Renewable energy power system supply-transmission-demand-storage collaborative optimization configuration method

Abstract

The invention provides a renewable energy power system supply-transmission-demand-storage collaborative optimization configuration method, and relates to the technical field of smart energy, and the method comprises the steps: obtaining basic data of a plurality of provinces and cities, including renewable energy resource data and auxiliary data; establishing a collaboration degree calculation model, respectively calculating the collaboration degree of each subsystem, calculating the overall collaboration degree based on the collaboration degree of each subsystem, and constructing a subsystem collaboration strength matrix to quantify the coupling relationship among the subsystems; establishing a multi-sub-population collaborative optimization architecture, and respectively generating an initial solution set of each sub-population by adopting a collaborative degree-oriented initialization strategy; adopting an improved sparrow optimization algorithm to carry out parallel optimization on each sub-population, and obtaining an optimal configuration result when a convergence condition is met; and outputting a supply-transmission-demand-storage collaborative optimization configuration scheme of the renewable energy power system according to the optimal configuration result. The overall operation efficiency and economical efficiency of the system can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart energy, and particularly relates to a renewable energy power system supply-distribution-demand-storage collaborative optimization configuration method. BACKGROUND

[0002] With the accelerated process of global energy transformation and carbon emission reduction, renewable energy has become the dominant direction of future energy development. However, the distribution of renewable energy resources has obvious regional characteristics. For example, wind energy resources are mainly concentrated in the "three north" regions, and there is a spatial mismatch between solar energy resource-rich areas and power load centers. At the same time, renewable energy generation has the characteristics of intermittency, volatility and uncertainty, and the renewable energy consumption capacity in a single region is limited. Therefore, it is urgent to realize the large-scale optimization configuration of renewable energy through cross-regional optimization scheduling. Cross-regional scheduling can not only smooth the fluctuation of renewable energy output and improve the overall power supply reliability of the system, but also can fully play the complementary advantages of resources in different regions and maximize the utilization efficiency of renewable energy. It has important technical significance and practical value for building a clean, low-carbon, safe and efficient modern energy system.

[0003] In the field of renewable energy optimization scheduling technology, intelligent optimization algorithms have become the main means to solve large-scale complex optimization problems. Traditional heuristic algorithms such as genetic algorithm and particle swarm optimization algorithm have been widely used in microgrid and distributed energy system optimization. In recent years, deep reinforcement learning algorithms have shown significant advantages in real-time optimization scheduling of integrated energy systems due to their ability to solve continuous sequence decision-making problems. However, existing algorithms have low search efficiency, are prone to local optimization, and have poor adaptability to high-dimensional problems when dealing with large-scale, multi-constrained complex power system collaborative optimization problems, which cannot meet the technical needs of new power systems for efficient collaborative optimization.

[0004] The invention patent with the Chinese publication number CN117151394A discloses a cross-regional hydrogen-electric coupling system optimization scheduling method based on supply-transport-demand coordination. The technology realizes the matching of hydrogen and electricity demand in remote energy consumption areas by constructing an electrolytic hydrogen production special zone in a renewable energy-rich area and adopting hydrogen energy cross-regional transportation or electric energy cross-regional transportation. The method establishes a multi-objective optimization scheduling model with the lowest total cost and the smallest carbon emission as the target, and uses the DDPG deep reinforcement learning algorithm to realize the on-demand allocation of hydrogen energy and electric energy. However, the technical scheme has the following problems: first, the DDPG algorithm used by the method is suitable for decision-making problems in continuous action space, but it has poor convergence stability and is sensitive to initialization when dealing with multi-subsystem coupling constraints and large-scale combinatorial optimization problems; in addition, the coordination mechanism of the method mainly reflects the hydrogen-electric conversion level, lacks quantitative evaluation and dynamic adjustment mechanism for the coordination degree of each subsystem in the power system, and it is difficult to realize real deep coordination optimization. SUMMARY

[0005] Therefore, the present application provides a renewable energy power system supply-transport-demand-storage collaborative optimization configuration method, which quantifies the coupling relationship between each subsystem by establishing a coordination degree calculation model, realizes parallel collaborative optimization of each subsystem by using an improved multi-subpopulation sparrow optimization algorithm, takes maximizing coordination degree and minimizing total cost as double objectives, and finally obtains a whole system collaborative optimization configuration scheme including renewable energy installed capacity, energy storage device configuration, power transmission scheme and demand response strategy of each province and city, thereby improving the overall operation efficiency and economy of the system.

[0006] The technical scheme of the present application is implemented as follows: The present application provides a renewable energy power system supply-transport-demand-storage collaborative optimization configuration method, comprising: S1, obtaining the basic data of a plurality of provinces and cities, including renewable energy resource data and auxiliary data, standardizing the renewable energy resource data, defining the decision variables of the four subsystems of supply, transport, demand and energy storage, and obtaining a unified resource evaluation basis and decision variable system; S2, establishing a coordination degree calculation model, calculating the coordination degree of each subsystem respectively, calculating the overall coordination degree based on the coordination degree of each subsystem, and constructing a subsystem coordination strength matrix to quantify the coupling relationship between each subsystem; S3, establishing a multi-subpopulation collaborative optimization architecture, dividing the population into four subpopulations according to the four subsystems of supply, transport, demand and energy storage, and using a coordination degree-oriented initialization strategy to generate an initial solution set for each subpopulation; S4, the improved sparrow optimization algorithm is used for parallel optimization of each sub-population, and the synergy degree and total cost are used as double targets for fitness evaluation, and the synergy optimization between sub-populations is realized through adaptive parameter adjustment and information exchange mechanism, and when the convergence condition is reached, the optimal configuration result is obtained; S5, according to the optimal configuration result, the renewable energy installed capacity of each province and city, the energy storage device configuration, the power transmission scheme and the demand response strategy are determined, and the supply-transmission-demand-storage collaborative optimization configuration scheme of the renewable energy power system is output.

[0007] Preferably, step S2 comprises: S21, the synergy degree calculation formula of the supply subsystem, the transmission subsystem, the demand subsystem and the energy storage subsystem is established respectively, wherein the synergy degrees of the transmission subsystem, the demand subsystem and the energy storage subsystem are obtained by weighted summation based on the influence indexes and weights of the subsystems, and the supply subsystem needs to deduct the uncertainty coefficient of renewable energy output additionally, and the uncertainty coefficient is quantified by calculating the variance of renewable energy power generation of each province and city; S22, the synergy degrees of each subsystem are calculated by the geometric mean method to obtain the overall synergy degree; S23, the internal uncertainty degree is evaluated by calculating the entropy value of each subsystem, the weight coefficient of each subsystem is calculated based on the entropy value, and the 4*4 symmetric synergy strength matrix is constructed combining the Pearson correlation coefficient and the weight coefficient between the subsystems, and the mutual influence degree between the subsystems is quantified.

[0008] Preferably, step S3 comprises: S31, a multi-sub-population collaborative optimization architecture is established, wherein the supply sub-population is responsible for optimizing the renewable energy installed capacity of each province and city, including the installed capacity decision variable of each type of renewable energy of each province and city; the transmission sub-population is responsible for optimizing the transmission capacity of the power transmission line; the demand sub-population is responsible for optimizing the power demand guarantee level of each province and city; and the energy storage sub-population is responsible for optimizing the energy storage installed capacity of each province and city; S32, the system characteristic factor of each subsystem is calculated based on the auxiliary data of each province and city, wherein the system characteristic factor of the supply sub-population is the resource endowment factor, the system characteristic factor of the transmission sub-population is the transmission demand coefficient, the system characteristic factor of the demand sub-population is the load characteristic coefficient, and the system characteristic factor of the energy storage sub-population is the configuration strategy coefficient; S33, based on the synergy strength matrix and the system characteristic factor of each subsystem, the initial solution set of each sub-population is generated respectively.

[0009] Preferably, step S4 comprises: S41, a double-target fitness function is constructed, considering both the maximization of synergy degree and the minimization of cost; S42. Calculate the initial fitness value of each individual in each subpopulation, assign roles based on the fitness ranking, and determine the number of discoverers, followers, and sentinels; S43, based on adaptive step size parameters Calculate dynamic warning value: , Where, Indicates the The warning value in the iteration, Indicates the initial warning value, Indicates the The adjustment coefficient of the subpopulation, Indicates the The subpopulation in Adaptive step size parameter for iterations; S44. The discoverer uses different search modes according to the warning value. A global search strategy is used when A local search strategy is used, where Represents the safety threshold; followers adopt different following methods according to individual rankings. and previous individuals follow the discoverer, ranked The individuals thereafter conduct random searches. Indicates the The size of the subpopulation; the vigilant adopts a conservative or aggressive position update strategy according to the relative level of fitness; S45. A hierarchical constraint processing mechanism is used to process constraints on individuals after position updates, including boundary layer constraint processing to ensure that decision variables are within the allowable range, balance layer constraint processing to process power system supply and demand balance constraints, and coordination layer constraint processing to process energy storage configuration ratio constraints and transmission capacity constraints; S46. Dynamically calculate the information exchange frequency based on the difference in convergence state of each subpopulation. When the exchange condition is met, combine the current best individuals of each subpopulation to form a candidate global solution, calculate its overall fitness, and if it is better than the current global optimal solution, update the global optimal solution and feed back the updated information to each subpopulation. S47. Check whether the convergence conditions are met. If the maximum number of iterations is reached or the fitness improvement is less than the set threshold, output the optimal configuration result; if not, set t=t+1 and return to S43 to continue iterative optimization.

[0010] Preferably, the adaptive step size parameter Taking into account the synergy feedback, convergence state and iterative process, the calculation formula is: , in, represents the initial step size parameter, Indicates the The subpopulation in The coordination degree of the iteration, Indicates the The maximum degree of cooperation among subpopulations, represents the maximum number of iterations, Indicates the The subpopulation in The convergence status evaluation value of iteration.

[0011] Preferably, step S46 includes: The information exchange frequency is dynamically calculated based on the difference in the convergence status of each subpopulation. When the convergence degree of each subpopulation is greatly different, the exchange frequency is increased to promote cooperation. When the convergence degree is similar, the exchange frequency is reduced to maintain diversity. The information exchange frequency calculation formula is: , in, represents the initial exchange frequency, Indicates the The subpopulation in The convergence status evaluation value of the iteration, represents a small constant that avoids the denominator being zero; When the information exchange conditions are met, the best individuals of each subpopulation are combined to form a candidate solution and its overall fitness is calculated. If the candidate solution is better than the current global optimal solution, the global optimal solution is updated and fed back to each subpopulation.

[0012] Preferably, the hierarchical constraint processing mechanism includes: boundary layer constraint processing to truncate decision variables that exceed the variable boundary, so that all decision variables are within the allowable range; balance layer constraint processing of the supply and demand balance constraint of the power system, based on the power demand data in the auxiliary data, when the power balance constraint is detected to be violated, the degree of violation is calculated and repaired through energy storage charging and discharging power adjustment and inter-provincial power exchange; collaborative layer constraint processing of energy storage configuration ratio constraint and transmission capacity constraint, so that the ratio of energy storage capacity to renewable energy installed capacity is within a reasonable range, and the fitness value of the individual that violates the constraint is adjusted through the penalty function.

[0013] Preferably, the calculation formula of the dual-objective fitness function is as follows: , , Where, Indicates the In the subpopulation individual; Represents an individual The fitness value of Represents an individual overall synergy degree of the corresponding configuration scheme; theoretical maximum value of the synergy degree, used for normalization; individual total cost of the corresponding configuration scheme; maximum value of the cost, used for normalization; and respectively represent the weight coefficients of the synergy degree and the cost, satisfying ; extracted from the position vector of the individual , the a-th subsystem corresponds to the decision variable vector; represent the cost function of the a-th subsystem.

[0014] Preferably, the auxiliary data in step S1 includes: power demand data, including historical power load data, load growth trend data, peak-valley load difference coefficient, load prediction data and demand response potential data of each province and city, data time granularity covering annual, monthly, daily and hourly data; power transmission network data, including the existing power transmission line rated transmission capacity, line impedance parameters, line length, power transmission loss coefficient between each province and city, and construction planning data and power transmission capacity expansion potential data of cross-regional power transmission channel; energy storage related data, including the installed capacity of existing energy storage facilities in each province and city, technical type distribution, charging and discharging efficiency parameters, and investment cost and operation and maintenance cost data of energy storage equipment; economic development data, energy consumption structure data and policy support intensity data of each province and city.

[0015] Preferably, step S5 includes: extracting specific configuration parameters of each subsystem from the global optimal solution, including renewable energy installed capacity of each type of each province and city of the supply subsystem, transmission capacity configuration of each power transmission line of the transmission subsystem, power demand guarantee level of each province and city of the demand subsystem, and energy storage installed capacity of each province and city of the energy storage subsystem; calculating the key performance indicators of the configuration scheme, including the overall synergy degree of the system, the total investment cost, the annual operation cost, the renewable energy utilization rate and the power supply reliability; and generating a complete supply-transmission-demand-storage collaborative optimization configuration report.

[0016] The present application has the following beneficial effects compared with the prior art: (1) The present application solves the problems of traditional single population optimization algorithm in dealing with high-dimensional complex search space, such as falling into local optimum and low search efficiency, by constructing a renewable energy power system supply-delivery-demand-storage collaborative optimization configuration method of multi-subpopulation coevolution. The method decomposes the single population into four specialized subpopulations responsible for the independent optimization of the four subsystems of supply, delivery, demand and energy storage, realizes global collaboration through a collaborative degree guided information exchange mechanism, improves search efficiency while maintaining search diversity, realizes the dual objective optimization of maximum collaborative degree and minimum total cost, and provides a technically feasible and economically reasonable configuration scheme for large-scale renewable energy power system; (2) The coupling relationship between subsystems is quantified by constructing a subsystem collaboration strength matrix, and an intelligent initialization strategy is designed combined with the resource endowment characteristics of each province and city, which solves the problems of poor initial population quality and slow convergence speed caused by completely random initialization of standard sparrow optimization algorithm. The strategy makes the initial population have better search starting point and directionality, improves the convergence efficiency of the algorithm and the quality of the solution; (3) By decomposing the traditional single population into four specialized subpopulations, each subpopulation evolves independently and realizes global collaboration through information exchange, effectively solving the problem of dimension disaster in high-dimensional complex search space. This mechanism not only maintains the professional characteristics of each subsystem optimization, but also realizes global collaboration through dynamic information exchange mechanism, avoiding the algorithm falling into local optimal solution and improving the global search ability; (4) Based on the feedback of collaborative degree and the convergence state of each subpopulation, the search step and the early warning value are dynamically adjusted, which solves the problem that the existing algorithm based on fixed parameter setting cannot adapt to different optimization stages and subsystem characteristics. This mechanism enables the algorithm to adaptively adjust the search strategy according to the real-time optimization state, maintain a larger search step in the early optimization stage to enhance the global search ability, and reduce the search step in the later optimization stage to improve the local search precision, improving the adaptability and convergence precision of the algorithm; (5) By establishing a three-layer constraint processing architecture of boundary layer, balance layer and collaboration layer, the complex coupled constraints in the supply-delivery-demand-storage system are effectively handled, such as power balance constraint and energy storage charging and discharging constraint. This mechanism combines hierarchical repair and penalty function in a way to ensure the feasibility of the solution while maintaining the search efficiency of the algorithm, avoiding the performance degradation of the algorithm caused by constraint violation in traditional constraint handling methods. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creating any inventive labor.

[0018] Figure 1 A method flowchart of the present application; Figure 2 An algorithm flowchart of the present application. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be clearly and completely described in combination with the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, but not all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0020] As Figure 1 shown, the present application provides a renewable energy power system supply-transport-demand-storage collaborative optimization configuration method, comprising: S1, obtaining the basic data of multiple provinces and cities, including renewable energy resource data and auxiliary data, performing standardization processing on the renewable energy resource data, defining the decision variables of the four subsystems of supply, transport, demand and energy storage, and obtaining a unified resource evaluation basis and decision variable system; S2, establishing a collaboration degree calculation model, calculating the collaboration degree of each subsystem respectively, calculating the overall collaboration degree based on the collaboration degree of each subsystem, and constructing a subsystem collaboration strength matrix to quantify the coupling relationship between each subsystem; S3, establishing a multi-subpopulation collaborative optimization architecture, dividing the population into four subpopulations according to the four subsystems of supply, transport, demand and energy storage, and using a collaboration degree oriented initialization strategy to generate an initial solution set for each subpopulation; S4, using an improved sparrow optimization algorithm to perform parallel optimization on each subpopulation, using collaboration degree and total cost as double objectives to evaluate fitness, and performing collaborative optimization between subpopulations through adaptive parameter adjustment and information exchange mechanism, and obtaining the optimal configuration result when the convergence condition is reached; S5, determining the renewable energy installed capacity, energy storage device configuration, power transport scheme and demand response strategy of each province and city according to the optimal configuration result, and outputting the renewable energy power system supply-transport-demand-storage collaborative optimization configuration scheme.

[0021] Specifically, in an embodiment of the present application, step S1 comprises: First, obtain the renewable energy resource data of multiple provinces and cities in an economic region, specifically including the basic resource endowment information of six main renewable energy types.

[0022] Wind energy resource data collection of different height layer of wind speed data, wind direction distribution data and wind energy density distribution information, data sources include the meteorological observation station data and satellite remote sensing data of the national meteorological bureau; Solar energy resource data collection of each province and city solar radiation intensity data, sunshine duration data and environmental temperature data, wherein the environmental temperature data is used for temperature correction calculation of photovoltaic power generation efficiency; Water energy resource data collection of each province and city river runoff data, water level change data and water energy theoretical reserves data; Biomass resource data collection of each province and city of agricultural and forestry waste production data, urban organic waste production data and energy crop planting area data; Geothermal resource data collection of each province and city of geothermal gradient distribution data, geothermal fluid temperature data and geothermal reservoir depth information; Marine energy resource data collection of coastal provinces and cities of tidal energy, wave energy and ocean current energy resource distribution data.

[0023] At the same time, the auxiliary data required for the operation of the power system is obtained synchronously. The power demand data collection of each province and city historical power load data, load growth trend data, peak valley load difference coefficient, load prediction data and demand response potential data, data time granularity covers annual, monthly, daily and hourly level data, to accurately reflect the electricity consumption characteristics and demand change law of each province and city. Transmission network data collection of existing transmission line parameters information between each province and city, including line rated transmission capacity, line impedance parameters, line length and transmission loss coefficient, at the same time, the construction planning data and transmission capacity expansion potential data of cross regional transmission channel are obtained. Energy storage related data collection of existing energy storage facilities in each province and city, including installed capacity, technical type distribution, charging and discharging efficiency parameters and investment cost and operation and maintenance cost data of energy storage equipment. In addition, the economic development data, energy consumption structure data and policy support intensity data of each province and city are also needed, which are used as constraint conditions and evaluation indexes of collaborative optimization configuration.

[0024] After obtaining the basic data, the renewable energy resource data needs to be standardized to eliminate the dimensional difference between different resource data: , Among them, Indicates the index of province and city, N is the number of selected provinces and cities, Indicates the index of renewable energy resource type, Corresponding to wind energy, solar energy, water energy, biomass energy, geothermal energy and marine energy, Indicates the original value of the first The first Kind of resource in the first Province and city,

[0025] For the processing of auxiliary data, the power demand data is also processed in a standardized manner, taking the per capita electricity consumption and the unit GDP electricity consumption of each province and city as the standardization benchmarks to ensure the comparability of the demand data. The transmission network data is classified and standardized according to the transmission distance and voltage level, and the transmission lines of different voltage levels adopt corresponding standardized benchmarks. The energy storage related data is classified and processed according to the type of energy storage technology and the application scenario, and different technical routes such as electrochemical energy storage, pumped storage and compressed air storage adopt corresponding technical and economic parameters for standardization.

[0026] On the basis of data acquisition and standardization, the decision variable system of the supply, transmission, demand and energy storage four subsystems is defined. The total decision variable vector is defined, and the decision variables of the supply-transmission-demand-storage four subsystems are uniformly represented as The supply subsystem variable includes the installed capacity of 6 kinds of renewable energy sources of multiple provinces and cities, with a total dimension of 186; the transmission subsystem variable includes the transmission capacity of each transmission line; the demand subsystem variable includes the power demand guarantee level of each province and city; and the energy storage subsystem variable includes the energy storage installed capacity of each province and city.

[0027] Specifically, in an embodiment of the present application, step S2 includes: S21, respectively establishing the coordination degree calculation formula of the supply subsystem, the transmission subsystem, the demand subsystem and the energy storage subsystem, wherein the coordination degrees of the transmission subsystem, the demand subsystem and the energy storage subsystem are obtained by weighted summation based on the influence indexes and weights of each subsystem, and the supply subsystem needs to additionally deduct the uncertainty coefficient of renewable energy output, and the uncertainty coefficient is quantified by calculating the variance of renewable energy power generation of each province and city; S22, calculating the coordination degrees of each subsystem by the geometric mean method to obtain the overall coordination degree; S23, evaluating the internal uncertainty degree by calculating the entropy value of each subsystem, calculating the weight coefficient of each subsystem based on the entropy value, and constructing a 4x4 symmetric coordination strength matrix combining the Pearson correlation coefficient and the weight coefficient between the subsystems to quantify the mutual influence degree between each subsystem.

[0028] In a specific example, the implementation process of step S2 is as follows: The coordination degree calculation of each subsystem reflects the coordination degree between the elements in the subsystem and the contribution degree of the subsystem to the overall coordination.

[0029] The coordination degree calculation formula of each subsystem is uniformly represented as: , wherein, represents a subsystem index, , represents the number of indicators of the th subsystem, represents the weight of the th indicator of the th subsystem, represents the normalized indicator value. Specifically, the influence indicators of each subsystem include: for the supply subsystem, indicators such as installed capacity matching degree, output stability, and resource utilization; for the transmission subsystem, indicators such as transmission capacity utilization, network connectivity, and uniformity of power flow distribution; for the demand subsystem, indicators such as load forecasting accuracy, peak-valley difference coefficient, and demand response capability; and for the energy storage subsystem, indicators such as capacity configuration rationality, charging and discharging efficiency, and frequency modulation and peak shaving capability. are determined according to the analytic hierarchy process and expert scoring method, while satisfying the normalization constraint condition .

[0030] Specifically, the calculation of the coordination degree of the supply subsystem has particularity. In addition to considering the conventional influence indicators such as installed capacity matching degree, output stability, and resource utilization, the influence of the renewable energy output uncertainty coefficient needs to be additionally deducted. This is because renewable energy has the characteristics of intermittency and volatility, and the uncertainty of its output will have an adverse impact on the coordinated operation of the system, so it must be considered when calculating the coordination degree of the supply subsystem. The calculation formula of the coordination degree of the supply subsystem is: , wherein, represents the renewable energy output uncertainty coefficient, and the calculation formula is: , wherein, represents the power generation of the th kind of renewable energy in the th province or city, represents the variance function. The fluctuation degree of the renewable energy output of each province or city is reflected by variance calculation, and the greater the variance, the higher the uncertainty and the more significant the negative impact on the coordination degree.

[0031] The calculation of the overall coordination degree uses the geometric mean method, and the calculation formula is: , wherein, represents the overall coordination degree based on the decision variable , The geometric mean method is chosen based on two considerations: first, to ensure that the importance of the synergy of each subsystem is equal, so that the advantages of one subsystem do not overcompensate for the disadvantages of others; and second, to promote a balanced improvement in the synergy of each subsystem, guiding the optimization algorithm to take into account the coordinated development of each subsystem while improving the overall synergy.

[0032] The present invention also constructs a subsystem synergy strength matrix to quantify the coupling relationship between the subsystems.

[0033] First, the entropy value of each subsystem is calculated to evaluate the degree of uncertainty within the subsystem. The entropy value reflects the degree of order of the information within the subsystem. The smaller the entropy value, the more ordered the system is and the lower the uncertainty is: , in, Indicates the subsystem index, , Indicates the Subsystem In the process of entropy calculation, the values ​​of each indicator are first normalized, and then calculated according to the definition formula of information entropy. The normalization coefficient Make sure the entropy value is in the range [0,1].

[0034] Calculate the weight coefficient of each subsystem based on the entropy value: , in, Represents the subsystem index. This calculation method is based on the basic principle of information entropy. Subsystems with smaller entropy values ​​have higher information content and lower uncertainty, and therefore should be given greater weight in overall coordination.

[0035] The construction of the synergy strength matrix combines the Pearson correlation coefficient and weight coefficient between subsystems. The calculation formula of the synergy strength matrix elements is: , in, Indicates the subsystem and the The Pearson correlation coefficient is calculated based on the historical statistical information of the indicator data of each subsystem, and the correlation strength is determined by analyzing the covariance and standard deviation between the indicators of different subsystems.

[0036] The final constructed synergy strength matrix Is a 4×4 symmetric matrix, the matrix elements The value range of the is [0, 1], and the larger the value, the stronger the influence of the a-th subsystem on the b-th subsystem. The symmetry of the matrix reflects the reciprocity of the mutual influence between the subsystems, that is, the influence strength of the a-th subsystem on the b-th subsystem is equal to the influence strength of the b-th subsystem on the a-th subsystem.

[0037] The traditional sparrow optimization algorithm generates an initial population by using a completely random initialization method, ignores the coupling relationship and constraint characteristics between the subsystems of the power system, and leads to poor initial population quality and low search efficiency. At the same time, a single population optimization mechanism is prone to fall into a local optimal solution when facing a high-dimensional complex search space, and it is difficult to fully explore different regions of the solution space. In view of these problems, the present application proposes a multi-subpopulation collaborative optimization architecture and a synergy degree oriented intelligent initialization strategy, which improves the global search ability and convergence efficiency of the algorithm through decomposition of the optimization problem and specialized search strategy.

[0038] Specifically, in an embodiment of the present application, step S3 comprises: S31, a multi-subpopulation collaborative optimization architecture is established, wherein the supply subpopulation is responsible for optimizing the renewable energy installed capacity of each province and city, including the installed capacity decision variables of each type of renewable energy in each province and city; the transmission subpopulation is responsible for optimizing the transmission capacity of the power transmission line; the demand subpopulation is responsible for optimizing the power demand guarantee level of each province and city; and the energy storage subpopulation is responsible for optimizing the energy storage installed capacity of each province and city; S32, system characteristic factors of each subsystem are calculated based on the auxiliary data of each province and city, wherein the system characteristic factor of the supply subpopulation is a resource endowment factor, the system characteristic factor of the transmission subpopulation is a transmission demand coefficient, the system characteristic factor of the demand subpopulation is a load characteristic coefficient, and the system characteristic factor of the energy storage subpopulation is a configuration strategy coefficient; S33, initial solution sets of each subpopulation are generated based on the synergy strength matrix and the system characteristic factors of each subsystem.

[0039] In the embodiment, a single population is decomposed into four specialized subpopulations, each of which is responsible for the optimization of the corresponding subsystem. Each subpopulation has an independent evolution mechanism, and global collaboration is realized through information exchange. The subpopulation structure is designed as , wherein represents the a-th subpopulation. , and represents the a-th individual in the a-th subpopulation. , represents the size of the a-th subpopulation.

[0040] The synergy degree oriented initialization strategy is based on the constructed synergy strength matrix, fully utilizes the coupling relationship between the subsystems and the system characteristic factors of each province and city, and specifically takes the supply subpopulation as an example. The system characteristic factor of the supply subpopulation is a supply resource endowment factor:​ , in, Indicates the The weight coefficient of renewable energy sources satisfies , Indicates the Provinces and cities Theoretical power generation of renewable energy sources, Indicates the The maximum theoretical power generation capacity of a renewable energy source.

[0041] The initialization strategy for the supply subpopulation combines historical data, synergy strength, and resource endowment characteristics: , in, Indicates the The historical mean of the decision variables of the supply subsystem, represents the disturbance factor, Indicates the supply subsystem and The synergy strength of the subsystems Express obedience Uniformly distributed random numbers, Represents the decision variable index.

[0042] The other subpopulations use similar initialization strategies, and the transport subpopulation is initialized based on the transmission demand characteristics: , represents the transmission demand coefficient; the demand subpopulation is initialized based on the load characteristics: , Represents the load characteristic coefficient; the energy storage subpopulation is initialized based on the configuration strategy: , Represents the energy storage configuration strategy coefficient.

[0043] The traditional sparrow optimization algorithm has significant deficiencies when dealing with multi-objective collaborative optimization problems in renewable energy power systems. First, the fixed parameter settings cannot adapt to different optimization stages and subsystem characteristics, resulting in low search efficiency; second, the lack of an effective constraint processing mechanism makes it difficult to handle complex coupling constraints in the power system; third, a single fitness evaluation system cannot balance the multi-objective requirements of collaborative performance and economy. To address these problems, the present invention proposes an adaptive parameter adjustment mechanism, a hierarchical constraint processing architecture, and a dual-objective fitness function. By dynamically adjusting the algorithm behavior, hierarchically processing complex constraints, and balancing multi-objective optimization, the performance of the algorithm in the collaborative optimization of complex power systems is significantly improved.

[0044] Specifically, if Figure 2 As shown, in one embodiment of the present invention, step S4 includes: S41. Construct a dual-objective fitness function, taking into account both the maximization of synergy and the minimization of cost.

[0045] Specifically, the fitness function is designed as a weighted summation, which balances the system's collaborative performance and economic requirements through weight coefficients, providing a clear optimization direction for the algorithm. The specific calculation formula is: , , Where, Indicates the In the subpopulation individual; Represents an individual The fitness value of Represents an individual The overall coordination of the corresponding configuration scheme; It represents the theoretical maximum value of synergy and is used for normalization; Represents an individual The total cost of the corresponding configuration solution; Indicates the maximum value of the cost, used for normalization; and Represent the weight coefficients of coordination and cost respectively, satisfying ; Indicates that from the individual The decision variable vector corresponding to the a-th subsystem extracted from the position vector; represents the cost function of the ath subsystem. The supply subsystem costs mainly include the investment cost and operation and maintenance costs of renewable energy equipment. The transmission subsystem costs include the construction and expansion costs of transmission lines. The demand subsystem costs include the demand response equipment investment and electricity purchase costs. The energy storage subsystem costs include the energy storage equipment investment cost and the charging and discharging loss costs.

[0046] S42. Calculate the initial fitness value of each individual in each subpopulation, assign roles based on the fitness ranking, and determine the number of discoverers, followers, and sentinels.

[0047] Specifically, the dual-objective fitness function is first used to calculate the initial fitness value of each individual in each subpopulation, and then roles are assigned based on the fitness ranking results. The discoverer is responsible for global search, the follower is responsible for local search, and the vigilant is responsible for boundary exploration. Role assignment is based on fitness ranking, with individuals with higher fitness acting as discoverers, individuals with medium fitness acting as followers, and individuals with lower fitness acting as vigilants. Number of discoverers: ; Number of followers: ; Number of Sentinels: . It represents the proportion of discoverers in the s-th subpopulation, and its value range is [0.1, 0.3].

[0048] S43, based on adaptive step size parameters Calculate dynamic warning value: , Where, Indicates the The warning value in the iteration, Indicates the initial warning value, Indicates the The adjustment coefficient of the subpopulation, Indicates the The subpopulation in Adaptive step size parameter for iterations.

[0049] Specifically, the adaptive step size parameter Taking into account the synergy feedback, convergence state and iterative process, the calculation formula is: , in, represents the initial step size parameter, Indicates the The subpopulation in The coordination degree of the iteration, Indicates the The maximum degree of cooperation among subpopulations, represents the maximum number of iterations, Indicates the The subpopulation in The convergence status evaluation value of iteration.

[0050] The convergence state evaluation is quantified by calculating the standard deviation change of the fitness value of each subpopulation: , in, Indicates the The subpopulation in The standard deviation of fitness at the iteration, Indicates the number of iterations: , in, Indicates the Subpopulation Individuals in The fitness value of the iteration, Indicates the The subpopulation in The average fitness value of iterations.

[0051] S44. The discoverer uses different search modes according to the warning value. A global search strategy is used when A local search strategy is used, where Represents the safety threshold; followers adopt different following methods according to individual rankings. and previous individuals follow the discoverer, ranked The individuals thereafter conduct random searches. Indicates the The size of the subpopulation; the sentinels adopt conservative or aggressive position updating strategies according to their relative fitness levels.

[0052] In this embodiment, the location update strategy of the discoverer adopts different search modes according to the warning value. When the warning value is small ( ), indicating a relatively safe environment, using a global search strategy: , in, represents the safety threshold, represents a uniform random number in (0,1]. In this embodiment, ST is set to 0.8. When the warning value is large ( ), indicating that there is a threat in the environment, and a local search strategy is adopted: , in, represents random numbers that follow a standard normal distribution, Represents the identity matrix.

[0053] The follower's position update strategy adopts different following methods according to the individual's position in the population. ), adopting the strategy of staying away from the worst individual: , For the top-ranked individuals ( ), adopting the strategy of learning from the optimal individual: , in, and Respectively represent The best and worst individuals in a subpopulation, Indicates a randomly selected symbol. Here, c represents the ranking number of the individual fitness.

[0054] The position update strategy of the vigilant adopts different behavior patterns according to the relationship between the individual's fitness value and the average level. The vigilant with fitness higher than the average level adopts a conservative strategy ( ): , Indicates the The subpopulation in The average fitness value of iterations.

[0055] Vigilants with below-average fitness adopt an aggressive strategy ( ): , in, and represents random numbers that follow a standard normal distribution, represents a small constant that prevents the denominator from being zero.

[0056] S45. A hierarchical constraint processing mechanism is used to constrain individuals after position updates, including boundary layer constraint processing to ensure that decision variables are within the allowable range, balance layer constraint processing to ensure the supply and demand balance constraints of the power system, and collaborative layer constraint processing to ensure the energy storage configuration ratio constraints and transmission capacity constraints.

[0057] In this embodiment, the boundary layer constraint process ensures that all decision variables are within the allowed range, and truncates variables that exceed the boundary: , in, and Respectively represent The lower and upper bounds of the decision variables.

[0058] The balancing layer constraints handle the supply and demand balance constraints of the power system to ensure the balance of power supply and demand in various provinces and cities: , in, and Respectively represent The power received and delivered by each province and city, and Represent the energy storage discharge and charging power respectively, Indicates the network loss power, Indicates the The electricity demand of provinces and cities.

[0059] When the power balance constraint is violated, the violation degree is calculated and repaired through energy storage adjustment: , When a power balance constraint violation is detected, the system first calculates the degree of violation , which represents the amount of electricity that is unbalanced between supply and demand in the i-th province or city. , indicating excess power supply, in this case priority is given to increasing energy storage charging power to accommodate the excess power while increasing the power export appropriately ; if , it indicates power shortage, at this time, the discharge power of the energy storage is increased preferentially to fill the power gap while increasing the power import appropriately The charging and discharging power constraints of the energy storage device and the transmission capacity constraints of the transmission line need to be satisfied simultaneously during the repair process. Through this combination repair strategy of energy storage regulation and power exchange, the power balance constraints are effectively satisfied, while the economy and technical feasibility of the solution are maintained.

[0060] The coordination layer handles the energy storage configuration ratio constraint to ensure that the ratio of energy storage capacity to renewable energy installed capacity is within a reasonable range: , wherein, and represent the lower and upper limits of the energy storage configuration ratio, respectively.

[0061] For individuals who violate the constraints, the fitness value is adjusted by a penalty function: , wherein, is the penalty coefficient, is the constraint violation degree.

[0062] S46, combine the current optimal individuals of each sub-population to form a candidate global solution, calculate its overall fitness, if better than the current global optimal solution, update the global optimal solution and feedback the update information to each sub-population.

[0063] In this embodiment, the information exchange frequency is dynamically calculated based on the convergence state difference of each sub-population. When the information exchange condition is met, the current optimal individuals of each sub-population are combined to form a candidate global solution, and its overall fitness is calculated. If better than the current global optimal solution, update the global optimal solution and feedback the update information to each sub-population.

[0064] The dynamic adjustment mechanism of information exchange frequency determines the exchange time based on the difference degree of the convergence state of each sub-population. When the convergence degree of each sub-population is greatly different, the exchange frequency is increased to promote cooperation, and when the convergence degree is similar, the exchange frequency is reduced to maintain diversity. The information exchange frequency calculation formula is: , wherein, is the initial exchange frequency, is the convergence state evaluation value of the th sub-population in the th iteration, is a small constant to avoid zero denominator.

[0065] The information exchange frequency dynamic adjustment mechanism determines the exchange time based on the difference degree of convergence state of each sub-population. Specifically, when the convergence degree of each sub-population is greatly different, it indicates that the optimization progress of different subsystems is uneven, at this time, the information exchange needs to be increased to promote the lagging sub-population to learn from the advanced sub-population, and realize collaborative optimization; on the contrary, when the convergence degree of each sub-population is similar, it indicates that the optimization of each subsystem is relatively balanced, at this time, the information exchange frequency should be reduced to maintain the search diversity of each sub-population, and avoid premature convergence to local optimal solution. The numerator part of the exchange frequency calculation formula reflects the imbalance degree of convergence state, when the difference between the minimum and maximum convergence degree is greater, the value is closer to 1, and the exchange frequency is higher; when the difference is smaller, the value is close to 0, and the exchange frequency is reduced.

[0066] When the information exchange condition is met (i.e. the difference of convergence state of each sub-population exceeds the threshold value), the optimal solution of each sub-population is combined to form a candidate solution: , and the overall fitness of the candidate solution is calculated . If the fitness of the candidate solution is better than the current global optimal solution , the global optimal solution is updated: , and the corresponding part of the global optimal solution is fed back to each sub-population: .

[0067] The implementation process of the global optimal solution feedback mechanism includes three key steps: first, a candidate global solution is formed by combining the current optimal solutions of each sub-population, which contains the latest optimization results of the four subsystems; second, the overall fitness of the candidate solution is calculated to evaluate its performance as a global solution; finally, if the fitness of the candidate solution is better than the current global optimal solution, the global optimal solution is updated, and the corresponding part of the updated global optimal solution is fed back to the corresponding sub-population. This feedback mechanism ensures that each sub-population maintains a relatively independent evolution process and can obtain timely guidance from the global optimal information, thereby maintaining search diversity while accelerating convergence to the global optimal solution. In the feedback process, each sub-population only receives the decision variable information related to its own subsystem, avoiding confusion between different subsystems.

[0068] S47, check whether the convergence condition is met, if the maximum iteration number is reached or the fitness improvement amplitude is less than the set threshold value, output the optimal configuration result; if not, let t=t+1, return to S43 to continue iteration optimization.

[0069] The convergence condition includes two conditions of reaching the maximum iteration number and the fitness improvement amplitude being less than the set threshold value ; wherein, represents the fitness of the i-th iteration​ the global optimal fitness value of the sub-iteration, denotes a convergence threshold). When any of the conditions is satisfied, the algorithm terminates and outputs the global optimal solution ; otherwise, t = t + 1, return to step S43 to continue iteration.

[0070] Specifically, in an embodiment of the present application, step S5 comprises: extracting the specific configuration parameters of each subsystem from the global optimal solution .

[0071] The extraction process of the supply subsystem configuration parameters involves reconstructing the first 186 elements in the global optimal solution according to the province and energy type, and the specific extraction formula is: , wherein, denotes the optimal installed capacity of the kth renewable energy in the ith province, denotes the element value at the corresponding position in the global optimal solution.

[0072] The extraction process of the transmission subsystem configuration parameters is based on the topological structure characteristics of the power transmission network. The transmission capacity configuration of each transmission line is determined by: , denotes the optimal transmission capacity of the ith transmission line, denotes the transmission line index, covering all inter-provincial power transmission channels and cross-regional transmission lines.

[0073] The extraction of the demand subsystem configuration parameters involves the determination of the power demand guarantee level of each province and city. The power demand guarantee level reflects the degree of satisfaction of the system to the power demand of each province and city, and the extraction formula is: , denotes the optimal power demand guarantee level of the ith province and city, the numerical range is usually between 0.9 and 1.1, and the value of 1 indicates complete satisfaction of the benchmark demand, greater than 1 indicates a certain demand growth margin, and less than 1 indicates a certain supply-demand gap but within an acceptable range.

[0074] The energy storage subsystem configuration parameters are extracted by: , denotes the optimal energy storage installed capacity of the ith province and city.

[0075] The performance indicators of the configuration scheme are calculated, including the overall coordination degree of the system, the total investment cost, the annual operation cost, the renewable energy utilization rate, and the power supply reliability. The overall coordination degree and the total investment cost are calculated according to the corresponding formulas, the annual operation cost mainly includes the operation and maintenance cost of the renewable energy equipment and the energy storage equipment, the power loss cost of the transmission line, and the cost generated by the power transaction. The operation and maintenance cost is calculated according to the installed capacity of various types of equipment and the corresponding annual operation and maintenance rate, the power loss cost is determined according to the transmission capacity, the loss rate and the power price of the line, and the power transaction cost is calculated based on the inter-provincial power exchange amount and the market price. The renewable energy utilization rate is measured by the ratio of the actual power generation capacity of each province and city to the installed capacity. The power supply reliability indicator is evaluated based on the comparison between the total power supply capacity of the system and the peak load demand.

[0076] The complete supply-transmission-demand-storage coordinated optimization configuration report includes the configuration scheme overview, the detailed configuration parameter table of each subsystem, the key performance indicator summary, the economic analysis, the technical feasibility evaluation and the implementation suggestion, etc.

[0077] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for collaborative optimization configuration of supply-transmission-demand-storage of renewable energy power system, characterized by: include: S1. Obtain basic data from multiple provinces and cities, including renewable energy resource data and auxiliary data, standardize the renewable energy resource data, define decision variables for the four subsystems of supply, transmission, demand, and energy storage, and obtain a unified resource evaluation basis and decision variable system; S2. Establish a synergy calculation model to calculate the synergy of each subsystem separately, calculate the overall synergy based on the synergy of each subsystem, and construct a subsystem synergy strength matrix to quantify the coupling relationship between each subsystem; S3. Establish a multi-subpopulation collaborative optimization architecture, decompose the population into four subpopulations according to the four subsystems of supply, transportation, demand, and energy storage, and use a synergy-guided initialization strategy to generate the initial solution set for each subpopulation; S4. Use the improved sparrow optimization algorithm to perform parallel optimization on each sub-population, evaluate the fitness with the dual objectives of coordination and total cost, and perform collaborative optimization between sub-populations through adaptive parameter adjustment and information exchange mechanism. When the convergence condition is reached, the optimal configuration result is obtained; S5. Based on the optimal configuration results, determine the renewable energy installed capacity, energy storage equipment configuration, power transmission plan and demand response strategy for each province and city, and output the supply-transmission-demand-storage coordinated optimization configuration plan for the renewable energy power system.

2. A renewable energy power system supply-transmission-demand-storage coordinated optimization configuration method according to claim 1, characterized in that: Step S2 includes: S21. Establish calculation formulas for the synergy of the supply subsystem, transmission subsystem, demand subsystem, and energy storage subsystem, respectively. The synergy of the transmission subsystem, demand subsystem, and energy storage subsystem is calculated by weighted summation of the impact indicators and weights of each subsystem. The uncertainty coefficient of renewable energy output needs to be additionally deducted from the supply subsystem. The uncertainty coefficient is quantified by calculating the variance of renewable energy power generation in each province and city. S22. Calculate the synergy of each subsystem using the geometric mean method to obtain the overall synergy; S23. Evaluate the degree of internal uncertainty by calculating the entropy value of each subsystem, calculate the weight coefficient of each subsystem based on the entropy value, and construct a 4×4 symmetrical synergy intensity matrix by combining the Pearson correlation coefficient and weight coefficient between subsystems to quantify the degree of mutual influence between subsystems.

3. A renewable energy power system supply-transmission-demand-storage coordinated optimization configuration method according to claim 1, characterized in that: Step S3 includes: S31. Establish a multi-subpopulation collaborative optimization architecture, in which the supply subpopulation is responsible for optimizing the installed capacity of renewable energy in each province and city, including the decision variables for the installed capacity of various types of renewable energy in each province and city; the transport subpopulation is responsible for optimizing the transmission capacity of transmission lines; the demand subpopulation is responsible for optimizing the power demand guarantee level in each province and city; and the energy storage subpopulation is responsible for optimizing the installed capacity of energy storage in each province and city; S32. Calculate the system characteristic factor of each subsystem based on the auxiliary data of each province and city, where the system characteristic factor of the supply subpopulation is the resource endowment factor, the system characteristic factor of the transport subpopulation is the transmission demand coefficient, the system characteristic factor of the demand subpopulation is the load characteristic coefficient, and the system characteristic factor of the energy storage subpopulation is the configuration strategy coefficient; S33. Based on the synergy intensity matrix and the system characteristic factors of each subsystem, an initial solution set of each subpopulation is generated respectively.

4. A renewable energy power system supply-transmission-demand-storage coordinated optimization configuration method according to claim 1, characterized in that: Step S4 includes: S41. Construct a dual-objective fitness function, taking into account both the maximization of synergy and the minimization of cost. S42. Calculate the initial fitness value of each individual in each subpopulation, assign roles based on the fitness ranking, and determine the number of discoverers, followers, and sentinels; S43, based on adaptive step size parameters Calculate dynamic warning value: ， Where, Indicates the The warning value in the iteration, Indicates the initial warning value, Indicates the The adjustment coefficient of each subpopulation, Indicates the The subpopulation in Adaptive step size parameter for iterations; S44. The discoverer uses different search modes according to the warning value. A global search strategy is used when A local search strategy is used, where Represents the safety threshold; followers adopt different following methods according to individual rankings. and previous individuals follow the discoverer, ranked The individuals thereafter conduct random searches. Indicates the The size of the subpopulation; the vigilant adopts a conservative or aggressive position update strategy according to the relative level of fitness; S45. A hierarchical constraint processing mechanism is used to process constraints on individuals after position updates, including boundary layer constraint processing to ensure that decision variables are within the allowable range, balance layer constraint processing to process power system supply and demand balance constraints, and coordination layer constraint processing to process energy storage configuration ratio constraints and transmission capacity constraints; S46. Dynamically calculate the information exchange frequency based on the difference in convergence state of each subpopulation. When the exchange condition is met, combine the current best individuals of each subpopulation to form a candidate global solution, calculate its overall fitness, and if it is better than the current global optimal solution, update the global optimal solution and feed back the updated information to each subpopulation. S47. Check whether the convergence conditions are met. If the maximum number of iterations is reached or the fitness improvement is less than the set threshold, output the optimal configuration result; if not, set t=t+1 and return to S43 to continue iterative optimization.

5. A method for coordinated optimization configuration of supply-transmission-demand-storage of a renewable energy power system according to claim 4, characterized in that: Adaptive step size parameter Taking into account the synergy feedback, convergence state and iterative process, the calculation formula is: ， in, represents the initial step size parameter, Indicates the The subpopulation in The coordination degree of the iteration, Indicates the The maximum degree of cooperation among subpopulations, represents the maximum number of iterations, Indicates the The subpopulation in The convergence status evaluation value of iteration.

6. A renewable energy power system supply-transmission-demand-storage coordinated optimization configuration method according to claim 4, characterized in that: Step S46 includes: The information exchange frequency is dynamically calculated based on the difference in the convergence status of each subpopulation. When the convergence degree of each subpopulation is greatly different, the exchange frequency is increased to promote cooperation. When the convergence degree is similar, the exchange frequency is reduced to maintain diversity. The information exchange frequency calculation formula is: ， in, represents the initial exchange frequency, Indicates the The subpopulation in The convergence status evaluation value of the iteration, represents a small constant that avoids the denominator being zero; When the information exchange conditions are met, the best individuals of each subpopulation are combined to form a candidate solution and its overall fitness is calculated. If the candidate solution is better than the current global optimal solution, the global optimal solution is updated and fed back to each subpopulation.

7. A renewable energy power system supply-transmission-demand-storage coordinated optimization configuration method according to claim 4, characterized in that: The hierarchical constraint processing mechanism includes: boundary layer constraint processing, which truncates decision variables that exceed the variable boundaries, so that all decision variables are within the allowable range; balance layer constraint processing, which processes the supply and demand balance constraints of the power system. Based on the power demand data in the auxiliary data, when a violation of the power balance constraint is detected, the degree of violation is calculated and repaired through energy storage charging and discharging power adjustment and inter-provincial power exchange; collaborative layer constraint processing, which processes energy storage configuration ratio constraints and transmission capacity constraints, keeps the ratio of energy storage capacity to renewable energy installed capacity within a reasonable range, and adjusts the fitness value of individuals that violate the constraints through a penalty function.

8. A renewable energy power system supply-transmission-demand-storage coordinated optimization configuration method according to claim 4, characterized in that: The calculation formula of the dual-objective fitness function is as follows: ， ， Where, Indicates the In the subpopulation individual; Represents an individual The fitness value of Represents an individual The overall coordination of the corresponding configuration scheme; It represents the theoretical maximum value of synergy and is used for normalization; Represents an individual The total cost of the corresponding configuration solution; Indicates the maximum value of the cost, used for normalization; and Represent the weight coefficients of coordination and cost respectively, satisfying ; Indicates that from the individual The decision variable vector corresponding to the a-th subsystem extracted from the position vector; represents the cost function of the a-th subsystem.

9. The method for coordinated optimization configuration of supply-transmission-demand-storage of a renewable energy power system according to claim 1, characterized in that: The auxiliary data in step S1 includes: Electricity demand data, including historical electricity load data, load growth trend data, peak-valley load difference coefficient, load forecast data, and demand response potential data for each province and city, with data time granularity covering annual, monthly, daily, and hourly levels; transmission network data, including the rated transmission capacity, line impedance parameters, line length, and transmission loss coefficient of existing transmission lines between provinces and cities, as well as construction planning data and transmission capacity expansion potential data for cross-regional transmission channels; energy storage-related data, including the installed capacity, technology type distribution, charging and discharging efficiency parameters, and investment and operation and maintenance cost data of existing energy storage facilities in each province and city; economic development data, energy consumption structure data, and policy support intensity data for each province and city.

10. The method for coordinated optimization configuration of supply-transmission-demand-storage of a renewable energy power system according to claim 1, characterized in that: Step S5 includes: The specific configuration parameters of each subsystem are extracted from the global optimal solution, including the installed capacity of various types of renewable energy in each province and city of the supply subsystem, the transmission capacity configuration of each transmission line in the transmission subsystem, the power demand guarantee level in each province and city of the demand subsystem, and the installed energy storage capacity in each province and city of the energy storage subsystem. The key performance indicators of the configuration scheme are calculated, including the overall system coordination, total investment cost, annualized operating cost, renewable energy utilization rate and power supply reliability. A complete supply-transmission-demand-storage coordinated optimization configuration report is generated.

Images