overall matching degree is calculated based on multi-level matching degrees, using the following formula: Among them, D s D is the number of days in season s (90 days in summer / 90 days in winter / 185 days in total for spring and autumn). year This represents the number of days in a year (365 days).

[0048] This embodiment focuses on single-level matching degree for single-level power grid, providing clear triggering basis for scheduling actions at each level and avoiding blindness in hierarchical control; multi-level matching degree quantifies cross-level energy utilization rate, guiding the synergistic complementarity of main-distribution-microgrid, thereby avoiding local optima but global imbalance; seasonal comprehensive matching degree takes into account the whole year scenario and is used to balance the optimization priorities of different seasons. The three factors, from local to global and from single season to whole year, comprehensively quantify the effect of new energy consumption and load matching, making subsequent constraint setting and strategy formulation more accurate, thus building a quantitative foundation for coordinated scheduling.

[0049] S3: Under the constraints of power grid operation and preset resource constraints, construct an objective function with the goal of maximizing the comprehensive benefits of coordinated dispatch of main-distribution-microgrid; solve the objective function to obtain the coordinated dispatch strategy under different seasonal scenarios.

[0050] In this embodiment, the objective function aims to minimize the sum of the main grid loss cost and the wind and solar curtailment penalty cost of the main-distribution-microgrid. The main grid loss cost is the product of the unit grid loss cost coefficient and the main grid loss; the wind and solar curtailment penalty cost is the product of the sum of the wind and solar curtailment of the main-distribution-microgrid and the unit wind and solar curtailment penalty cost coefficient.

[0051] Specifically, the objective function is expressed as follows: Among them, c loss P is the unit network loss cost coefficient. loss Main network loss, c curt P is the unit cost coefficient for the penalty of wind and solar power curtailment. wind,curt P pv,curt ω1 and ω2 are the curtailment of wind and solar power, respectively, and the weighting coefficients of grid loss and curtailment penalty costs are the main grid loss and the main grid loss and the main grid loss penalty costs, respectively.

[0052] This embodiment focuses directly on two key costs: grid loss and wind / solar curtailment. The objectives are clear and easily quantifiable. At the same time, it aligns with seasonal scheduling needs and can guide scheduling strategies to prioritize the consumption of new energy sources, thereby reducing the penalty for curtailment and making scheduling both economical and efficient.

[0053] In this embodiment, the process of solving the objective function specifically includes: An improved particle swarm optimization algorithm is adopted, and several output curves and load curves are randomly generated as scheduling schemes. The scheduling schemes are set as particles, and the main grid loss cost and the wind and solar curtailment penalty cost of the main-distribution-microgrid are calculated as the fitness value for each particle. The number of iterations, the initial learning factor and the initial velocity of the particles are set. Set the minimum fitness value as the population optimum, adjust the learning factor based on the number of iterations, update the generation rate of each particle based on the learning factor and the population optimum, and update the particles. The fitness value of each particle is calculated until the change is less than the change threshold or the iteration ends when the population reaches its optimal value after n consecutive iterations, thus obtaining the optimal output-load curve of the main-distribution-microgrid.

[0054] Specifically, first, the particle swarm size is set. Maximum number of iterations Inertia weight The initial value or range of variation ( and ), and individual learning factors Social learning factors This embodiment sets initial values ​​and adjustment strategies, taking... It is 100. It is 100; First, 100 random scheduling schemes are generated as initial particles, and the initial velocity v of the particles is also randomly generated. i and initial position x i Then, each initial particle is substituted into the objective function to calculate the fitness value f(x) of each particle. i ); Set the initial position of each particle to its individual historical best position. Find the particle with the best fitness among all particles and set its position as the global best position for the entire swarm. ; The learning factor is adjusted based on the number of iterations. The adjustment strategy is that c1 decreases linearly with the number of iterations, while c2 increases linearly. The formula is expressed as follows: Where t is the current iteration number. and Individual learning factors The maximum and minimum values, and For social learning factors The maximum and minimum values; through the above adjustment strategy, in the early stage of the search, the particles focus more on individual exploration (larger c1); as the search progresses, the particles gradually focus more on learning from the group's optimal experience (larger c2), promoting convergence; The generation rate of each particle is updated based on the learning factor and the population optimum, as expressed by the following formula: in Let be the velocity of particle i at iteration number t. Let be the position of particle i at iteration number t. To generate random numbers between 0 and 1, the speed value ranges from [-0.2, 0.2]. The particle position is then updated using the formula: x i (t+1)=x i (t)+v i (t+1), after the update, it is necessary to check whether the new position exceeds the solution space boundary. If it does, boundary processing is performed, and then the new position x of each particle is calculated. i The fitness value at (t+1) is determined if the fitness of the current new position is better than that of the particle itself. Then update with the new location. Update the fitness values ​​of all particles and compare them with the current fitness values. The fitness of the particles is compared, and if a particle provides a better solution, then the fitness is updated. ; If the maximum number of iterations T is reached max ,or If the improvement level is continuously less than a preset small threshold ϵ multiple times, the loop will exit and the result will be output.

[0055] This embodiment employs an improved particle swarm optimization algorithm to solve the objective function, covering all possible resource combinations across the main grid, distribution grid, and microgrid. By incorporating seasonal characteristics, it avoids missing optimal solutions suitable for seasonal scenarios, thus circumventing the local optima trap of traditional algorithms. The design of dynamically adjusting the learning factor balances the exploratory nature of the initial search with the convergence of later stages. Combined with the iteration termination condition, this improves both solution efficiency and result stability. Particle updates are based on the swarm optimum, guiding all scheduling schemes towards the global optimum. This ensures that the final optimal output-load curve satisfies both single-level optimization and cross-level coordination between the main grid, distribution grid, and microgrid, making the scheduling strategy both economical and aligned with seasonal characteristics and hierarchical requirements, thereby improving the feasibility of the scheduling strategy.

[0056] In this embodiment, the process of obtaining the collaborative scheduling strategy under different seasonal scenarios includes: The optimal output-load curve of the main-distribution-microgrid is divided into seasonal scenarios to obtain the output-load curves for different seasons; The optimal output-load matching degree for different seasons is determined based on the spatiotemporal matching degree of the output-load curves for different seasons. The corresponding scheduling strategy is queried from the preset scheduling strategy comparison table based on the optimal output-load matching degree, which serves as the collaborative scheduling strategy for different seasonal scenarios.

[0057] Specifically, the scheduling strategy comparison table is shown in Table 1: Table 1. Comparison of Scheduling Strategies The single-layer matching degree of the optimal output-load curve can be calculated, and the corresponding scheduling action can be selected from the reference table. Subsequently, the multi-layer matching degree and seasonal comprehensive matching degree can be calculated to evaluate the scheduling strategy.

[0058] This embodiment divides the optimal output-load curve by season, ensuring that the strategy conforms to the energy and load patterns of different seasons and avoiding the rigidity of using the same strategy throughout the year. The optimal matching degree is determined by the spatiotemporal matching degree, which is directly linked to a multi-dimensional matching degree index system, so that the strategy triggering has a clear quantitative basis rather than subjective judgment, thus improving the scheduling accuracy. The overall approach combines the optimal solution of the objective function with seasonal characteristics, so that the final strategy not only meets the core goal of cost reduction and energy consumption, but also has strong engineering feasibility and is perfectly adapted to the scheduling needs of the new power system.

[0059] In this embodiment, the resource constraints include thermal power unit constraints, wind power / photovoltaic unit constraints, energy storage system constraints, and grid operation safety constraints; The power grid operation constraints include load peak-valley difference constraints, minimum matching degree, and output boundary.

[0060] Specifically, the constraints for thermal power units are: Among them, P th,max P th,min P represents the upper and lower limits of the output of thermal power units. th (t) represents the real-time output of the thermal power unit, ΔP th,up ΔP th,down These are the limits for the uphill and downhill ramps of thermal power units.

[0061] The constraints for wind power / solar power units are: Among them, P w,max (t) Wind turbine output limit, P w (t) represents the real-time output of the wind turbine, P pv,max (t) represents the output limit of the photovoltaic unit, P pv (t) represents the real-time output of the photovoltaic unit, ΔP w,down ΔP is the downhill ramp limit for wind turbines. w,up ΔP is the uphill climbing limit for wind turbines. pv,down ΔP is the downhill ramp limit for photovoltaic units. pv,up The ramp-up limit for photovoltaic units.

[0062] The constraints of the energy storage system are: Where Emin ess and Emax ess are the lower and upper limits of the energy storage capacity, respectively; Pmax ch and Pmax dis are the charging and discharging power limits, respectively; and η ch η dis These represent the energy storage charge and discharge efficiency, u ch (t), u dis (t) is the charge / discharge state parameter, which is a 0-1 variable, and T is one charge / discharge cycle.

[0063] The power grid operation safety constraints are: Among them, P gen (t) represents the total power output of the system, P load (t) represents the total load in the system, P loss (t) represents the network loss value, P line,i (t) represents the line power, P line,i,max For line power limits, V max V min V represents the upper and lower limits of the node voltage. i (t) represents the node voltage value, P tie,main-dist (t) represents the power of the tie line between the main and distribution network coupling nodes, P tie,main-dist,max Power limits for main and distribution network coupling node tie lines, P tie,main-micro (t) represents the power of the main and microgrid coupled node tie lines, P tie,main-micro,max Power limits for main and microgrid coupled node tie lines.

[0064] In this embodiment, constraints such as thermal power units and energy storage are used to avoid equipment overload and overcharging / over-discharging, while grid safety constraints are used to prevent load fluctuations and matching imbalance risks. Constraints such as load peak-valley difference and minimum matching degree take into account seasonal characteristics and matching degree indicators, making the constraints more in line with the needs of seasonal and multi-level coordination, thereby ensuring that the dispatch strategy does not break through the safety boundaries of equipment and grid while reducing costs and absorbing the load, and improving the reliability of the solution implementation.

[0065] Example 2: Figure 5 As shown, this embodiment also provides a main-distribution-microgrid coordinated dispatching system based on seasonal characteristics, including a data acquisition module, a data processing module, a feature extraction module, a calculation module, and a coordinated dispatching module; The data acquisition module is used to collect operational data of the main-distribution-microgrid. The data processing module processes the power grid's operating data based on the typical day screening principle to obtain the typical daily operating curves of the main-distribution-microgrid. The feature extraction module extracts seasonal features from typical daily operating curves to obtain seasonal characteristic quantitative indicators of the main-distribution-microgrid, and generates grid operation constraints based on the seasonal characteristic quantitative indicators. Under the constraints of power grid operation, the calculation module constructs an objective function with the goal of maximizing the comprehensive benefits of coordinated scheduling of the main grid, distribution grid, and microgrid, and solves it to obtain the coordinated scheduling strategies under different seasonal scenarios. The coordinated scheduling module adjusts the power output allocation and energy storage operation status of each level of the power grid in the main-distribution-microgrid based on the coordinated scheduling strategy.

[0066] By constructing a corresponding system to integrate the collaborative scheduling method in this solution, human-computer interaction is achieved, improving the user experience.

[0067] As a further supplement to this embodiment, the following scenario will be used as an example to further illustrate this solution: like Figure 6 The diagram shows a power grid simulation structure for a specific scenario. The main grid uses an IEEE 14-node system, the distribution network uses an IEEE 33-node system, and the microgrid uses a 5-node system. Distribution network node 1 is connected to node 9 of the main grid's IEEE 14-node system, and microgrid node 1 is connected to node 10 of the main grid's IEEE 14-node system. The specific structure is as follows: Main grid: Thermal power access nodes 1, 2, 3, 6, 8; Wind power access node 6; Photovoltaic power access node 9; Load access nodes 12, 13; Energy storage access node 14; Distribution network: Wind power access node 16; Photovoltaic power access node 18; Load access node 24; Energy storage access nodes 17, 25; Microgrid: Wind power access node 2; Photovoltaic power access node 3; Load access node 4; Energy storage access node 3.

[0068] Under the scheduling scheme of this invention, the matching degree analysis results are shown in Table 2: Table 2. Seasonal Matching Analysis and Scheduling Strategy Triggering The table above shows the single-level matching degree, coordinated matching degree, and whether corresponding dispatch strategies are triggered for the main grid, distribution grid, and microgrid under four typical seasons: spring, summer, autumn, and winter. In summer, due to significant photovoltaic output, the microgrid matching degree is generally higher than 0.8, triggering the midday charging mode for energy storage. In winter, wind power dominates, and the microgrid matching degree is also high, triggering nighttime energy storage charging. In spring and autumn, due to generally lower renewable energy output, the matching degree is generally lower than 0.5, mainly triggering the tie-line reinforcement mode. This table reflects that the proposed dispatch strategy can effectively respond to the characteristics of renewable energy in different seasons and achieve dynamic adjustment.

[0069] Figure 7 This chart illustrates the changing trends in the matching degree of the main grid, distribution grid, and microgrid across the four seasons. The multi-level matching degree is highest in summer, peaking during midday when sunlight intensity is high. In winter, the multi-level matching degree is higher at night, consistent with wind power output characteristics. In spring and autumn, the overall matching degree is lower and fluctuates significantly, especially during morning and evening load peaks when the matching degree troughs occur. This chart verifies that the temporal coupling characteristics of renewable energy and load exhibit significant seasonal differences, highlighting the necessity of seasonal regulation.

[0070] Figure 8The total dispatch cost was compared across the four seasons. Summer saw the lowest total cost due to abundant solar resources and a low curtailment rate. Winter, while characterized by high wind power output, also saw higher costs due to greater load demand and frequent mobilization of energy storage. Spring and autumn, with weakened wind and solar resources and lower matching rates, required frequent use of tie lines for reinforcement and adjustment of energy storage, leading to significantly higher costs. This indicates a strong correlation between dispatch cost and the availability and timing matching of renewable energy sources.

[0071] Figure 9 The data shows the matching degree distribution of each level in different seasons. The main network has the highest matching degree stability, followed by the distribution network, while the microgrid has the most discrete matching degree distribution due to the large fluctuations in distributed resources.

[0072] Figure 10 , Figure 11 and Figure 12 The scheduling scheme of this invention is compared with the ordinary scheduling scheme. Figure 10 The results show that the total annual cost of this invention is reduced by approximately 1.71% compared to the conventional scheduling scheme, with particularly significant cost reductions during the summer. Figure 11 The results show that, due to optimized energy storage and tie-line scheduling, network losses decreased by approximately 0.53% compared to conventional scheduling schemes. Figure 12 Compared to conventional scheduling schemes, the wind and solar curtailment rate decreased by approximately 2.66%, highlighting the effectiveness of the proposed strategy in improving renewable energy consumption. The comparative results validate the superiority of the seasonal characteristic analysis and dynamic scheduling strategy of this invention.

[0073] As can be seen from the above embodiments, it has at least the following substantial effects: (1) By screening typical days, this invention condenses the 8760 hours of the whole year into typical day curves for each season. It does not require solving the massive data of the whole year hour by hour, but only optimizes the typical day scenario, thereby greatly reducing the amount of calculation of the objective function in the subsequent solution, avoiding the solution delay caused by data overload, and enabling the scheduling scheme to respond quickly to seasonal changes. (2) This invention extracts seasonal features and transforms them into power grid operation constraints, thereby enabling the constraints to dynamically adapt to seasonal changes. This also allows multi-level power grids to be interconnected, strengthening the coordination of each level of power grid. As a result, the final coordinated scheduling strategy optimizes the operation of a single level while ensuring overall coordination. (3) By aiming at maximizing comprehensive benefits, the present invention obtains seasonal scheduling strategies under constraints, which not only maximizes the efficiency of resource utilization in different seasons and reduces ineffective costs, but also prioritizes the optimization of high-potential new energy sources, significantly reducing the curtailment rate of wind and solar power. At the same time, it can respond to different scheduling strategies according to different seasons, making collaborative scheduling more flexible and changeable.

[0074] The specific embodiments described above are preferred embodiments of the main-distribution-microgrid coordinated dispatch method and system based on seasonal characteristics of the present invention, and are not intended to limit the specific scope of the present invention. The scope of the present invention includes but is not limited to the specific embodiments described above. All equivalent changes made in accordance with the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A coordinated dispatch method for primary-distribution-microgrid based on seasonal characteristics, characterized in that: Includes the following steps: S1. Based on the typical day screening principle, the power grid operation data is processed to obtain the typical daily operation curves of the main-distribution-microgrid; S2. Extract seasonal characteristics from typical daily operating curves to obtain seasonal characteristic quantitative indicators of the main-distribution-microgrid, and determine grid operation constraints based on seasonal characteristic quantitative indicators; S3. Under the constraints of power grid operation and preset resource constraints, construct an objective function with the goal of maximizing the comprehensive benefits of coordinated dispatch of main-distribution-microgrid; solve the objective function to obtain the coordinated dispatch strategy under different seasonal scenarios.

2. The main-distribution-microgrid coordinated dispatch method based on seasonal characteristics according to claim 1, characterized in that: In S1, the power grid operation data is processed based on the typical day screening principle to obtain the typical daily operation curves of the main-distribution-microgrid, including the following steps: The annual renewable energy output data and load power data of the main-distribution-microgrid are collected and normalized to obtain the operation dataset; The number of clusters is determined to be k based on the seasonal quantity. A data point is randomly selected from the running dataset as the initial cluster center. The shortest distance between all other data points and the initial cluster center is calculated to obtain the probability distribution of all other data points being selected as cluster centers. The top k-1 data points with the highest probability are taken as target cluster centers. Clustering is performed on the data in the running dataset based on the initial cluster centers and target cluster centers to obtain clustered data. The clustered data is then divided into seasonal scenarios and recorded based on time series to obtain the typical daily operating curves of the main-distribution-microgrid.

3. The main-distribution-microgrid coordinated dispatch method based on seasonal characteristics according to claim 2, characterized in that: The clustered data is divided into seasonal scenarios and recorded based on time series to obtain typical daily operation curves for the main-distribution-microgrid system, including the following steps: Clustered data with photovoltaic output greater than or equal to the output threshold and the proportion of output during midday hours greater than or equal to the proportion threshold are defined as summer scenario data; Clustered data with wind power output greater than or equal to the output threshold and the proportion of output during nighttime hours greater than or equal to the proportion threshold are defined as winter scenario data; The remaining clustering data will be classified as spring and autumn scene data; The summer, winter, and spring / autumn scenario data were recorded based on time series to obtain the typical daily operation curves of the main-distribution-microgrid in summer, winter, and spring / autumn respectively.

4. The main-distribution-microgrid coordinated dispatch method based on seasonal characteristics according to claim 3, characterized in that: In S2, seasonal characteristic quantitative indicators of the main-distribution-microgrid are obtained by extracting seasonal features from typical daily operating curves, including the following steps: The correction coefficient for new energy output is determined based on the average daily output of new energy in different seasons and the average daily output of new energy throughout the year in the typical daily operation curve. The load peak-valley difference threshold is determined based on the load peak-valley difference and peak-valley ratio in different seasons in the typical daily operation curve; The output-load matching degree is determined based on the spatiotemporal matching degree of new energy output and load in different seasons in typical daily operation curves; The new energy output correction coefficient, the load peak-valley difference threshold, and the output-load matching degree are used as quantitative indicators of the seasonal characteristics of the main-distribution-microgrid.

5. The main-distribution-microgrid coordinated dispatch method based on seasonal characteristics according to claim 4, characterized in that: The output-load matching degree includes single-level matching degree, multi-level matching degree, and seasonal comprehensive matching degree; The single-level matching degree is used to quantify the renewable energy absorption capacity and load matching effect of a certain level in the main-distribution-microgrid in different seasons; The multi-level matching degree is used to quantify the cross-level energy utilization rate between various levels in the main-distribution-microgrid; The seasonal comprehensive matching degree is used to quantify the annual renewable energy absorption capacity and load matching effect in the main-distribution-microgrid.

6. The main-distribution-microgrid coordinated dispatch method based on seasonal characteristics according to claim 1, characterized in that: In S3, the objective function aims to minimize the sum of the main grid loss cost and the wind and solar curtailment penalty cost of the main-distribution-microgrid. The main grid loss cost is the product of the unit grid loss cost coefficient and the main grid loss; the wind and solar curtailment penalty cost is the product of the sum of the wind and solar curtailment of the main-distribution-microgrid and the unit wind and solar curtailment penalty cost coefficient.

7. The main-distribution-microgrid coordinated dispatch method based on seasonal characteristics according to claim 1, characterized in that: In S3, the process of solving the objective function specifically includes: An improved particle swarm optimization algorithm is adopted, and several output curves and load curves are randomly generated as scheduling schemes. The scheduling schemes are set as particles, and the main grid loss cost and the wind and solar curtailment penalty cost of the main-distribution-microgrid are calculated as the fitness value for each particle. The number of iterations, the initial learning factor and the initial velocity of the particles are set. Set the minimum fitness value as the population optimum, adjust the learning factor based on the number of iterations, update the generation rate of each particle based on the learning factor and the population optimum, and update the particles. The fitness value of each particle is calculated until the change is less than the change threshold or the iteration ends when the population reaches its optimal value after n consecutive iterations, thus obtaining the optimal output-load curve of the main-distribution-microgrid.

8. The main-distribution-microgrid coordinated dispatch method based on seasonal characteristics according to claim 7, characterized in that: In S3, the process of obtaining the collaborative scheduling strategy for different seasonal scenarios includes: The optimal output-load curve of the main-distribution-microgrid is divided into seasonal scenarios to obtain the output-load curves for different seasons; The optimal output-load matching degree for different seasons is determined based on the spatiotemporal matching degree of the output-load curves for different seasons. The corresponding scheduling strategy is queried from the preset scheduling strategy comparison table based on the optimal output-load matching degree, which serves as the collaborative scheduling strategy for different seasonal scenarios.

9. The main-distribution-microgrid coordinated dispatch method based on seasonal characteristics according to claim 1, characterized in that: The resource constraints include constraints on thermal power units, constraints on wind / photovoltaic units, constraints on energy storage systems, and constraints on grid operation safety. The power grid operation constraints include load peak-valley difference constraints, minimum matching degree, and output boundary.

10. A primary-distribution-microgrid coordinated dispatch system based on seasonal characteristics, applicable to the primary-distribution-microgrid coordinated dispatch method based on seasonal characteristics as described in any one of claims 1-9, characterized in that: It includes a data acquisition module, a data processing module, a feature extraction module, a computing module, and a collaborative scheduling module; The data acquisition module is used to collect operational data of the main-distribution-microgrid. The data processing module processes the power grid's operating data based on the typical day screening principle to obtain the typical daily operating curves of the main-distribution-microgrid. The feature extraction module extracts seasonal features from typical daily operating curves to obtain seasonal characteristic quantitative indicators of the main-distribution-microgrid, and generates grid operation constraints based on the seasonal characteristic quantitative indicators. Under the constraints of power grid operation, the calculation module constructs an objective function with the goal of maximizing the comprehensive benefits of coordinated scheduling of the main grid, distribution grid, and microgrid, and solves it to obtain the coordinated scheduling strategies under different seasonal scenarios. The coordinated scheduling module adjusts the power output allocation and energy storage operation status of each level of the power grid in the main-distribution-microgrid based on the coordinated scheduling strategy.

Images