Wind power plant renting and recruitment sharing heat energy storage optimization operation method based on master-slave game

Abstract

The invention discloses a wind power plant leasing and recruitment sharing heat energy storage optimization operation method based on a master-slave game, and the method comprises the following steps: building a WFCO optimization model, building a CSESSO optimization model, building a CSESSO leasing service pricing model based on the master-slave game according to the game relation between the WFCO optimization model and the CSESSO optimization model, determining the leasing price and the leasing capacity, and carrying out the leasing and recruitment sharing heat energy storage optimization operation. The number of iterations is initialized; using an improved Shapley value to optimize lease cost distribution of the CSESSO lease service pricing model based on the master-slave game, calculating the global optimal lease income of the CSESSO in each time period, and recording the lease price and lease capacity in each time period; and repeatedly calculating the lease income according to the relationship between the calculated lease income and the income threshold until the number of iterations reaches the iteration threshold, obtaining the CSESSO optimal lease income, and completing the lease and recruitment sharing heat energy storage optimization of the wind power plant based on the master-slave game.

Description

Technical Field

[0001] The present invention belongs to the technical field of shared thermal energy storage in power systems, and in particular relates to a method for optimizing the operation of shared thermal energy storage in wind farms based on master-slave game. Background Art

[0002] The new power system is a key way to achieve global net zero carbon emissions. As the penetration rate of renewable energy continues to increase, emerging energy has posed a challenge to the reliability of the power system. In order to meet this challenge, the renewable energy system uses the flexible characteristics of energy storage to smooth out the fluctuations and deviations of renewable energy output and achieve reliable and high-quality power supply. However, some areas with a high proportion of renewable energy power generation are faced with the dilemma of the conflict between the mandatory configuration of new energy storage policies and the low utilization rate and poor economic efficiency after energy storage configuration. There is an urgent need to optimize the commercial operation model of energy storage to reasonably increase energy storage revenue while smoothing out the fluctuations and deviations of renewable energy output.

[0003] In order to improve the efficiency and economic feasibility of energy storage power stations, the shared energy storage business model has become a research hotspot. New energy stations lease shared energy storage in a "rent-to-build" manner. Based on the principle of shared economy, a symbiotic environment is formed through a multi-user and multi-agent collaboration model, which has high dispatchability and significant economic benefits. Important national directives such as the "Electricity Ancillary Service Management Measures" and the "Grid Connection Operation Management Regulations" issued by the National Energy Administration emphasize strengthening the competitive market pricing mechanism of energy storage and advocating the improvement of the collective investment and allocation framework of energy storage. Researchers have conducted in-depth research on sharing models including capacity leasing, peak-shaving income, capacity compensation and other ancillary services. The prior art constructs a two-stage optimization model of the master-slave game between leasing users and shared energy storage, introduces a user preference mechanism for multiple objectives such as emission reduction and distributed power generation, and optimizes the shared energy storage leasing price according to the user's leasing behavior. The prior art proposes a P2P energy trading scheme, in which distributed photovoltaic operators and users participate in the community sharing market, guides leasing users to participate in shared energy storage leasing services through a dynamic leasing service price mechanism, and uses a reinforcement learning algorithm to solve a reasonable leasing price. The existing technology also proposes a distributed photovoltaic community energy sharing optimization operation strategy based on a two-layer structure. The upper layer dynamically optimizes the charging and discharging decisions of centralized energy storage and the energy storage sharing price, and the lower layer optimizes the energy sharing income distribution among the entities of the distributed photovoltaic community to achieve a reasonable distribution of the additional income of the alliance. However, the ongoing challenges related to market dynamics make the low utilization and high cost of electrochemical energy storage a persistent problem.

[0004] Electrochemical energy storage power stations face limitations in directly addressing the different energy consumption needs of heat, electricity and other forms. Thermal energy dominates in final energy consumption and has great potential to balance energy supply and demand. By 2020, thermal energy will account for 73% of the world's total energy consumption.

[18] . Therefore, the idea of ​​shared heat storage with heat storage as the main form of energy storage has become an important solution, which can supply the energy consumption of end users in a combined mode of electricity and heat, thereby maximizing the benefits of shared energy storage power stations. In 2023, demonstration projects of shared thermal energy storage power stations based on solid thermal storage will be built in Hami and Kuitun, Xinjiang, respectively, to meet the peak and voltage regulation needs of the power grid. The prior art proposes an optimized scheduling strategy for the electric-thermal combined system considering the consumption behavior characteristics of distributed solid electric thermal storage users, and establishes a constraint model that meets the thermal storage heat required by users to improve the wind power absorption capacity. The prior art also studies the key parameter design and economic evaluation method of electric heating solid thermal storage devices, and analyzes the influence of valley electricity utilization coefficient on the economic efficiency of electric heating solid thermal storage devices. The above literature has conducted research from the perspective of solid thermal storage, and failed to fully consider the shared economic value of solid thermal storage. The prior art has also developed a typical solid-state thermal storage cogeneration shared energy storage system, established an optimization model for maximizing economic benefits, and analyzed the feasibility of multi-mode operation under given scenarios. Currently, researchers have explored shared energy storage based on solid thermal storage from the perspectives of optimized scheduling and economic evaluation, but have failed to consider the energy storage leasing needs of power generation side resources.

[0005] In summary, existing studies rarely consider the energy sharing and coordinated optimization operation of wind farm clusters and shared thermal energy storage on the power generation side. Summary of the invention

[0006] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method for optimizing the operation of wind farm leasing shared thermal energy storage based on master-slave game, which solves the problem that existing studies rarely consider the energy sharing and coordinated optimization operation of wind farm clusters and shared thermal energy storage on the power generation side, and realizes the optimization of the benefits of the source-storage alliance participating in the peak-shaving and dispatching of the distribution network.

[0007] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a method for optimizing the operation of wind farm leasing and sharing thermal energy storage based on master-slave game, comprising the following steps:

[0008] S1. Establish a WFCO (wind farm group operator) optimization model, and optimize the leased capacity in response to the lease price through the WFCO optimization model;

[0009] S2. Establish a CSESSO (Shared Thermal Energy Storage Operator) optimization model, and adjust the lease price according to the optimized lease capacity through the CSESSO optimization model;

[0010] S3. According to the game relationship between the WFCO optimization model and the CSESSO optimization model, a CSESSO leasing service pricing model based on the master-slave game is established to determine the leasing price and leasing capacity, and initialize the number of iterations;

[0011] S4. Use the improved Shapley value to optimize the rental cost allocation of the CSESSO rental service pricing model based on the master-slave game, calculate the global optimal rental income of CSESSO in each period according to the allocated rental cost, rental capacity and rental price, and record the rental price and rental capacity in each period;

[0012] S5. Set the income threshold and iteration threshold, and repeat S4 according to the relationship between the calculated rental income and the income threshold until the number of iterations reaches the iteration threshold, so as to obtain the optimal rental income of CSESSO and complete the optimization of renting and sharing thermal energy storage in wind farms based on master-slave game.

[0013] Further: In S1, the leasing capacity demand is optimized according to the calculated WFCO total revenue, and the expression of the objective function of the WFCO optimization model is specifically:

[0014]

[0015]

[0016] In the formula, I W represents the total revenue of WFCO; They represent the WFCO grid-connected power revenue and the wind power sales revenue respectively; C L , C b They represent the leasing cost and the grid-connected fluctuation assessment cost respectively; p t , P t ADN,w They represent WFCO on-grid electricity price and grid-connected power respectively; t represents time; ε represents the unit grid-connected power fluctuation assessment cost; P t , P t-1 They represent the predicted output of WFCO at time t and the predicted output of WFCO at time t-1 respectively; P limit It indicates the upper limit of power fluctuation during the process of wind farm grid connection; Δt indicates the time interval;

[0017] The constraints of the WFCO optimization model include WFCO output constraints, WFO (wind farm operator) output constraints, and WFCO wind curtailment power constraints;

[0018] The expression of the WFCO output constraint is specifically:

[0019] 0≤P t ADN,w ≤P t L

[0020] Where P t L It indicates the maximum power that the WFCO grid-connected node can accommodate;

[0021] The expression of the WFO output constraint is specifically:

[0022]

[0023] In the formula, They represent the dispatch output and forecast output of the mth WFO respectively;

[0024] The expression of the WFCO wind power curtailment constraint is specifically:

[0025] 0≤P t W,Q,c ≤max(P t F -P t L ,0)

[0026] Where P t F Indicates the forecast output of WFCO.

[0027] Further: In S2, the rental price is adjusted according to the calculated total operating income of CSESSO, and the expression of the objective function of the CSESSO optimization model is specifically:

[0028]

[0029] In the formula, max represents the maximum value, I CSESSO represents the total operating revenue of CSESSO, Respectively represent the rental income, frequency regulation service income, and heating service income of CSESSO; C loss , C w Respectively represent the operating cost of CSESSO and the cost of wind power purchase. C loss , C w The specific expression is:

[0030]

[0031] In the formula, They represent the capacity rental price and electricity rental price of CSESSO respectively; L t , H t Respectively represent the rental capacity and rental mileage of WFCO; P t ADN,d , P t ADN,c They represent the interaction power between CSESSO and the main grid, and the grid-connected power respectively; They represent the heating power price and grid-connected electricity price of the interaction between CSESSO and the main grid respectively; κ t , They represent the CSESSO heating electricity price and heating power respectively; γ represents the CSESSO operating cost, represents the heat loss of the shared thermal energy storage at time t; They represent the wind power price and wind power amount abandoned by WFCO respectively;

[0032] The constraints of the CSESSO optimization model include the electric and thermal power balance constraint and the CSESSO operation constraint. The expression of the electric and thermal power balance constraint is specifically:

[0033]

[0034] P t ADN,d ⊥P t ADN,c =0

[0035] Where P t W represents the leased power of the wind power cluster, α represents the heating power and heating efficiency of the CSESSO resistance wire respectively; ⊥ is an operator, indicating that at least one of the parameters is 0;

[0036] The expression of the operation constraint of the CSESSO is specifically:

[0037]

[0038] In the formula, μ min , μ max Respectively represent the lower limit and upper limit of the heat load inertia adjustment ratio; H t L represents the heat load dispatched by CSESSO; Respectively represent the maximum efficiency of CSESSO charging and heating.

[0039] Further: In S3, the expression of the CSESSO leasing service pricing model G based on the master-slave game is specifically:

[0040] G={N;S {CSESSO∪WFCO} ;I {CSESSO∪W}}

[0041] Where N represents the set of participants, and N = {CSESSO ∪ WFCO}, where CSESSO is the principal participant, WFCO is the subordinate participant, S {CSESSO∪WFCO} Represents a policy set, which includes the principal CSESSO policy set and the slave WFCO policy set. {CSESSO∪W} represents the benefits, which include the total benefits of WFCO and the total operating benefits of CSESSO;

[0042] Subject CSESSO strategy set S CSESSO The specific expression is:

[0043]

[0044] Slave WFCO policy set S WFCO The specific expression is:

[0045]

[0046] In the formula, R t represents the rental capacity, Indicates the grid-connected power, represents the revenue from curtailed wind power sales, C L It represents the grid-connected fluctuation assessment cost;

[0047] The constraint functions of the CSESSO leasing service pricing model based on the master-slave game include WFCO game constraints and CSESSO game constraints;

[0048] The expression of the WFCO game constraint is specifically:

[0049]

[0050] In the formula, Assess the cost of grid connection fluctuation for WFCO;

[0051] The expression of the CSESSO game constraint is specifically:

[0052]

[0053] In the formula, is the maximum price difference in the spot market, and ψ is the conversion factor.

[0054] Further: In S4, the method of optimizing the leasing cost allocation of the CSESSO leasing service pricing model based on the master-slave game using the improved Shapley value comprises the following sub-steps:

[0055] S41, using the equal power-successive load method to calculate the equivalent output of each wind farm, calculating the waveform similarity between the equivalent output and the actual output curve according to the cosine similarity algorithm, and then calculating the first to third redistribution sub-coefficients;

[0056] S42. Calculate the redistribution coefficient according to the first to third redistribution sub-coefficients, and then establish a rental cost allocation model based on the improved Shapley value, calculate the allocation result of the improved Shapley value, and complete the rental cost allocation.

[0057] Further: In S41, the waveform similarity S between the equivalent output curve and the actual output curve is calculated.i,wave The specific expression is:

[0058]

[0059] In the formula, represents the equivalent output of each wind farm, represents the predicted output of the mth wind farm in the wind farm cluster in time period t, i represents the ordinal number of the wind farm;

[0060] The specific expression for calculating the first redistribution sub-coefficient K1 is:

[0061]

[0062] The specific expression for calculating the second redistribution sub-coefficient K2 is:

[0063]

[0064] The specific expression for calculating the third redistribution sub-coefficient K3 is:

[0065]

[0066] In the formula, r mn 、r m represent the correlation coefficient and mean, respectively.

[0067] Further: In S42, the redistribution coefficient K is calculated m The specific expression is:

[0068] K m =θ1K1+θ2K2+θ3K3

[0069] In the formula, θ1, θ2 and θ3 all represent the weights of the apportionment coefficients;

[0070] Calculate the amortized result of the improved Shapley value The specific expression is:

[0071]

[0072] In the formula, ΔK m represents the allocation coefficient, M represents the number of wind farms, x m represents the allocation result of the original Shapley value; χ represents the control factor; C represents the WFCO rental cost.

[0073] Further: S5 includes the following sub-steps:

[0074] S51. Determine whether the rental income is greater than a preset income threshold;

[0075] If yes, the rental income is used as the income threshold, and it is determined whether the number of iterations is greater than the iteration threshold. If yes, the income threshold is used as the optimal rental income of CSESSO, and the process goes to S53. If no, the number of iterations is increased by 1, and the process goes to S52.

[0076] If not, proceed to S52;

[0077] S52, adjust the parameters of the CSESSO leasing service pricing model based on the master-slave game, determine the new leasing price and leasing capacity, repeat S4 to calculate the global optimal leasing income of the new CSESSO in each period, and return to S51;

[0078] S53. Output the calculated rental price and rental capacity corresponding to the optimal rental income of CSESSO, and complete the optimization of renting and hiring shared thermal energy storage in wind farms based on master-slave game.

[0079] The beneficial effects of the present invention are:

[0080] (1) The present invention provides a method for optimizing the operation of wind farm leasing shared thermal energy storage based on master-slave game, which realizes efficient utilization of new energy storage and new energy resources by leasing a shared thermal energy storage joint operation mechanism taking solid thermal storage into consideration through a wind farm cluster.

[0081] (2) The present invention constructs a CSESSO leasing service pricing model based on master-slave game, with shared thermal energy storage operators as the main body and wind farm cluster alliances as the slaves. It introduces a shared thermal energy storage leasing service pricing mechanism based on a two-part electricity price by referring to the frequency modulation market. It uses an improved genetic algorithm nested with CPLEX commercial software to solve the game equilibrium solution, determine the shared thermal energy storage leasing price and energy use strategy, and the follower leasing demand, so as to maximize the operating benefits of each entity.

[0082] (3) Based on the uncertainty of wind farm clusters and dynamic leasing demand, the present invention proposes a two-stage optimization operation strategy for shared thermal energy storage. The correlation between the leasing members of each wind farm and the correlation between individual members and the alliance are fully considered, and the improved Shapley value is used to solve the problem, so as to achieve a reasonable allocation of leasing costs for each member. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 The present invention provides a method for optimizing the operation of wind farm leasing and sharing thermal energy storage based on master-slave game.

[0084] Figure 2 Predict power diagrams for each wind farm.

[0085] Figure 3 Predict output and node maximum capacity power diagram for wind farm clusters.

[0086] Figure 4 This is a graph of electricity and heat loads and market clearing prices. DETAILED DESCRIPTION

[0087] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0088] Embodiment 1:

[0089] like Figure 1 As shown, in one embodiment of the present invention, a method for optimizing the operation of a wind farm leasing and sharing thermal energy storage based on a master-slave game comprises the following steps:

[0090] S1. Establish a WFCO optimization model, and optimize the leasing capacity by responding to the leasing price through the WFCO optimization model;

[0091] S2. Establish a CSESSO optimization model, and adjust the lease price according to the optimized lease capacity through the CSESSO optimization model;

[0092] S3. According to the game relationship between the WFCO optimization model and the CSESSO optimization model, a CSESSO leasing service pricing model based on the master-slave game is established to determine the leasing price and leasing capacity, and initialize the number of iterations;

[0093] S4. Use the improved Shapley value to optimize the rental cost allocation of the CSESSO rental service pricing model based on the master-slave game, calculate the global optimal rental income of CSESSO in each period according to the allocated rental cost, rental capacity and rental price, and record the rental price and rental capacity in each period;

[0094] S5. Set the income threshold and iteration threshold, and repeat S4 according to the relationship between the calculated rental income and the income threshold until the number of iterations reaches the iteration threshold, so as to obtain the optimal rental income of CSESSO and complete the optimization of renting and sharing thermal energy storage in wind farms based on master-slave game.

[0095] In S1, WFCO benefits from selling planned electricity to the grid and selling abandoned wind power to shared thermal energy storage; in addition, WFCO uses leased energy storage capacity to offset grid-connected fluctuation assessment costs to maximize economic benefits.

[0096] In S1, the leasing capacity demand is optimized according to the calculated total WFCO revenue, and the expression of the objective function of the WFCO optimization model is specifically:

[0097]

[0098] In the formula, I W represents the total revenue of WFCO; They represent the WFCO grid-connected power revenue and the wind power sales revenue respectively; C L , C b They represent the leasing cost and the grid-connected fluctuation assessment cost respectively; p t , P t ADN,w represents the WFCO on-grid electricity price and grid-connected power respectively; t represents time; ε represents the unit grid-connected power fluctuation assessment cost, which is 0.3654 yuan / kWh in this embodiment; P t , P t-1 They represent the predicted output of WFCO at time t and the predicted output of WFCO at time t-1 respectively; P limit represents the upper limit of power fluctuation during the process of wind farm grid connection; Δt represents the time interval, which is 1h in this embodiment;

[0099] The constraints of the WFCO optimization model include WFCO output constraints, WFO output constraints, and WFCO wind power curtailment constraints;

[0100] The expression of the WFCO output constraint is specifically:

[0101] 0≤P t ADN,w ≤P t L

[0102] Where P t L It indicates the maximum power that the WFCO grid-connected node can accommodate;

[0103] The expression of the WFO output constraint is specifically:

[0104]

[0105] In the formula, They represent the dispatch output and forecast output of the mth WFO respectively;

[0106] The expression of the WFCO wind power curtailment constraint is specifically:

[0107] 0≤P t W,Q,c ≤max(P t F -P t L ,0)

[0108] Where P t F Indicates the forecast output of WFCO.

[0109] In S2, CSESSO formulates a leasing price strategy based on the power leasing demand for each period provided by WFCO and the wind power abandoned by WFCO, and optimizes the CSESSO leasing price. Improve the operating efficiency of CSESSO and promote the consumption of new energy. Shared thermal energy storage uses the frequency regulation auxiliary service market mechanism to trade with the upper-level power grid and realize arbitrage by clearing the price difference. At the same time, it provides heat source services to thermal loads, promotes the diversified value of shared thermal energy storage, and maximizes economic benefits.

[0110] In S2, the rental price is adjusted according to the calculated total operating income of CSESSO. The expression of the objective function of the CSESSO optimization model is specifically:

[0111]

[0112] In the formula, max represents the maximum value, I CSESSO represents the total operating revenue of CSESSO, Respectively represent the rental income, frequency regulation service income, and heating service income of CSESSO; C loss , C w Respectively represent the operating cost of CSESSO and the cost of wind power purchase. C loss , C w The specific expression is:

[0113]

[0114] In the formula, They represent the capacity rental price and electricity rental price of CSESSO respectively; L t , H t Respectively represent the rental capacity and rental mileage of WFCO; P t ADN,d , P t ADN,c They represent the interaction power between CSESSO and the main grid, and the grid-connected power respectively; They represent the heating power price and grid-connected electricity price of the interaction between CSESSO and the main grid respectively; κ t , They represent the CSESSO heating electricity price and heating power respectively; γ represents the CSESSO operating cost, represents the heat loss of the shared thermal energy storage at time t; They represent the wind power price and wind power amount abandoned by WFCO respectively;

[0115] The constraints of the CSESSO optimization model include the electric and thermal power balance constraint and the CSESSO operation constraint. The expression of the electric and thermal power balance constraint is specifically:

[0116]

[0117] P t ADN,d ⊥P t ADN,c =0

[0118] Where P t W represents the leased power of the wind power cluster, α represents the heating power and heating efficiency of the CSESSO resistance wire, and the value of α is 0.3; ⊥ is an operator, indicating that at least one of the parameters is 0;

[0119] The expression of the operation constraint of the CSESSO is specifically:

[0120]

[0121]

[0122] In the formula, μ min , μ max They respectively represent the lower limit and upper limit of the heat load inertia adjustment ratio, which are set to 0.9 and 1.1 respectively in this embodiment; represents the heat load dispatched by CSESSO; Respectively represent the maximum efficiency of CSESSO charging and heating.

[0123] In S3, WFCO and CSESSO are two interest entities. CSESSO optimizes the rental price, WFCO optimizes the rental capacity demand by responding to the rental price, and CSESSO adjusts the rental price by optimizing the rental demand. The two parties have a game relationship, and each entity has a decision-making power. Therefore, a master-slave game optimization model is established with CSESSO as the priority to determine the rental price. The game consists of three elements: the participant set, the strategy set, and the benefit. This embodiment describes the master-slave game as a CSESSO rental service pricing model based on the master-slave game.

[0124] In S3, the expression of the CSESSO rental service pricing model G based on the master-slave game is specifically:

[0125] G={N;S {CSESSO∪WFCO} ;I {CSESSO∪W}}

[0126] Where N represents the set of participants, and N = {CSESSO ∪ WFCO}, where CSESSO is the principal participant, WFCO is the subordinate participant, S {CSESSO∪WFCO} Represents a policy set, which includes the principal CSESSO policy set and the slave WFCO policy set.{CSESSO∪W} represents the benefits, which include the total benefits of WFCO and the total operating benefits of CSESSO;

[0127] Subject CSESSO strategy set S CSESSO The specific expression is:

[0128]

[0129] Slave WFCO policy set S WFCO The specific expression is:

[0130]

[0131] In the formula, R t represents the rental capacity, Indicates the grid-connected power, represents the revenue from curtailed wind power sales, C L It represents the grid-connected fluctuation assessment cost;

[0132] In this embodiment, in order to avoid the potential possibility that CSESSO and WFCO deliberately raise or lower the rental price, the present invention introduces WFCO game constraints and CSESSO game constraints to ensure the feasibility and effectiveness of the rental price and the fairness of the respective interests of CSESSO and WFCO.

[0133] The constraint functions of the CSESSO leasing service pricing model based on the master-slave game include WFCO game constraints and CSESSO game constraints;

[0134] The expression of the WFCO game constraint is specifically:

[0135]

[0136] In the formula, Assess the cost of grid connection fluctuation for WFCO;

[0137] The WFCO game constraint means that WFCO leases capacity from CSESSO to cope with grid fluctuation assessment, and the total leasing cost of WFCO is not higher than the assessment cost.

[0138] The expression of the CSESSO game constraint is specifically:

[0139]

[0140] In the formula, is the maximum price difference in the spot market, ψ is the conversion factor, and in this embodiment, it is 0.85.

[0141] The CSESSO game constraint means that CSESSO obtains income by leasing energy storage capacity, and the operating costs and opportunity costs of CSESSO are not higher than the leasing income.

[0142] In S4, the method for optimizing the leasing cost allocation of the CSESSO leasing service pricing model based on the master-slave game using the improved Shapley value includes the following sub-steps:

[0143] S41, using the equal power-successive load method to calculate the equivalent output of each wind farm, calculating the waveform similarity between the equivalent output and the actual output curve according to the cosine similarity algorithm, and then calculating the first to third redistribution sub-coefficients;

[0144] S42. Calculate the redistribution coefficient according to the first to third redistribution sub-coefficients, and then establish a rental cost allocation model based on the improved Shapley value, calculate the allocation result of the improved Shapley value, and complete the rental cost allocation.

[0145] In this embodiment, the redistribution coefficient is set to improve the fairness of cost sharing among wind farms. The sharing strategy is more in line with practical applications. The improved Shapley value is used to optimize the cost distribution in the CSESSO leasing service pricing model based on the master-slave game, and the traditional Shapley value is improved by considering the output and load demand of each wind farm. The following factors are mainly considered:

[0146] 1) The similarity between the output of each wind farm and the total load demand waveform.

[0147] The higher the similarity between the output change trend of each wind farm and the waveform of the load demand curve, the less the transaction volume with CSESSO and the lower the transaction cost. By making the output curve of each wind farm follow the load demand to achieve mutual assistance between surplus and power shortage, a distribution coefficient considering the waveform similarity is designed.

[0148] First, the equivalent output of each wind farm is calculated using the equal power-successful load method. Its equivalent output is equal to the cumulative output of the wind farm, and its change trend follows the load demand change:

[0149]

[0150] Then the cosine similarity algorithm is used to calculate the waveform similarity S between the equivalent output and actual output curves. i,wave The greater the waveform similarity, the greater the contribution of each wind farm to meeting load demand and reducing leasing costs, and the smaller the corresponding redistribution coefficient should be.

[0151] In S41, the waveform similarity S between the equivalent output curve and the actual output curve is calculated. i,wave The specific expression is:

[0152]

[0153] In the formula, represents the equivalent output of each wind farm, represents the predicted output of the mth wind farm in the wind farm cluster in time period t, i represents the ordinal number of the wind farm;

[0154] The specific expression for calculating the first redistribution sub-coefficient K1 is:

[0155]

[0156] 2) The correlation between WFO and WFCO.

[0157] The amplitude of WFO output has a certain impact on WFCO output. The greater the WFO output, the greater the impact. The greater the capacity demand for horizontal power fluctuations, the higher the actual output of the wind farm needs to share the higher rental cost, and the greater its redistribution coefficient.

[0158] The specific expression for calculating the second redistribution sub-coefficient K2 is:

[0159]

[0160] 3) Correlation between WFOs.

[0161] The output correlation between WFOs will affect the power mutual assistance capability of wind farms. The smaller the output correlation between WFOs, the stronger the mutual assistance capability and the stronger the ability to smooth the power fluctuations of other wind farms. At this time, the smaller the capacity to be leased to CSESSO, the smaller the corresponding redistribution coefficient should be.

[0162] The specific expression for calculating the third redistribution sub-coefficient K3 is:

[0163]

[0164] In the formula, r mn 、r m represent the correlation coefficient and mean, respectively.

[0165] 4) Improve the Shapley value cost sharing strategy.

[0166] Set the weight of the apportionment coefficient and introduce the entropy weight method to calculate the weight of each redistribution coefficient. The greater the information entropy of the apportionment coefficient, the greater the weight ratio.

[0167] In S42, the redistribution coefficient K is calculated. m The specific expression is:

[0168] K m =θ1K1+θ2K2+θ3K3

[0169] In the formula, θ1, θ2 and θ3 all represent the weights of the apportionment coefficients;

[0170] Calculate the amortized result of the improved Shapley value The specific expression is:

[0171]

[0172] In the formula, ΔK m represents the allocation coefficient, M represents the number of wind farms, x m represents the allocation result of the original Shapley value; χ represents the control factor, which is 0.8 in this embodiment; and C represents the WFCO rental cost.

[0173] The S5 comprises the following sub-steps:

[0174] S51. Determine whether the rental income is greater than a preset income threshold;

[0175] If yes, the rental income is used as the income threshold, and it is determined whether the number of iterations is greater than the iteration threshold. If yes, the income threshold is used as the optimal rental income of CSESSO, and the process goes to S53. If no, the number of iterations is increased by 1, and the process goes to S52.

[0176] If not, proceed to S52;

[0177] S52, adjust the parameters of the CSESSO leasing service pricing model based on the master-slave game, determine the new leasing price and leasing capacity, repeat S4 to calculate the global optimal leasing income of the new CSESSO in each period, and return to S51;

[0178] S53. Output the calculated rental price and rental capacity corresponding to the optimal rental income of CSESSO, and complete the optimization of renting and hiring shared thermal energy storage in wind farms based on master-slave game.

[0179] In this embodiment, the CSESSO leasing service pricing model based on the master-slave game is solved by genetic algorithm nested with CPLEX commercial software. The main body CSESSO optimizes the leasing price for each time period through genetic algorithm, takes the leasing income as the fitness function, and uses CPLEX commercial software to solve the remaining capacity of CSESSO to be put into the spot market for trading and provide thermal energy for the heat load; the slave WFCO optimizes the leasing capacity for each time period through CPLEX commercial software to ensure the solution efficiency.

[0180] Embodiment 2:

[0181] This embodiment provides a multi-case comparative analysis experiment for Embodiment 1.

[0182] The master-slave game optimization framework proposed in this paper consists of one shared thermal energy storage power station and three wind farms. The effectiveness of the proposed optimization strategy is verified through simulation. Figure 2The predicted output curves of each wind farm on a typical day are given. In the figure, the horizontal axis is time in hours, and the vertical axis is power in megawatts. Figure 3 The predicted output of the wind farm cluster, power fluctuation, maximum power curve of the grid-connected power node, and the leasing demand of the wind farm cluster are given respectively. In the figure, the horizontal axis is time in hours, and the vertical axis is power in megawatts. Figure 4 The curves of electric load, thermal load and spot market clearing price are given. In the figure, the horizontal axis is time in hours, the left vertical axis is power in megawatts, and the right vertical axis is price in yuan / MWh. The electricity price of the distribution network is 1.25 times the spot market clearing price. The wind farm cluster abandoned wind power price is 1 / 2 times the clearing price. The references for the parameters of shared thermal energy storage are shown in Table 1.

[0183] Table 1 Relevant parameters of wind farm cluster and shared thermal energy storage

[0184]

[0185] The rationality of the master-slave game optimization operation strategy of CSESSO and WFCO proposed in this paper is verified by setting up comparative examples.

[0186] Case 1: CSESSO and WFCO trade directly at a pre-determined price, using an average price cap to settle WFCO’s leased capacity.

[0187] Case 2: WFO conducts transactions with CSESSO based on its own leasing needs, forming a master-slave game optimization model with one master and multiple slaves. CSESSO aggregates the leasing needs of various wind farms and optimizes the leasing prices for each period.

[0188] Case 3: Based on the master-slave game, the CSESSO leasing service pricing model, each wind farm reduces the demand for CSESSO leasing through mutual assistance.

[0189] The leasing costs, CSESSO benefits and WFCO benefits under the three cases are shown in Table 2. It can be found that compared with Case 1 and Case 2, Case 3 has the best CSESSO benefits and WFCO benefits have been improved to a certain extent. Compared with Case 1, Case 3 introduces the master-slave game strategy. It can be seen that the master-slave game can reduce the leasing price slightly to obtain more leasing demand from WFCO, indicating that the master-slave game can take into account the overall benefits of both CSESSO and WFCO and achieve fairness in market competition. Compared with Case 2, Case 3 takes into account the mutual assistance characteristics of each wind farm, reduces the leasing capacity of WFCO to CSESSO, increases the benefits of WFCO, and reduces the benefits of CSESSO by 8,400 yuan; however, the overall benefits of CSESSO have not decreased, because CSESSO obtains benefits by trading the remaining capacity with the power grid and heat users; at the same time, the strategy of Case 3 avoids the improper leasing demand of CSESSO by each wind farm, which is conducive to the fairness of the economic benefits of CSESSO and WFCO. Therefore, compared with Case 1 and Case 2, Case 3 has the best benefits. It increases WFCO benefits by reducing CSESSO leasing benefits, and realizes CSESSO's diversified benefit channels through the frequency regulation spot service market and heat users.

[0190] Table 2 Comparative analysis of the benefits of each entity

[0191]

[0192] The beneficial effects of the present invention are as follows: the present invention provides a method for optimizing the operation of shared thermal energy storage for leasing wind farms based on master-slave game, and realizes efficient utilization of new energy storage and new energy resources by leasing a shared thermal energy storage joint operation mechanism taking solid thermal storage into consideration through a wind farm cluster.

[0193] The present invention constructs a CSESSO leasing service pricing model based on master-slave game, with shared thermal energy storage operators as the main body and wind farm cluster alliances as the slaves. The price mechanism of shared thermal energy storage leasing services based on a two-part electricity price is introduced by referring to the frequency modulation market. The improved genetic algorithm is nested with CPLEX commercial software to solve the game equilibrium solution, determine the shared thermal energy storage leasing price and energy use strategy, and the follower leasing demand, so as to maximize the operating efficiency of each subject.

[0194] According to the uncertainty of wind farm clusters and dynamic leasing requirements, the present invention proposes a two-stage optimization operation strategy for shared thermal energy storage. The correlation between leasing members of each wind farm and the correlation between individual members and the alliance are fully considered, and the improved Shapley value is used for solution to achieve a reasonable allocation of leasing costs for each member.

[0195] In the description of the present invention, it is necessary to understand that the orientation or positional relationship indicated by the terms "center", "thickness", "upper", "lower", "horizontal", "top", "bottom", "inner", "outer", "radial", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", and "third" are used only for descriptive purposes, and cannot be understood as indicating or implying the relative importance or the number of implicitly specified technical features. Therefore, the features defined by "first", "second", and "third" may explicitly or implicitly include one or more of the features.

Claims

1. A method for optimizing the operation of wind farm leasing and sharing thermal energy storage based on master-slave game, characterized in that: The following steps are involved: S1. Establish a WFCO optimization model, and optimize the leasing capacity by responding to the leasing price through the WFCO optimization model; S2. Establish a CSESSO optimization model, and adjust the lease price according to the optimized lease capacity through the CSESSO optimization model; S3. According to the game relationship between the WFCO optimization model and the CSESSO optimization model, a CSESSO leasing service pricing model based on the master-slave game is established to determine the leasing price and leasing capacity, and initialize the number of iterations; S4. Use the improved Shapley value to optimize the rental cost allocation of the CSESSO rental service pricing model based on the master-slave game, calculate the global optimal rental income of CSESSO in each period according to the allocated rental cost, rental capacity and rental price, and record the rental price and rental capacity in each period; S5. Set the income threshold and iteration threshold, and repeat S4 according to the relationship between the calculated rental income and the income threshold until the number of iterations reaches the iteration threshold, so as to obtain the optimal rental income of CSESSO and complete the optimization of renting and sharing thermal energy storage in wind farms based on master-slave game.

2. The wind farm leasing and sharing thermal energy storage optimization operation method based on master-slave game according to claim 1 is characterized in that: In S1, the leasing capacity demand is optimized according to the calculated total WFCO revenue, and the expression of the objective function of the WFCO optimization model is specifically: In the formula, I W represents the total revenue of WFCO; They represent the WFCO grid-connected power revenue and the wind power sales revenue respectively; C L , C b They represent the leasing cost and the grid-connected fluctuation assessment cost respectively; p t , P t ADN,w They represent WFCO on-grid electricity price and grid-connected power respectively; t represents time; ε represents the unit grid-connected power fluctuation assessment cost; P t , P t-1 They represent the predicted output of WFCO at time t and the predicted output of WFCO at time t-1 respectively; P limit It indicates the upper limit of power fluctuation during the process of wind farm grid connection; Δt indicates the time interval; The constraints of the WFCO optimization model include WFCO output constraints, WFO output constraints, and WFCO wind power curtailment constraints; The expression of the WFCO output constraint is specifically: 0≤P t ADN,w ≤P t L Where P t L It indicates the maximum power that the WFCO grid-connected node can accommodate; The expression of the WFO output constraint is specifically: In the formula, They represent the dispatch output and forecast output of the mth WFO respectively; The expression of the WFCO wind power curtailment constraint is specifically: 0≤P t W,Q,c ≤max(P t F -P t L ,0) Where P t F Indicates the forecast output of WFCO.

3. The wind farm leasing and sharing thermal energy storage optimization operation method based on master-slave game according to claim 2 is characterized in that: In S2, the rental price is adjusted according to the calculated total operating income of CSESSO. The expression of the objective function of the CSESSO optimization model is specifically: In the formula, max represents the maximum value, I CSESSO represents the total operating revenue of CSESSO, Respectively represent the rental income, frequency regulation service income, and heating service income of CSESSO; C loss , C w Respectively represent the operating cost of CSESSO and the cost of wind power purchase. C loss , C w The specific expression is: In the formula, They represent the capacity rental price and electricity rental price of CSESSO respectively; L t , H t Respectively represent the rental capacity and rental mileage of WFCO; P t ADN,d , P t ADN,c They represent the interaction power between CSESSO and the main grid, and the grid-connected power respectively; They represent the heating power price and grid-connected electricity price of the interaction between CSESSO and the main grid respectively; κ t , They represent the CSESSO heating electricity price and heating power respectively; γ represents the CSESSO operating cost, represents the heat loss of the shared thermal energy storage at time t; They represent the wind power price and wind power amount abandoned by WFCO respectively; The constraints of the CSESSO optimization model include the electric and thermal power balance constraint and the CSESSO operation constraint. The expression of the electric and thermal power balance constraint is specifically: P t ADN,d ⊥P t ADN,c =0 Where P t W represents the leased power of the wind power cluster, α represents the heating power and heating efficiency of the CSESSO resistance wire respectively; ⊥ is an operator, indicating that at least one of the parameters is 0; The expression of the operation constraint of the CSESSO is specifically: In the formula, μ min , μ max They respectively represent the lower limit and upper limit of the heat load inertia adjustment ratio; represents the heat load dispatched by CSESSO; Respectively represent the maximum efficiency of CSESSO charging and heating.

4. The wind farm leasing and sharing thermal energy storage optimization operation method based on master-slave game according to claim 3 is characterized in that: In S3, the expression of the CSESSO rental service pricing model G based on the master-slave game is specifically: G={N;S {CSESSO∪WFCO };I {CSESSO∪W }} Where N represents the set of participants, and N = {CSESSO ∪ WFCO}, where CSESSO is the principal participant, WFCO is the subordinate participant, S {CSESSO∪WFCO} Represents a policy set, which includes the principal CSESSO policy set and the slave WFCO policy set. {CSESSO∪W} represents the benefits, which include the total benefits of WFCO and the total operating benefits of CSESSO; Subject CSESSO strategy set S CSESSO The specific expression is: Slave WFCO policy set S WFCO The specific expression is: In the formula, R t represents the lease capacity, Indicates the grid-connected power, represents the revenue from curtailed wind power sales, C L It represents the grid-connected fluctuation assessment cost; The constraint functions of the CSESSO leasing service pricing model based on the master-slave game include WFCO game constraints and CSESSO game constraints; The expression of the WFCO game constraint is specifically: In the formula, Assess the cost of grid connection fluctuation for WFCO; The expression of the CSESSO game constraint is specifically: In the formula, is the maximum price difference in the spot market, and ψ is the conversion factor.

5. The wind farm leasing and sharing thermal energy storage optimization operation method based on master-slave game according to claim 4 is characterized in that: In S4, a method for optimizing the leasing cost allocation of the CSESSO leasing service pricing model based on the master-slave game using the improved Shapley value is used. The following steps are included: S41, using the equal power-successive load method to calculate the equivalent output of each wind farm, calculating the waveform similarity between the equivalent output and the actual output curve according to the cosine similarity algorithm, and then calculating the first to third redistribution sub-coefficients; S42. Calculate the redistribution coefficient according to the first to third redistribution sub-coefficients, and then establish a rental cost allocation model based on the improved Shapley value, calculate the allocation result of the improved Shapley value, and complete the rental cost allocation.

6. The wind farm leasing and sharing thermal energy storage optimization operation method based on master-slave game according to claim 5 is characterized in that: In S41, the waveform similarity S between the equivalent output curve and the actual output curve is calculated. i,wave The specific expression is: In the formula, represents the equivalent output of each wind farm, represents the predicted output of the mth wind farm in the wind farm cluster in time period t, i represents the ordinal number of the wind farm; The specific expression for calculating the first redistribution sub-coefficient K1 is: The specific expression for calculating the second redistribution sub-coefficient K2 is: The specific expression for calculating the third redistribution sub-coefficient K3 is: In the formula, r mn 、r m represent the correlation coefficient and mean, respectively.

7. The method for optimizing the operation of wind farm leasing and sharing thermal energy storage based on master-slave game according to claim 6 is characterized in that: In S42, the redistribution coefficient K is calculated. m The specific expression is: K m =θ1K1+θ2K2+θ3K3 In the formula, θ1, θ2 and θ3 all represent the weights of the apportionment coefficients; Calculate the amortized result of the improved Shapley value The specific expression is: In the formula, ΔK m represents the allocation coefficient, M represents the number of wind farms, x m represents the allocation result of the original Shapley value; χ represents the control factor; C represents the WFCO rental cost.

8. The wind farm leasing and sharing thermal energy storage optimization operation method based on master-slave game according to claim 7 is characterized in that: The S5 comprises the following sub-steps: S51. Determine whether the rental income is greater than a preset income threshold; If yes, the rental income is used as the income threshold, and it is determined whether the number of iterations is greater than the iteration threshold. If yes, the income threshold is used as the optimal rental income of CSESSO, and the process goes to S53. If no, the number of iterations is increased by 1, and the process goes to S52. If not, proceed to S52; S52, adjust the parameters of the CSESSO leasing service pricing model based on the master-slave game, determine the new leasing price and leasing capacity, repeat S4 to calculate the global optimal leasing income of the new CSESSO in each period, and return to S51; S53. Output the calculated rental price and rental capacity corresponding to the optimal rental income of CSESSO, and complete the optimization of renting and hiring shared thermal energy storage in wind farms based on master-slave game.

Images