Comprehensive energy system optimal configuration method and system considering medium and long term characteristics of hydrogen energy storage

Abstract

The invention relates to the technical field of comprehensive energy optimal configuration, in particular to a comprehensive energy system optimal configuration method and system considering medium and long term characteristics of hydrogen energy storage, and the method comprises the steps: constructing a multi-energy coupling mathematical model of a rural comprehensive energy system; a typical day containing multi-dimensional data is generated by a typical day generation method based on spectrum joint clustering, a hydrogen energy storage long-period operation model is constructed, and the hydrogen storage amount at any moment in the whole-year operation period is decomposed into dynamic superposition of the initial hydrogen storage amount and a typical day scene; constructing a double-layer optimization configuration model comprising capacity configuration and scheduling operation, wherein the double-layer optimization configuration model comprises an upper-layer capacity configuration model and a lower-layer scheduling operation model; and solving the double-layer optimal configuration model by adopting a Benders decomposition method, splitting the double-layer optimal configuration model into a main problem and a sub-problem, and carrying out iterative approximation to an optimal solution by utilizing a plane cutting method. The system capacity configuration is optimized, the solving efficiency is improved, and the consumption capability of renewable energy sources and the overall flexibility of the system are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of integrated energy optimization configuration, and particularly relates to a method and system for integrated energy optimization configuration considering medium and long term characteristics of hydrogen energy storage. BACKGROUND

[0002] With the transformation of energy structure, the sustainable development of rural energy system faces new opportunities and challenges. The traditional rural energy supply mode often relies on single energy, which has problems such as low energy utilization efficiency, environmental pollution and low energy supply reliability. Integrated energy system (IES) realizes the cascade utilization and collaborative optimization of energy by integrating various forms of energy, which is an important way to improve energy utilization efficiency, promote renewable energy consumption and ensure energy supply safety.

[0003] Hydrogen energy, as a clean and efficient secondary energy, plays an increasingly important role in building a new energy system. Hydrogen energy storage has the advantages of large storage capacity, long storage period and high energy density, which can effectively solve the intermittency and volatility of renewable energy (such as photovoltaic), and realize the transfer of energy in time and space. The introduction of hydrogen energy and its related conversion equipment (such as electrolytic tank, fuel cell) into rural integrated energy system is expected to further improve the economy, environmental protection and reliability of the system.

[0004] However, the optimization configuration of rural integrated energy system is a complex decision-making process, which needs to consider the selection and capacity configuration of various energy devices, as well as the operation strategy of the system at different time scales. In particular, the long-period operation characteristics of hydrogen energy storage (such as cross-seasonal energy storage) bring new challenges to the optimization modeling and solving of the system. How to effectively build an optimization model containing multiple energy coupling characteristics and long-period operation characteristics of hydrogen energy storage, and use efficient solving algorithm to realize the economic, environmental protection and reliable operation of rural integrated energy system is the focus and difficulty of current research.

[0005] The information disclosed in this BACKGROUND section is only intended to enhance the understanding of the general background of the application, and should not be construed as recognition or any form of suggestion that this information constitutes prior art that is known to those skilled in the art. SUMMARY

[0006] The present application provides a method and system for integrated energy optimization configuration considering medium and long term characteristics of hydrogen energy storage, thereby effectively solving the problems in the background art.

[0007] In order to achieve the above purpose, the technical scheme adopted by the present application is as follows: a method for integrated energy optimization configuration considering medium and long term characteristics of hydrogen energy storage, comprising the following steps:

[0008] a mathematical model of rural integrated energy system including hydrogen energy storage is constructed;

[0009] The typical day containing multi-dimensional data is generated by a typical day generation method based on spectral joint clustering, a hydrogen storage long-period operation model is constructed, and the hydrogen storage amount at any time in the annual operation period is decomposed into the initial hydrogen storage amount and the dynamic superposition of the typical day scenario;

[0010] Based on the multi-energy coupling mathematical model and the hydrogen storage long-period operation model, a double-layer optimization configuration model containing capacity configuration and scheduling operation is constructed, the double-layer optimization configuration model includes an upper-layer capacity configuration model and a lower-layer scheduling operation model, the upper-layer capacity configuration model takes the minimum equal annual value cost in the whole life cycle as the optimization objective, and takes the capacity of each device as the decision variable, and the lower-layer scheduling operation model takes the minimum annual operation cost as the optimization objective, and takes the output of each device and the energy storage operation strategy as the decision variable;

[0011] The double-layer optimization configuration model is solved by using a Benders decomposition method, the double-layer optimization configuration model is split into a master problem and a sub-problem, the optimal solution is iteratively approximated by using a cut plane method, and the capacity configuration and the scheduling operation strategy are obtained.

[0012] Further, the multi-energy coupling mathematical model includes:

[0013] The mathematical model of the energy supply device includes a power generation model of a distributed photovoltaic unit, a waste power supply model, and a natural gas power model of a biogas tank;

[0014] The mathematical model of the energy conversion device includes a hydrogen production model of an electrolytic cell, a hydrogen fuel cell power model, and a combined heat and power unit output model;

[0015] The mathematical model of the energy storage device includes a charge-discharge model of an energy storage battery, a charge-discharge heat model of a heat storage tank, and a charge-discharge hydrogen model of a hydrogen storage tank.

[0016] Further, the mathematical model of the energy supply device includes:

[0017] The power generation model of the distributed photovoltaic unit includes:

[0018]

[0019] In the formula, P PV (t) is the actual power generation of the photovoltaic unit at the t period; P STC is the rated output power of the photovoltaic unit; G ING (t) is the solar radiation intensity at the t period; G STC is the solar radiation intensity under a standard environment; k PV is a temperature coefficient; T c (t) is the surface temperature of the photovoltaic panel; T r is a reference temperature;

[0020] Garbage power generation model:

[0021] H WI (t) = W gar (t) α WI β WI q WI η WI ;

[0022] H WG (t) = W gar (t) α WG q WG η WG ;

[0023] H gar (t) = H WI (t) + H WG (t);

[0024] P gar (t) = (H WI (t) + H WG (t)) η WHB (1 - η aux );

[0025] In the formula: H WI (t) and H WG (t) are the heat power generated by garbage incineration and gasification at time t, respectively; H gar (t) is the heat power generated by garbage at time t; P gar (t) is the garbage power generation at time t; W gar (t) is the amount of garbage treated at time t; α WI and α WG are the proportions of incinerable garbage and gasifiable garbage in household garbage, respectively; β WI is the proportion of incinerable garbage remaining after the leachate is discharged from the storage pool; q WI and q WG are the low-grade heat values of incinerable garbage and gasifiable garbage, respectively; η WI , η WG , η WHB , η aux are the incinerator efficiency, pyrolysis gasifier thermal efficiency, turbine power generation efficiency, and auxiliary power rate, respectively;

[0026] Natural gas power model of the biogas tank:

[0027] G BD (t) = V BD (t) η B2G q CH4 ;

[0028] In the formula: GBD (t) is the natural gas power obtained after purification treatment of biogas; η B2G is the purification coefficient of biogas for making natural gas; q CH4 is the natural gas heat value, V BD (t) is the gas production model of the biogas tank;

[0029] V BD (t) = a BD M BD (t) + b BD H BD (t) ;

[0030] H BD (t) = (T z (t + 1) - T z (t)) V B c m p m + H loss (t) ;

[0031] H loss (t) = c m M BD (t) (T z (t) - T e ) + k o A o (T z (t) - T e ) ;

[0032] wherein: a BD is the gas production coefficient of unit organic matter; b BD is the gas production coefficient of unit heat energy; H BD (t) is the heat power supplied to the biogas tank at time t; H loss (t) is the heat dissipation power of the biogas tank at time t; V B is the volume of the biogas tank; c m is the heat capacity of the feed; p m is the density of the feed; M BD (t) is the mass of the organic waste fed into the biogas tank at time t; A o is the heat dissipation area in the tank; k o is the heat transfer coefficient; T e is the external temperature of the biogas tank.

[0033] Further, the mathematical model of the energy conversion device comprises:

[0034] Electrolytic tank hydrogen production model:

[0035]

[0036] wherein: m EL(t) is the hydrogen mass produced by the electrolyzer in the time period t; η EL,H2 is the hydrogen production efficiency of the electrolyzer plant; P EL (t) is the power consumed by the electrolyzer in the time period t; q H is the hydrogen heating value;

[0037] Electrolyzer heat recovery model:

[0038] H EL (t) = η EL,h P EL,H2 (t) q EL (t);

[0039] where: H EL (t) is the heat power recovered by the electrolyzer in the time period t; η EL,h is the heat exchange efficiency of the heat exchanger;

[0040] Hydrogen fuel cell power model:

[0041] P FC (t) = η FC,e m FC (t) q H ;

[0042] where: P FC (t) is the power generated by the hydrogen fuel cell in the time period t; η FC,e is the hydrogen fuel cell power generation efficiency; m FC (t) is the hydrogen mass consumed by the hydrogen fuel cell in the time period t;

[0043] Hydrogen fuel cell heat recovery model:

[0044] H FC (t) = η FC,h m FC (t) q H ;

[0045] where: H FC (t) is the heat power generated by the hydrogen fuel cell in the time period t; η FC,h is the hydrogen fuel cell heat generation efficiency;

[0046] Cogeneration unit output model:

[0047]

[0048] where: G CHP (t), P CHP (t), H CHP (t) are the input gas power, output electric and heat power of the CHP unit in the time period t, respectively; η CHP , α CHP are the electric conversion efficiency and thermal energy efficiency ratio of the CHP unit, respectively.

[0049] Further, the mathematical model of the energy storage device comprises:

[0050] Charging and discharging model of energy storage battery:

[0051]

[0052] wherein S ES (t) represents the amount of electricity stored by the energy storage battery at time period t; σ ES represents the self-discharge rate of the energy storage battery; η ES represents the charging and discharging efficiency of the energy storage battery; P ES,ch (t) and P ES,dis (t) represent the charging and discharging power of the energy storage battery at time period t, respectively; Δt represents the time span from t to t+1.

[0053] Charging and discharging model of heat storage tank:

[0054]

[0055] wherein S HS (t) represents the amount of heat stored by the heat storage tank at time period t; σ HS represents the self-discharge rate of the heat storage tank; η HS represents the charging and discharging efficiency of the heat storage tank; H HS,ch (t) and H HS,dis (t) represent the charging and discharging power of the heat storage tank at time period t, respectively.

[0056] Charging and discharging model of hydrogen storage tank:

[0057]

[0058] wherein S H2S (t) represents the amount of hydrogen stored by the hydrogen storage tank at time period t; σ H2S represents the self-discharge rate of the hydrogen storage tank; η H2S represents the charging and discharging efficiency of the hydrogen storage tank; m H2S,ch (t) and m H2S,dis (t) represent the charging and discharging amount of hydrogen of the hydrogen storage tank at time period t, respectively.

[0059] Further, the typical day generation method based on spectral joint clustering generates a typical day containing multi-dimensional data, comprising the following steps:

[0060] Let the original data matrix be X = (xij)m×n wherein m represents the number of days, and n represents the variable dimension; construct a row similarity matrix W = (wij)m×m describe the multivariate time series similarity between days:

[0061]

[0062] where x i is the multivariate time series feature vector of the i-th day, x j is the multivariate time series feature vector of the j-th day, and σ is the bandwidth parameter of the Gaussian kernel, controlling the speed of similarity decay.

[0063] Construct the column similarity matrix Describe the association between variable-hour joint features:

[0064]

[0065] where k and l are column indices, corresponding to different variable-hour combinations; x :,k is the k-th column of the data matrix, representing the data of all days at a certain variable at a certain hour; x :,l is the l-th column of the data matrix, corresponding to another different variable-hour combination.

[0066] Calculate the normalized Laplacian matrix for rows and columns respectively:

[0067]

[0068] where L r , L c are the row and column normalized Laplacian matrices respectively; D r , D c are the row and column degree matrices respectively.

[0069] Perform singular value decomposition on the double-normalized matrix X T L r X+XL c X T , and select the first s eigenvectors to form the low-dimensional embedding space; perform K-means clustering on the rows and columns of the low-dimensional space respectively to obtain the row cluster C r and the column cluster C c , and the objective function is to minimize the residual error between the row and column clusters:

[0070]

[0071] where μ Cr,Cc represents the row-column cluster interaction center; X(i,j) directly represents the value of the i-th time point, j-th variable-hour feature in the original data matrix, compared with the interaction center.

[0072] For each row cluster C k , calculate the multivariate time series mean of all days it contains as the typical day:

[0073]

[0074] where representing the average value of the h hour, the v variable in the k typical day, characterizing the common mode of the cluster; x (d,h,v) representing the specific value of the d day, the h hour, the v variable in the original data;

[0075] According to the proportion of samples in the cluster, the typical day probability is calculated:

[0076]

[0077] Based on the selected typical day, the corresponding photovoltaic output, electrical load demand, thermal load demand, gas load demand, and hydrogen load demand data are extracted, and s typical day data are output; each typical day retains the photovoltaic output data and the coupling characteristics of electrical, thermal, and hydrogen multi-energy loads for 24 hours within the day.

[0078] Further, the construction of the hydrogen energy storage long-period operation model comprises:

[0079] In a typical day, the relationship between the hydrogen storage amounts of two consecutive time points is:

[0080]

[0081] ΔS H2S (k,0)=0;

[0082] In the formula: ΔS H2S (k,t) represents the hydrogen mass stored by the hydrogen storage tank at time t of the k typical day, kg, and the initial time within the day is set to 0, which is irrelevant to the actual hydrogen storage amount of the hydrogen storage tank; m H2S,ch (k,t) and m H2S,dis (k,t) respectively represent the hydrogen charging amount and the hydrogen discharging amount of the hydrogen storage tank at time t of the k typical day;

[0083] Between two consecutive operation days, the initial hydrogen storage mass of the hydrogen storage tank needs to satisfy:

[0084] S H2S,0 (r+1)=(1-σ H2S ×24)S H2S,0 (r)+ΔS H2S (k=h(r),24);

[0085] In the formula: S H2S,0 (r+1) represents the initial hydrogen storage mass of the hydrogen storage tank in the r+1 operation day; h(r) is a mapping function between the typical day k and the operation day r;

[0086] In order to ensure that the hydrogen storage amount of the hydrogen storage tank is the same within a year, it needs to satisfy:

[0087] S H2S,0 (1)=(1-σ H2S ×24)SH2S,0 (N r )+ΔS H2S (k=h(N r ),twenty four);

[0088] Where: N r This refers to the number of operating days throughout the year.

[0089] The actual hydrogen storage capacity of the hydrogen storage tank at any time throughout the year can be decomposed into the dynamic superposition of the initial hydrogen storage capacity and a typical daily scenario, expressed as:

[0090] S H2S (r,t)=S H2S,0 (r)+ΔS H2S (k = h(r), t);

[0091] In the formula: S H2S (r,t) represents the actual mass of hydrogen stored in the hydrogen storage tank at time t on operating day r, which is the sum of the initial hydrogen storage on operating day r and the mass of hydrogen stored at time t on operating day r.

[0092] The hydrogen stored in the hydrogen storage tank must meet upper and lower limit constraints:

[0093] α H2S,min S H2S,cap ≤S H2S (r,t)≤α H2S,max S H2S,cap ;

[0094] Since the above formula holds true at every moment within the scheduling period, the following equivalent processing is performed to reduce constraints:

[0095]

[0096] In the formula: ΔS H2S,max (k) and ΔS H2S,min (k) represents the maximum and minimum mass of hydrogen stored in the hydrogen storage tank on k typical days, in kg.

[0097] Furthermore, the objective function of the upper-level capacity configuration model, which aims to minimize the annual cost over the entire lifecycle, is as follows:

[0098] minC=C inv +C op +C scrap +C loss ;

[0099] In the formula, C inv For the equivalent annual value of investment cost, C op For operating costs, C scrap For scrapping costs, C loss For loss and damage costs;

[0100]

[0101] wherein: y represents the service life of the equipment; r is the discount rate; μ i is the unit capacity investment cost of the i-th equipment; S i represents the installed capacity of the i-th equipment; X represents a set of equipment in the system for which capacity is configured, including waste power generation, biogas fermentation, electrolytic tank, hydrogen fuel cell, cogeneration unit, energy storage battery, heat storage tank, and hydrogen storage tank;

[0102]

[0103] wherein: s is the total number of typical days in a year; p k is the probability of k typical days occurring in a year; N r is the number of operating days in a year; C om,k is the typical day operation and maintenance cost, C buy,k is the energy purchase cost, C CO2,k is the carbon trading cost, C waste,k is the waste treatment subsidy, C sell,k is the energy sale income.

[0104] Further, the upper layer capacity configuration model further includes the following constraints:

[0105] Equipment installed capacity limit constraint:

[0106] 0≤S i ≤S i,max ;

[0107] wherein: S i represents the installed capacity of the i-th equipment, S i,max represents the upper limit of the capacity that the i-th equipment can allow to be installed.

[0108] Further, the lower layer scheduling operation model has the following objective function with the minimum annual operation cost as the optimization objective:

[0109] min C op =C om +C buy +C CO2 -C waste -C sell ;

[0110] The objective function includes the annual operation and maintenance cost C om of the system, the energy purchase cost C buy , the carbon trading cost C CO2 , the waste treatment subsidy C waste , and the energy sale income C sell ;

[0111] Wherein:

[0112]

[0113] Wherein: λ i is the unit capacity operation and maintenance cost of equipment i; P i (k, t) represents the output power of the i-th equipment at typical day k time period t; X represents the set of equipment in the system, including photovoltaic units, waste power generation, biogas fermentation, electrolytic cells, hydrogen fuel cells, cogeneration units, electric boilers, energy storage batteries, heat storage tanks and hydrogen storage tanks;

[0114]

[0115] Wherein: P buy (k, t) and G buy (k, t) are the purchased electricity power and natural gas power at typical day k time period t, respectively;

[0116]

[0117] Wherein: W gar (k, t) is the amount of waste treated at typical day k time period t; M BD (r, t) is the mass of organic waste input into the biogas tank at typical day k time period t;

[0118]

[0119] Wherein: m sell (k, t) is the amount of hydrogen sold by the system at typical day k time period t.

[0120] Further, the lower layer scheduling operation model further comprises the following constraints:

[0121] Electric power balance constraint:

[0122] P buy (r, t) + P PV (r, t) + P gar (r, t) + P FC (r, t) + P CHP (r, t) + P ES,dis (r, t);

[0123] = P EL (r, t) + P EB (r, t) + P ES,ch (r, t) + P L (r, t)

[0124] Wherein: P L (r, t) is the electrical load of the system at r operating day t time period, P buy ​(r, t) is the purchased electricity power of r operation day t period, P PV (r, t) is the actual power generation of photovoltaic unit of r operation day t period; P gar (r, t) is the waste power generation of r operation day t period; P FC (r, t) is the hydrogen fuel cell power generation of r operation day t period; P CHP (r, t) is the input gas power of CHP unit of r operation day t period, P ES,ch (r, t) and P ES,dis (r, t) respectively represent the charging and discharging power of energy storage battery r operation day t period; P EL (r, t) is the electrolytic cell power consumption of r operation day t period;

[0125] Thermal power balance constraint:

[0126] H gar (r, t) + H CHP (r, t) + H EL (r, t) + H FC (r, t) + H EB (r, t) + H HS,dis (r, t);

[0127] = H BD (r, t) + H HS,ch (r, t) + H L (r, t)

[0128] In the formula, H L (r, t) is the thermal load of the system of r operation day t period; H gar (r, t) is the waste heat generation power of r operation day t period; H FC (r, t) is the hydrogen fuel cell heat generation power of r operation day t period; H CHP (r, t) is the thermal power of CHP unit of r operation day t period, H HS,ch (r, t) and H HS,dis (r, t) respectively represent the heat storage and release power of r operation day t period; H EL (r, t) is the heat recovery power of electrolytic cell of r operation day t period; H BD (r, t) is the heat supply to the biogas tank of r operation day t period;

[0129] Natural gas balance constraint:

[0130] G buy (r, t) + G BD (r, t) = G CHP (r, t) + G L (r, t);

[0131] G L (r,t) is the gas load of the system at time period t on day r, G buy (r,t) is the purchased natural gas power at time period t on day r; G BD (r,t) is the natural gas power obtained after purification treatment of biogas at time period t on day r; G CHP (r,t) is the input gas power of the CHP unit at time period t on day r;

[0132] Hydrogen energy balance constraint:

[0133] m EL (r,t) + m H2S,dis (r,t) = m FC (r,t) + m sell (r,t) + m H2S,ch (r,t) + m L (r,t);

[0134] wherein m L (r,t) is the hydrogen load of the system at time period t on day r; m EL (r,t) is the hydrogen mass produced by the electrolyzer at time period t on day r; m H2S,ch (r,t) and m H2S,dis (r,t) respectively represent the hydrogen charging amount and hydrogen discharging amount of the hydrogen storage tank at time period t on day r; m FC (r,t) is the hydrogen mass consumed by the hydrogen fuel cell at time period t on day r; m sell (r,t) is the hydrogen sales amount of the system at time period t on day r;

[0135] System energy purchase constraint:

[0136] P buy,min ≤ P buy (t) ≤ P buy,max ;

[0137] G buy,min ≤ G buy (r,t) ≤ G buy,max ;

[0138] Each device operation constraint, as described above, can be mapped to the constraint of a typical day k.

[0139] Further, the Benders decomposition method is used to solve the double-layer optimization configuration model, the double-layer optimization configuration model is split into a master problem and a sub-problem, and a cutting plane method is used to iteratively approximate the optimal solution, including the following steps:

[0140] Step 1: initialize the decision variable x 0 of the master problem, set the initial lower bound LB = E Tx 0 , set a convergence threshold value epsilon, set iteration count i = 0;

[0141] Step 2: in the i-th iteration, solve the main problem, i.e. the upper configuration problem, to obtain the current device capacity x i , update the lower bound LB = E T x i , and pass it to the subproblem;

[0142] Step 3: given the device capacity x i , solve the feasible subproblem to obtain Solve the convex feasible subproblem to obtain

[0143] Step 4: if the optimal value of the problem optimization objective is not equal to 0, i.e. the subproblem is infeasible, generate a feasibility cut plane, constrain the search space of the main problem, so that subsequent iterations avoid infeasible solutions, update the iteration number i = i + 1 and return to step 2;

[0144] If the optimal value of the problem optimization objective is equal to 0, i.e. the subproblem is feasible, solve the optimal subproblem to obtain And Generate an optimality cut plane, add it to the constraints of the main problem, and update the upper bound UB according to the formula;

[0145] Step 5: if the upper and lower bounds satisfy the convergence condition |UB-LB|<epsilon, stop iteration and output the optimal solution; if not converged, update the iteration number i = i + 1 and return to step 2 to continue solving the main problem;

[0146] Step 6: output the result. Finally output the optimal device capacity configuration x * .

[0147] The application also includes a hydrogen energy storage medium and long-term characteristic comprehensive energy system optimization configuration system, which uses the method as described above, and the system comprises:

[0148] A mathematical model construction unit is configured to construct a multi-energy coupling mathematical model of a rural comprehensive energy system including hydrogen energy storage;

[0149] A typical day generation unit is configured to generate a typical day containing multi-dimensional data based on a spectrum joint clustering typical day generation method, construct a hydrogen energy storage long-period operation model, and decompose the hydrogen storage amount at any time within the annual operation period into an initial hydrogen storage amount and a dynamic superposition of a typical day scenario;

[0150] A double-layer model construction unit is configured to construct a double-layer optimization configuration model including capacity configuration and dispatching operation based on the multi-energy coupling mathematical model and the hydrogen storage energy long-period operation model, the double-layer optimization configuration model including an upper-layer capacity configuration model and a lower-layer dispatching operation model, the upper-layer capacity configuration model taking minimum life cycle equivalent annual cost as an optimization objective and taking each device capacity as a decision variable, and the lower-layer dispatching operation model taking minimum annual operation cost as an optimization objective and taking each device output and energy storage operation strategy as a decision variable.

[0151] A solution unit is configured to solve the double-layer optimization configuration model by using a Benders decomposition method, split the double-layer optimization configuration model into a master problem and a sub-problem, and obtain capacity configuration and dispatching operation strategy by using a cut plane method to iteratively approximate an optimal solution.

[0152] The application further includes a computer device including a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor implements the method as described above when executing the computer program.

[0153] The application further includes a storage medium having a computer program stored thereon, and the computer program is executable on a processor to implement the method as described above.

[0154] The application has the beneficial effects that: by introducing hydrogen energy and its coupling devices, using a typical day generation method based on spectral joint clustering and a hydrogen storage energy long-period operation model, the operation state of a rural integrated energy system can be more accurately simulated, system capacity configuration can be optimized, the calculation complexity of the model is effectively reduced, the solving efficiency is improved, and the renewable energy consumption capacity and the overall flexibility of the system are effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0155] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments described in the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0156] Figure 1 It is a flowchart of the method in embodiment 1.

[0157] Figure 2 It is a structural schematic diagram of the system in embodiment 1.

[0158] Figure 3 It is a framework diagram of the integrated energy system in embodiment 2.

[0159] Figure 4This is a schematic diagram of the long-term hydrogen energy storage modeling method in Example 2;

[0160] Figure 5 Here is a framework diagram for the two-layer model configuration in Example 2;

[0161] Figure 6 This is a schematic diagram of the structure of the computer device of the present invention. Detailed Implementation

[0162] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0163] Example 1:

[0164] like Figure 1 As shown: A comprehensive energy system optimization configuration method considering the medium- and long-term characteristics of hydrogen energy storage, comprising the following steps:

[0165] Construct a multi-energy coupling mathematical model for a rural integrated energy system that includes hydrogen energy storage;

[0166] A typical day generation method based on spectrum joint clustering generates typical days containing multidimensional data, constructs a long-term operation model for hydrogen energy storage, and decomposes the hydrogen storage at any time during the annual operation cycle into a dynamic superposition of the initial hydrogen storage and typical day scenarios.

[0167] Based on the multi-energy coupling mathematical model and the long-cycle operation model of hydrogen energy storage, a two-layer optimization configuration model is constructed, which includes capacity configuration and scheduling operation. The two-layer optimization configuration model includes an upper-layer capacity configuration model and a lower-layer scheduling operation model. The upper-layer capacity configuration model takes the minimum annual cost over the whole life cycle as the optimization objective and the capacity of each device as the decision variable. The lower-layer scheduling operation model takes the minimum annual operating cost as the optimization objective and the output of each device and the energy storage operation strategy as the decision variables.

[0168] The Benders decomposition method is used to solve the bi-level optimization configuration model, which is divided into a main problem and sub-problems. The cutting plane method is used to iteratively approximate the optimal solution to obtain the capacity configuration and scheduling operation strategy.

[0169] By introducing hydrogen energy and its coupling devices, and adopting a typical daily generation method based on spectral joint clustering and a long-term operation model for hydrogen energy storage, the operating status of rural integrated energy systems can be simulated more accurately, the system capacity configuration can be optimized, the computational complexity of the model can be effectively reduced, the solution efficiency can be improved, and the absorption capacity of renewable energy and the overall flexibility of the system can be effectively enhanced.

[0170] In this embodiment, the multi-energy coupling mathematical model includes:

[0171] Mathematical models of energy supply equipment: power generation model of distributed photovoltaic units, energy supply model of waste-to-energy power generation, and natural gas power model of biogas digesters;

[0172] Mathematical models of energy conversion equipment: hydrogen production model of electrolyzer, power model of hydrogen fuel cell, and output model of combined heat and power unit;

[0173] Mathematical models of energy storage devices: energy storage battery charging and discharging model, thermal storage tank charging and discharging model, and hydrogen storage tank charging and discharging model.

[0174] The mathematical models of energy supply equipment include:

[0175] Power generation model of distributed photovoltaic units:

[0176]

[0177] In the formula: P PV (t) represents the actual power generation of the photovoltaic unit during time period t; P STC G is the rated output power of the photovoltaic unit. ING (t) represents the solar radiation intensity during time period t; G STC The solar radiation intensity under standard conditions; k PV T is the temperature coefficient; c (t) represents the surface temperature of the photovoltaic panel; T r For reference temperature;

[0178] Waste-to-energy power generation model:

[0179] H WI (t)=W gar (t)α WI β WI q WI η WI ;

[0180] H WG (t)=W gar (t)α WG q WG η WG ;

[0181] H gar (t)=H WI (t)+H WG (t);

[0182] P gar (t)=(H WI (t)+H WG (t))η WHB (1-η aux );

[0183] Where: HWI (t) and H WG (t) are the heat power generated by the incineration and gasification of domestic waste respectively in the period t; H gar (t) is the heat power generated by the waste in the period t; P gar (t) is the power generated by the waste in the period t; W gar (t) is the amount of waste treated in the period t; α WI and α WG are the proportions of incinerable waste and gasifiable waste in the domestic waste respectively; β WI is the proportion of incinerable waste remaining after the discharge of the leachate from the storage pool; q WI and q WG are the low-grade calorific values of the incinerable waste and the gasifiable waste respectively; η WI , η WG , η WHB , η aux are the efficiency of the incinerator, the thermal efficiency of the pyrolysis gasifier, the efficiency of the steam turbine generator, and the rate of auxiliary power respectively;

[0184] Natural gas power model of the biogas tank:

[0185] G BD (t) = V BD (t) η B2G q CH4 ;

[0186] In the formula: G BD (t) is the power of natural gas obtained after purification treatment of the biogas in the period t; η B2G is the purification coefficient of the biogas for producing natural gas; q CH4 is the calorific value of the natural gas, V BD (t) is the gas production model of the biogas tank;

[0187] V BD (t) = α BD M BD (t) + β BD H BD (t);

[0188] H BD (t) = (T z (t+1) - T z (t)) V B c m ρ m + H loss (t);

[0189] H loss (t) = c m M BD (t) (T z (t) - T e ) + ko A o (T z (t)-T e );

[0190] In the formula: α BD β is the gas production coefficient per unit organic matter; BD H is the gas production coefficient per unit heat energy; BD (t) represents the thermal power supplied to the biogas digester during time period t; H loss (t) represents the heat dissipation power of the biogas digester during time period t; V B c is the volume of the biogas digester. m For the heat capacity of the feed material; ρ m M represents the feed density. BD (t) represents the mass of organic waste fed into the biogas digester during time period t; A o k is the heat dissipation area inside the pool. o T is the heat transfer coefficient; e The external temperature of the biogas digester.

[0191] The mathematical model of the energy conversion device includes:

[0192] Electrolyzer hydrogen production model:

[0193]

[0194] Where: m EL (t) represents the mass of hydrogen produced by the electrolyzer during time period t; η EL,H2 The hydrogen production efficiency of the electrolyzer equipment; P EL (t) represents the power consumption of the electrolytic cell during time period t; q H This refers to the calorific value of hydrogen.

[0195] Electrolytic cell heat recovery model:

[0196] H EL (t)=η EL,h (1-η EL,H2 )P EL (t);

[0197] Where: H EL (t) represents the thermal power recovered by the electrolytic cell during time period t; η EL,h The heat exchange efficiency of the heat exchange device;

[0198] Hydrogen fuel cell power model:

[0199] P FC (t)=η FC,e m FC (t)q H ;

[0200] In the formula: P FC(t) is the hydrogen fuel cell power generation power at t time period; η FC,e is the hydrogen fuel cell power generation efficiency; m FC (t) is the hydrogen mass consumed by the hydrogen fuel cell at t time period;

[0201] Hydrogen fuel cell heat recovery model:

[0202] H FC (t) = η FC,h m FC (t)q H ;

[0203] In the formula: H FC (t) is the hydrogen fuel cell heat generation power at t time period; η FC,h is the hydrogen fuel cell heat generation efficiency;

[0204] Cogeneration unit output model:

[0205]

[0206] In the formula: G CHP (t), P CHP (t), H CHP (t) are the input gas power, output electric and heat power of the CHP unit at t time period respectively; η CHP , α CHP are the electric conversion efficiency and thermal energy efficiency ratio of the CHP unit respectively.

[0207] The mathematical model of the energy storage device includes:

[0208] Energy storage battery charging and discharging model:

[0209]

[0210] In the formula: S ES (t) represents the amount of electricity stored by the energy storage battery at t time period; σ ES represents the self-discharge rate of the energy storage battery; η ES represents the charging and discharging efficiency of the energy storage battery; P ES,ch (t) and P ES,dis (t) represent the charging and discharging power of the energy storage battery at t time period respectively; Δt represents the time span from t to t+1;

[0211] Thermal storage tank charging and discharging model:

[0212]

[0213] In the formula: S HS (t) represents the thermal energy stored by the thermal storage tank at t time period; σ HS represents the self-heat dissipation rate of the thermal storage tank; η HSdenotes the charging and discharging efficiency of the hydrogen storage tank; H HS,ch (t) and H HS,dis (t) respectively denote the charging and discharging power of the hydrogen storage tank at time t;

[0214] Hydrogen charging and discharging model of the hydrogen storage tank:

[0215]

[0216] wherein: S H2S (t) denotes the hydrogen mass stored in the hydrogen storage tank at time t; σ H2S denotes the self-loss rate of the hydrogen storage; η H2S denotes the charging and discharging efficiency of the hydrogen storage tank; m H2S,ch (t) and m H2S,dis (t) respectively denote the charging and discharging amount of the hydrogen storage tank at time t.

[0217] The method for generating a typical day containing multi-dimensional data based on spectral joint clustering comprises the following steps:

[0218] Let the original data matrix be wherein m represents the number of days, and n represents the variable dimension; construct a row similarity matrix describe the multivariate time series similarity between days:

[0219]

[0220] wherein: x i is the multivariate time series feature vector of the i-th day, x j is the multivariate time series feature vector of the j-th day, and σ is a Gaussian kernel bandwidth parameter, which controls the similarity decay rate;

[0221] construct a column similarity matrix describe the correlation between variable-hour joint features:

[0222]

[0223] wherein: k and l are column indices, corresponding to different variable combinations at different hours; x :,k is the k-th column of the data matrix, representing the data of all days at a certain variable at a certain hour; x :,l is the l-th column of the data matrix, which is similar to x :,k but corresponds to another different variable-hour combination;

[0224] calculate the normalized Laplacian matrix for the rows and columns respectively:

[0225]

[0226] wherein: L r , Lc are the row and column normalized Laplacian matrices respectively; D r , D c are the row and column degree matrices respectively;

[0227] singular value decomposition is performed on the double normalized matrix X T L r X+XL c X T singular value decomposition is performed on the double normalized matrix X r and column clusters C c , the objective function is to minimize the residual error between row and column clusters:

[0228]

[0229] wherein: μ Cr,Cc represents the interactive center of row and column clusters; X(i,j) directly represents the value of the i-th time point, j-th variable-hour feature in the original data matrix, which is compared with the interactive center to optimize the clustering division.

[0230] For each row cluster C k , calculate the multivariate time series mean of all days it contains as a typical day:

[0231]

[0232] wherein: represents the mean value of the h-th hour, v-th variable in the k-th typical day, representing the common mode of the cluster; x (d,h,v) represents the specific value of the d-th day, h-th hour, v-th variable in the original data;

[0233] According to the proportion of samples in the cluster, the typical day probability is calculated:

[0234]

[0235] Based on the selected typical day, the corresponding photovoltaic output, electrical load demand, thermal load demand, gas load demand, and hydrogen load demand data are extracted, and s typical day data are output; each typical day retains the photovoltaic output data and the coupling characteristics of electrical, thermal, and hydrogen multi-energy loads within 24 hours.

[0236] In this embodiment, a hydrogen energy storage long-period operation model is constructed, including:

[0237] In a typical day, the relationship between the hydrogen storage amount at two consecutive time points is:

[0238]

[0239] ΔSH2S (k, 0) = 0;

[0240] wherein: ΔS H2S (k, t) represents the hydrogen mass stored in the hydrogen storage tank at time t of the k typical day, kg, and the initial time within a day is set to 0, which is irrelevant to the actual hydrogen storage amount of the hydrogen storage tank; m H2S,ch (k, t) and m H2S,dis (k, t) represent the hydrogen charging amount and the hydrogen discharging amount of the hydrogen storage tank at time t of the k typical day, respectively;

[0241] Between two consecutive operation days, the initial hydrogen mass stored in the hydrogen storage tank needs to satisfy:

[0242] S H2S,0 (r+1) = (1-σ H2S ×24)S H2S,0 (r) + ΔS H2S (k = h(r), 24);

[0243] wherein: S H2S,0 (r+1) represents the initial hydrogen mass stored in the hydrogen storage tank at the r+1 operation day; h(r) is a mapping function between the k typical day and the operation day r;

[0244] In order to ensure that the hydrogen storage amount of the hydrogen storage tank is the same within a year, it needs to satisfy:

[0245] S H2S,0 (1) = (1-σ H2S ×24)S H2S,0 (N r )+ΔS H2S (k = h(N r ), 24);

[0246] wherein: N r is the number of operation days in a year;

[0247] The actual hydrogen mass stored in the hydrogen storage tank at any time within a year can be decomposed into the initial hydrogen storage amount and the dynamic superposition of the typical day scenario, which is represented as:

[0248] S H2S (r, t) = S H2S,0 (r) + ΔS H2S (k = h(r), t);

[0249] wherein: S H2S (r, t) represents the actual hydrogen mass stored in the hydrogen storage tank at time t of the r operation day, which is the sum of the initial hydrogen storage amount of the r operation day and the hydrogen mass stored at time t of the r operation day;

[0250] The hydrogen stored in the hydrogen storage tank needs to satisfy the upper and lower limit constraints:

[0251] αH2S,min S H2S,cap ≤S H2S (r,t)≤α H2S,max S H2S,cap ;

[0252] Since the above formula is true for each time in the scheduling period, to reduce constraints, the following equivalent processing is performed:

[0253]

[0254]

[0255] In the formula: ΔS H2S,max (k) and ΔS H2S,min (k) represent the maximum and minimum values of the hydrogen storage tank storing hydrogen mass in k typical days, respectively, in kg.

[0256] The objective function of the upper layer capacity configuration model with the minimum life cycle equivalent annual cost as the optimization objective is:

[0257] min C = C inv +C op +C scrap +C loss ;

[0258] In the formula, C inv is the equivalent annual investment cost, C op is the operation cost, C scrap is the scrap cost, and C loss is the loss cost.

[0259]

[0260] In the formula: y represents the service life of the equipment; r is the discount rate; μ i is the unit capacity investment cost of the i-th equipment; S i represents the configuration capacity of the i-th equipment; X represents the set of equipment for capacity configuration in the system, including waste power generation, biogas fermentation, electrolytic tank, hydrogen fuel cell, combined heat and power unit, energy storage battery, heat storage tank, and hydrogen storage tank;

[0261]

[0262] In the formula: s is the total number of typical days in a year; p k is the probability of k typical days appearing in a year; N r is the number of operating days in a year; C om,k is the typical day operation and maintenance cost, C buy,k is the energy purchase cost, C CO2,k is the carbon trading cost, C waste,k is the waste treatment subsidy.sell,k To generate revenue.

[0263] The upper-layer capacity configuration model also includes the following constraints:

[0264] Device installation capacity limit constraints:

[0265] 0≤S i ≤S i,max ;

[0266] In the formula: S i represents the installed capacity of the i-th device, S i,max represents the upper limit of the capacity that the i-th device can allow to be installed.

[0267] The lower-layer scheduling operation model takes the minimum annual operation cost as the optimization objective, and the objective function is:

[0268] minC op =C om +C buy +C CO2 -C waste -C sell ;

[0269] The objective function includes the annual operation cost C om , the energy purchase cost C buy , the carbon trading cost C CO2 , the waste treatment subsidy C waste , and the energy sale revenue C sell ;

[0270] Wherein:

[0271]

[0272] In the formula: λ i is the unit capacity operation cost of device i; P i (k, t) represents the output power of the i-th device at time period t of typical day k; X represents the set of devices in the system, including photovoltaic units, waste power generation, biogas fermentation, electrolytic cells, hydrogen fuel cells, combined heat and power units, electric boilers, energy storage batteries, heat storage tanks, and hydrogen storage tanks;

[0273]

[0274] In the formula: P buy (k, t) and G buy (k, t) are the purchased electricity power and natural gas power at time period t of typical day k, respectively;

[0275]

[0276] In the formula: W gar(k, t) is the amount of waste treated in period t of typical day k; M BD (r, t) is the amount of organic waste put into the biogas tank in period t of typical day k;

[0277]

[0278] wherein: m sell (k, t) is the amount of hydrogen sold by the system in period t of typical day k.

[0279] The lower-level scheduling operation model further comprises the following constraints:

[0280] The electrical power balance constraint:

[0281] P buy (r, t) + P PV (r, t) + P gar (r, t) + P FC (r, t) + P CHP (r, t) + P ES,dis (r, t) + P

[0282] = P EL (r, t) + P EB (r, t) + P ES,ch (r, t) + P L (r, t)

[0283] wherein: P L (r, t) is the electrical load of the system in period t of r operating day, P buy (r, t) is the purchased electrical power in period t of r operating day, P PV (r, t) is the actual power generated by the photovoltaic unit in period t of r operating day; P gar (r, t) is the power generated by the waste in period t of r operating day; P FC (r, t) is the power generated by the hydrogen fuel cell in period t of r operating day; P CHP (r, t) is the input gas power of the CHP unit in period t of r operating day, P ES,ch (r, t) and P ES,dis (r, t) respectively represent the charging and discharging power of the energy storage battery in period t of r operating day; P EL (r, t) is the power consumption of the electrolytic cell in period t of r operating day;

[0284] The thermal power balance constraint:

[0285] H gar (r, t) + H CHP (r, t) + H EL (r, t) + H FC (r, t) + H EB (r, t) + HHS,dis (r,t);

[0286] =H BD (r,t)+H HS,ch (r,t)+H L (r,t)

[0287] wherein H L (r,t) is the heat load of the system at time period t on day r; H gar (r,t) is the heat generation power of the garbage at time period t on day r; H FC (r,t) is the heat generation power of the hydrogen fuel cell at time period t on day r; H CHP (r,t) is the heat power of the CHP unit at time period t on day r, H HS,ch (r,t) and H HS,dis (r,t) represent the heat storage and release power of the heat storage tank at time period t on day r, respectively; H EL (r,t) is the heat power recovered by the electrolytic tank at time period t on day r; H BD (r,t) is the heat power supplied to the biogas tank at time period t on day r;

[0288] Natural gas balance constraint:

[0289] G buy (r,t)+G BD (r,t)=G CHP (r,t)+G L (r,t);

[0290] wherein G L (r,t) is the gas load of the system at time period t on day r, G buy (r,t) is the purchased natural gas power at time period t on day r; G BD (r,t) is the natural gas power obtained after biogas purification treatment at time period t on day r; G CHP (r,t) is the input gas power of the CHP unit at time period t on day r;

[0291] Hydrogen energy balance constraint:

[0292] m EL (r,t)+m H2S,dis (r,t)=m FC (r,t)+m sell (r,t)+m H2S,ch (r,t)+m L (r,t);

[0293] wherein: m L (r,t) is the hydrogen load of the system at time period t on day r; m EL(r, t) is the hydrogen mass produced by electrolyzer r at time period t on day t; m H2S,ch (r, t) and m H2S,dis (r, t) respectively represent the hydrogen charging mass and hydrogen discharging mass of hydrogen storage tank r at time period t on day t; m FC (r, t) is the hydrogen mass consumed by hydrogen fuel cell r at time period t on day t; m sell (r, t) is the hydrogen mass sold by the system at time period t on day t; m

[0294] System energy purchasing constraint:

[0295] P buy,min ≤ P buy (t) ≤ P buy,max ;

[0296] G buy,min ≤ G buy (r, t) ≤ G buy,max ;

[0297] Each device operation constraint, as well as the above-mentioned operation day constraint, can be mapped to a constraint of typical day k.

[0298] As a preferred embodiment of the above-mentioned embodiment, the Benders decomposition method is used to solve the bi-level optimization configuration model, the bi-level optimization configuration model is split into a master problem and a sub-problem, and the cutting plane method is used to iteratively approach the optimal solution, including the following steps:

[0299] Step 1: initialize the decision variable x 0 of the master problem, set the initial lower bound LB = E T x 0 , the upper bound UB = +∞, set the convergence threshold ε, and set the iteration count i = 0;

[0300] Step 2: in the i-th iteration, solve the master problem, i.e., the upper-level configuration problem, to obtain the current device capacity x i , update the lower bound LB = E T x i , and pass it to the sub-problem;

[0301] Step 3: under the given device capacity x i , solve the feasible sub-problem to obtain solve the convex feasible sub-problem to obtain

[0302] Step 4: if the optimal value of the problem optimization objective is not equal to 0, i.e., the sub-problem is infeasible, generate a feasibility cutting plane, constrain the search space of the master problem, so that subsequent iterations avoid infeasible solutions, update the iteration count i = i + 1, and return to step 2;

[0303] If the optimal value of the problem optimization objective is equal to 0, that is, the sub-problem is feasible, solve the optimal sub-problem to obtain and Generate an optimality cut plane, add it to the main problem constraints, and update the upper bound UB according to the formula;

[0304] Step 5: If the upper and lower bounds satisfy the convergence condition |UB-LB|<ε, stop iteration and output the optimal solution; if not converged, update the iteration number i=i+1 and return to step 2 to continue solving the main problem;

[0305] Step 6: Output the result. Finally, output the optimal device capacity configuration x * .

[0306] As shown in Figure 2 , the embodiment also includes a kind of comprehensive energy system optimization configuration system considering hydrogen energy medium and long term characteristics, using the method as described above, the system includes:

[0307] A mathematical model construction unit for constructing a multi-energy coupling mathematical model of a rural comprehensive energy system including hydrogen energy storage;

[0308] A typical day generation unit for generating a typical day containing multi-dimensional data based on a spectrum joint clustering typical day generation method, constructing a hydrogen energy storage long-period operation model, and decomposing the hydrogen storage amount at any time within the annual operation period into an initial hydrogen storage amount and a dynamic superposition of typical day scenarios;

[0309] A double-layer model construction unit for constructing a double-layer optimization configuration model containing capacity configuration and dispatching operation based on the multi-energy coupling mathematical model and the hydrogen energy storage long-period operation model, the double-layer optimization configuration model including an upper-layer capacity configuration model and a lower-layer dispatching operation model, the upper-layer capacity configuration model taking the minimum life cycle equivalent annual cost as the optimization objective and taking each device capacity as the decision variable, and the lower-layer dispatching operation model taking the minimum annual operation cost as the optimization objective and taking each device output and energy storage operation strategy as the decision variable;

[0310] A solving unit for solving the double-layer optimization configuration model using a Benders decomposition method, splitting the double-layer optimization configuration model into a main problem and a sub-problem, and using a cut plane method to iteratively approximate the optimal solution to obtain the capacity configuration and dispatching operation strategy.

[0311] Embodiment 2:

[0312] The embodiment includes a comprehensive energy system optimization configuration method considering hydrogen energy medium and long term characteristics, introduces hydrogen energy and its coupled devices, and optimizes system capacity configuration considering its long-period operation characteristics, aiming to improve the economy, environmental protection and reliability of rural energy systems, including the following steps:

[0313] A mathematical model of multi-energy coupling of rural integrated energy system is constructed, and the system mathematical model includes mathematical models of energy supply equipment, energy conversion equipment and energy storage equipment, as shown in Figure 3

[0314] A typical day containing multi-dimensional data is generated based on a typical day generation method of spectral joint clustering, a hydrogen storage energy long-period operation model is constructed, and the hydrogen storage amount at any time in the annual operation period is decomposed into an initial hydrogen storage amount and a dynamic superposition of a typical day scenario;

[0315] A double-layer optimization configuration model including capacity configuration and scheduling operation is constructed: the double-layer optimization configuration model includes two parts of capacity configuration and scheduling operation, the upper-layer capacity configuration model takes the minimum equal annual cost in the whole life cycle as the optimization objective, takes the capacity of each equipment as the decision variable, and the constraint condition is the installation capacity limit of the equipment, and the lower-layer scheduling operation model takes the minimum annual operation cost as the optimization objective, and the decision variable is the output of each equipment and the operation strategy of energy storage;

[0316] The Benders decomposition method is used to solve the double-layer optimization model, the double-layer optimization configuration model is split into a master problem and a sub-problem, and the cut plane method is used to iteratively approximate the optimal solution, so as to effectively reduce the solving complexity and improve the calculation convergence. Based on the simulation analysis of certain large rural data, the reasonable configuration and economic operation of the equipment are realized.

[0317] In this embodiment, the models of various key equipment are constructed as follows:

[0318] The power generation model of the distributed photovoltaic unit can be expressed as:

[0319]

[0320] In the formula, P PV (t) is the actual power generation of the photovoltaic unit at the t period; P STC is the rated output power of the photovoltaic unit; G ING (t) is the solar radiation intensity at the t period; G STC is the solar radiation intensity under standard environment; k PV is the temperature coefficient; T c (t) is the surface temperature of the photovoltaic panel; T r is the reference temperature.

[0321] Garbage power generation is an effective way of rural waste utilization, which can convert production and living garbage into electric energy and thermal energy, including pyrolysis gasification and direct incineration. The garbage power supply model can be expressed as:

[0322] H WI (t) = W gar (t) α WI β WI q​WI η WI ;

[0323] H WG (t)=W gar (t)α WG q WG η WG ;

[0324] H gar (t)=H WI (t)+H WG (t);

[0325] P gar (t)=(H WI (t)+H WG (t))η WHB (1-η aux );

[0326] Where: H WI (t) and H WG (t) represents the thermal power generated by municipal solid waste incineration and gasification during time period t; H gar (t) represents the heat generation power of waste during time period t; P gar (t) represents the waste-to-energy power generation capacity during time period t; W gar (t) represents the amount of waste processed during time period t; α WI and α WG These represent the proportions of incinerable and gasifiable waste in municipal solid waste; β WI The percentage of incinerable waste remaining after leachate is discharged from the storage tank; q WI and q WG These are the low-grade calorific values ​​of incinerable waste and gasifiable waste, respectively; η WI η WG η WHB η aux These are incinerator efficiency, pyrolysis gasification furnace thermal efficiency, steam turbine power generation efficiency, and plant power consumption rate.

[0327] Abundant biomass resources in rural areas can be ecologically transformed through anaerobic fermentation biogas technology. This is an effective method for treating organic waste and recovering energy, characterized by process controllability, high utilization rate, and good environmental protection. By mixing organic waste such as manure and kitchen waste, biogas can be produced through microbial fermentation in an anaerobic environment. The purified biogas can then be converted into electricity and heat using a CHP (Clean Gasification) unit, which can also supply rural gas loads such as cooking and lighting.

[0328] The biogas production model of a biogas digester can be represented as follows:

[0329] V BD (t)=a|T z(t)-T o |+b;

[0330] In the formula: VBD(t) is the biogas digester output during time period t; Tz(t) and To are the actual reaction temperature and the optimal reaction temperature, respectively; a and b are coefficients obtained from data fitting.

[0331] Because biogas production is highly sensitive to temperature changes, thermal control measures are needed to maintain a constant temperature environment required for the fermentation process in order to improve gas production efficiency. The heat power supplied to the biogas digester is affected by factors such as the actual reaction temperature, heat dissipation, and digester volume, and its expression is as follows:

[0332] H BD (t)=(T z (t+1)-T z (t))V B c m ρ m +H loss (t);

[0333] H loss (t)=c m M BD (t)(T z (t)-T e )+k o A o (T z (t)-T e );

[0334] Where: H BD (t) represents the thermal power supplied to the biogas digester during time period t; H loss (t) represents the heat dissipation power of the biogas digester during time period t; V B c is the volume of the biogas digester. m For the heat capacity of the feed material; ρ m M represents the feed density. BD (t) represents the mass of organic waste fed into the biogas digester during time period t; A o k is the heat dissipation area inside the pool. o T is the heat transfer coefficient; e The external temperature of the biogas digester.

[0335] Based on empirical data, biogas production is mainly affected by the amount of organic matter added and the amount of heat supplied. The biogas model is simplified as follows:

[0336] V BD (t)=α BD M BD (t)+β BD H BD (t);

[0337] In the formula: α BDGas production coefficient of unit organic matter; β BD Gas production coefficient of unit heat energy.

[0338] Using biogas purification technology, biogas produced by fecal pollution treatment facilities can be made into natural gas to supply rural gas load. The power of natural gas obtained after biogas purification treatment is:

[0339] G BD (t)=V BD (t)η B2G q CH4 ;

[0340] In the formula: G BD (t) is the power of natural gas obtained after biogas purification treatment at t; η B2G is the purification coefficient of biogas to natural gas; q CH4 is the calorific value of natural gas.

[0341] The energy conversion equipment includes electrolytic cell, fuel cell and combined heat and power unit, and the model is established as follows:

[0342] The electrolytic cell is the core device for hydrogen energy production, which realizes the conversion of electric energy into hydrogen and oxygen through electrochemical reaction. The electrolytic hydrogen production technology can convert excess renewable energy into hydrogen energy storage, and cooperate with hydrogen fuel cell power generation technology to realize the bidirectional conversion of electricity-hydrogen-electricity, which can not only smooth short-term power fluctuations, but also solve the long-term seasonal imbalance between supply and demand. In addition, hydrogen energy can realize flexible conversion of energy forms through hydrogen-to-gas, hydrogen-to-electricity and other technologies, and strengthen the coupling and interaction of power, heat and natural gas systems.

[0343] The hydrogen production model of electrolytic cell can be expressed as:

[0344]

[0345] In the formula: m EL (t) is the mass of hydrogen produced by the electrolytic cell at t; η EL,H2 is the hydrogen production efficiency of the electrolytic cell equipment; P EL (t) is the power consumption of the electrolytic cell at t; q H is the calorific value of hydrogen. Since a large amount of high-grade waste heat will be generated during the electrolytic hydrogen production process of the electrolytic cell, heat recovery can be carried out, and the heat recovery model can be expressed as:

[0346] H EL (t)=η EL,h (1-η EL,H2 )P EL (t);

[0347] In the formula: H EL (t) is the heat power recovered by the electrolytic cell at t; η EL,h is the heat exchange efficiency of the heat exchange device.

[0348] Similarly, the mathematical model of the hydrogen fuel cell can be expressed as:

[0349] P FC (t) = η FC,e m FC (t) q H ;

[0350] wherein P FC (t) is the power generated by the hydrogen fuel cell at time t; η FC,e is the power generation efficiency of the hydrogen fuel cell; and m FC (t) is the mass of hydrogen consumed by the hydrogen fuel cell at time t. Since the hydrogen fuel cell releases a large amount of high-grade waste heat during operation, multi-stage utilization of energy can be achieved through waste heat recovery, and the heat recovery model can be expressed as:

[0351] H FC (t) = η FC,h m FC (t) q H ;

[0352] wherein H FC (t) is the heat generation power of the hydrogen fuel cell at time t; and η FC,h is the heat generation efficiency of the hydrogen fuel cell.

[0353] A combined heat and power (CHP) unit can realize multi-energy coupling of electricity-heat-gas and is the core equipment in a comprehensive energy system. The CHP unit adopts an integrated architecture of a gas turbine and a waste heat boiler, compresses air and combusts natural gas to drive a turbine to generate electricity, and recovers low-grade heat energy from exhaust gas using the waste heat boiler to realize electricity-heat cascade utilization. The output model can be expressed as:

[0354] P MT (t) = G MT (t) η gas η MT ;

[0355] wherein P MT (t) and G MT (t) are the electric power output by the gas turbine and the natural gas power consumed by the gas turbine at time t, respectively; and η gas and η MT are the natural gas utilization rate and the power generation efficiency of the gas turbine, respectively.

[0356] The waste heat boiler recovers low-grade exhaust gas discharged from the gas turbine and converts the waste heat into usable heat energy through a heat exchange device to realize heat cascade utilization. The heat power output by the waste heat boiler can be expressed as:

[0357] H HE (t) = G MT (t)(t)η gas η CHP (1-η CHP,loss )η HE η h ;

[0358] H HE (t) is the output heat power of the waste heat boiler at time t; η CHP , η CHP,loss , η HE and η h are the heat conversion efficiency, heat loss rate, heat exchange device efficiency and heating coefficient of the waste heat boiler, respectively.

[0359] Based on the above, the power supply and heat supply model of the CHP unit can be equivalent to:

[0360]

[0361] G CHP (t), P CHP (t) and H CHP (t) are the input gas power, output electric power and output heat power of the CHP unit at time t, respectively; η CHP and α CHP are the electric conversion efficiency and thermal energy efficiency ratio of the CHP unit, respectively.

[0362] The energy storage device includes an energy storage battery, a heat storage tank and a hydrogen storage tank, and the model is established as follows:

[0363] The charge and discharge model of the energy storage battery can be expressed as:

[0364]

[0365] S ES (t) represents the amount of electricity stored by the energy storage battery at time t; σ ES represents the self-discharge rate of the energy storage battery; η ES represents the charge and discharge efficiency of the energy storage battery; P ES,ch (t) and P ES,dis (t) represent the charge and discharge power of the energy storage battery at time t, respectively; Δt represents the time span from t to t+1.

[0366] The amount of electricity stored by the energy storage battery needs to satisfy the upper and lower limit constraints:

[0367] α ES,min S ES,cap ≤ S ES (t) ≤ α ES,max S ES,cap ;

[0368] S ES,cap is the maximum capacity of the energy storage battery; αES,min , α ES,max are minimum and maximum energy storage ratio respectively.

[0369] Since the energy storage battery cannot charge and discharge at the same time, it needs to meet the charge and discharge power constraints:

[0370] 0≤P ES,ch (t)≤z ES,ch (t)β ES,ch S ES,cap ;

[0371] 0≤P ES,dis (t)≤z ES,dis (t)β ES,dis S ES,cap ;

[0372] z ES,ch (t)+z ES,dis (t)≤1;

[0373] In the formula: z ES,ch (t), z ES,dis (t) is the state variable of the energy storage battery charging and discharging at t, which is a 0-1 variable; β ES,ch is the ratio of maximum charging power to energy storage battery capacity, indicating how fast it can charge at most under normal conditions; β ES,dis is the ratio of maximum discharging power to energy storage battery capacity, indicating how fast it can discharge at most under normal conditions.

[0374] At the same time, in order to facilitate periodic control of energy storage, the total energy change of the energy storage battery within the scheduling period T should be balanced:

[0375]

[0376] Similar to the energy storage battery, the charging and discharging model of the heat storage tank can be represented as:

[0377]

[0378] In the formula: S HS (t) represents the heat energy stored in the heat storage tank at t; σ HS represents the self-heating rate of the heat storage tank; η HS represents the charging and discharging efficiency of the heat storage tank; H HS,ch (t) and H HS,dis (t) represent the charging and discharging power of the heat storage tank at t respectively.

[0379] The heat energy stored in the heat storage tank also needs to meet the upper and lower limit constraints:

[0380] α HS,min S HS,cap ≤S HS(t)≤α HS,max S HS,cap ;

[0381] wherein S HS,cap is the maximum capacity of the heat storage tank; α HS,min , α HS,max are the minimum and maximum heat storage ratios, respectively.

[0382] Since the heat storage tank cannot simultaneously charge and discharge heat, the charge and discharge power constraints must be satisfied:

[0383] 0≤P HS,ch (t)≤z HS,ch (t)β HS,ch S HS,cap ;

[0384] 0≤P HS,dis (t)≤z HS,dis (t)β HS,dis S HS,cap ;

[0385] z HS,ch (t)+z HS,dis (t)≤1;

[0386] wherein z HS,ch (t), z HS,dis (t) are the charging and discharging state variables of the heat storage tank at time t, which are 0-1 variables; β HS,ch is the ratio of the maximum heat storage power to the capacity of the heat storage tank; β HS,dis is the ratio of the maximum discharging power to the capacity of the heat storage tank.

[0387] At the same time, in order to facilitate periodic control of the heat storage tank, the total energy change of the heat storage tank within the dispatching period T should be balanced:

[0388]

[0389] Hydrogen energy storage is a new type of energy storage technology that has shown significant advantages in new power systems. Hydrogen energy storage not only enables GW-level large-scale storage, but also breaks through the daily cycle limit of traditional energy storage to achieve cross-season regulation, and has gradually become an important technical support for balancing the fluctuations of intermittent power sources and achieving energy space-time transfer.

[0390] The hydrogen storage tank charging and discharging model can be represented as:

[0391]

[0392] wherein S H2S (t) represents the mass of hydrogen stored in the hydrogen storage tank at time t; σ H2S represents the self-loss rate of hydrogen storage; η H2S represents the charging and discharging efficiency of the hydrogen storage tank; mH2S,ch (t) and m H2S,dis (t) represent the hydrogen charging and discharging amounts of the hydrogen storage tank t at time period, respectively.

[0393] The hydrogen stored in the hydrogen storage tank also needs to meet the upper and lower limit constraints:

[0394] α H2S,min S H2S,cap ≤S H2S (t)≤α H2S,max S H2S,cap ;

[0395] In the formula: S H2S,cap is the maximum capacity of the hydrogen storage tank; α H2S,min , α H2S,max are the minimum and maximum hydrogen storage ratios, respectively.

[0396] Since the hydrogen storage tank cannot charge and discharge hydrogen at the same time, it needs to meet the charging and discharging power constraints:

[0397] 0≤P H2S,ch (t)≤z H2S,ch (t)β H2S,ch S H2S,cap ;

[0398] 0≤P H2S,dis (t)≤z H2S,dis (t)β H2S,dis S H2S,cap ;

[0399] z H2S,ch (t)+z H2S,dis (t)≤1;

[0400] In the formula: z H2S,ch (t), z H2S,dis (t) are the charging and discharging state variables of the hydrogen storage tank at time period t, which are 0-1 variables; β H2S,ch is the ratio of the maximum hydrogen charging power to the capacity of the hydrogen storage tank; β H2S,dis is the ratio of the maximum hydrogen discharging power to the capacity of the hydrogen storage tank.

[0401] In this example, the hydrogen energy storage long-period operation model acquisition method is as follows:

[0402] The optimization of rural integrated energy system considering long-period operation of hydrogen energy storage is a large-scale mixed integer optimization problem, which is difficult to solve due to long optimization duration, large number of decision variables and model complexity. To reduce the computational complexity and improve the solving efficiency, a clustering algorithm is usually used to extract representative typical scenarios from massive historical data, and the weighted sum of the operation cost of each scenario is used to replace the production simulation of 8760 hours in a year, so that the high-dimensional optimization problem is reduced to the combination solving of a limited number of scenarios. The typical day generation method based on spectral joint clustering used in the invention has the following process:

[0403] Let the original data matrix be Where m represents the number of days, and n represents the variable dimension. Construct the row similarity matrix Describe the multivariate time series similarity between days:

[0404]

[0405] In the formula: x i is the multivariate time series feature vector of the i-th day, x j is the multivariate time series feature vector of the j-th day, and sigma is the Gaussian kernel bandwidth parameter, which controls the similarity decay rate.

[0406] Construct the column similarity matrix Describe the correlation between variable-hour joint features:

[0407]

[0408] In the formula: k and l are column indices corresponding to different variable combinations at different hours; x :,k is the k-th column of the data matrix, representing the data of all days at a certain variable at a certain hour. Similarly, x :,l is the l-th column of the data matrix, which is similar to x :,k , but corresponds to another different variable-hour combination.

[0409] Calculate the normalized Laplacian matrix for rows and columns respectively:

[0410]

[0411] In the formula: L r , L c are the row and column normalized Laplacian matrices respectively; D r , D c are the row and column degree matrices respectively.

[0412] To the double-normalized matrix X T L r X+XL c X TSingular value decomposition is performed, and the first s eigenvectors are selected to form a low-dimensional embedding space. K-means clustering is performed on the rows and columns in the low-dimensional space, respectively, to obtain row clusters C r and column clusters C c The objective function is to minimize the residual error between row and column clusters:

[0413]

[0414] In the formula: μ Cr,Cc represents the interaction center of row and column clusters.

[0415] For each row cluster C k , the multivariate time series mean of all days contained therein is calculated as a typical day:

[0416]

[0417] In the formula: represents the mean of the hth hour, vth variable in the kth typical day, representing the common mode of the cluster; x (d,h,v) represents the specific value of the dth day, hth hour, and vth variable in the original data;

[0418] The typical day probability is calculated according to the proportion of samples in the cluster:

[0419]

[0420] Based on the selected typical day, the corresponding photovoltaic output, electrical load demand, thermal load demand, gas load demand, and hydrogen load demand data are extracted, and s typical day data are output. Each typical day retains the coupling characteristics of photovoltaic output data and electrical, thermal, and hydrogen multi-energy loads for 24 hours within the day, and can represent a similar daily scene mode in the data set.

[0421] On this basis, energy storage state models are established for "a single typical day" and "multiple typical days coupled", as shown in Figure 4 The hydrogen storage amount at any time of the year is decomposed into an initial hydrogen storage amount and a dynamic superposition of typical day scenarios, the hydrogen charging and discharging strategy of the hydrogen storage tank is accurately described within a single typical day, the hydrogen storage amount is linked between typical days, and the calculation complexity is effectively reduced.

[0422] Since the hydrogen storage tank can be operated for a long period of time, its hydrogen storage state has time series coupling constraints between different typical days and within a typical day. Within a typical day, the relationship between the hydrogen storage amounts at two consecutive times is:

[0423]

[0424] ΔS H2S (k,0)=0;

[0425] In the formula: ΔSH2S (k,t) represents the hydrogen mass stored in the hydrogen storage tank at k typical day t, kg, and the initial time in a day is set to 0, which is irrelevant to the actual hydrogen storage amount of the hydrogen storage tank; m H2S,ch (k,t) and m H2S,dis (k,t) respectively represent the hydrogen charging amount and the hydrogen discharging amount of the hydrogen storage tank at k typical day t.

[0426] Between two consecutive operation days, the initial hydrogen mass stored in the hydrogen storage tank needs to meet:

[0427] S H2S,0 (r+1)=(1-σ H2S ×24)S H2S,0 (r)+ΔS H2S (k=h(r),24);

[0428] In the formula: S H2S,0 (r+1) represents the initial hydrogen mass stored in the hydrogen storage tank at r+1 operation day; h(r) is a mapping function between k typical day and operation day r.

[0429] In order to ensure that the hydrogen storage amount of the hydrogen storage tank is the same within a year, it needs to meet:

[0430] S H2S,0 (1)=(1-σ H2S ×24)S H2S,0 (N r )+ΔS H2S (k=h(N r ),24);

[0431] In the formula: N r is the number of operation days in a year.

[0432] The actual hydrogen storage mass of the hydrogen storage tank at any time in a year can be decomposed into the initial hydrogen storage amount and the dynamic superposition of the typical day scenario, which can be represented as:

[0433] S H2S (r,t)=S H2S,0 (r)+ΔS H2S (k=h(r),t);

[0434] In the formula: S H2S (r,t) represents the actual hydrogen mass stored in the hydrogen storage tank at r operation day t, which is the sum of the initial hydrogen storage amount at r operation day and the hydrogen mass stored at r operation day t.

[0435] In order to ensure the safe operation of hydrogen energy storage, the hydrogen stored in the hydrogen storage tank needs to meet the upper and lower limit constraints:

[0436] α H2S,min S H2S,cap ≤S H2S(r, t)≤α H2S,max S H2S,cap ;

[0437] Since the above formula is true for each time in the scheduling period, to reduce the constraints, the following equivalent processing is performed:

[0438]

[0439] In the formula: ΔS H2S,max (k) and ΔS H2S,min (k) represent the maximum and minimum values of the hydrogen storage mass of the hydrogen storage tank on the k typical day, kg.

[0440] In this example, the method for obtaining the bi-level optimization configuration model including capacity configuration and scheduling operation is as follows:

[0441] The bi-level optimization configuration model includes two parts of capacity configuration and scheduling operation, as shown in Figure 5 .

[0442] The upper layer capacity configuration model takes the minimum life cycle equivalent annual cost as the objective function, and the life cycle equivalent annual cost includes equivalent annual investment cost C inv , operation cost C op , scrap cost C scrap and loss cost C loss :

[0443] min C = C inv + C op + C scrap + C loss ;

[0444] The equivalent annual investment cost C inv refers to the investment cost of the equipment per year converted by the annual capital present value coefficient, and the conversion model can be expressed as:

[0445]

[0446] In the formula: y represents the service life of the equipment; r is the discount rate; μ i is the unit capacity investment cost of the i-th equipment; S i represents the configuration capacity of the i-th equipment; and X represents the set of devices for capacity configuration in the system, including waste power generation, biogas fermentation, electrolytic tank, hydrogen fuel cell, combined heat and power unit, energy storage battery, heat storage tank and hydrogen storage tank.

[0447] The operation cost C op refers to the total of the annual operation cost of each device in the system, including the annual operation and maintenance cost C om , energy purchase cost C buy , carbon trading cost C CO2Waste disposal subsidy C waste and sales revenue C sell :

[0448]

[0449] In the formula: s is the total number of typical days in a year; p k Let k be the probability of a typical day occurring throughout the year; N r C represents the number of operating days throughout the year. om,k Typical daily maintenance cost, C buy,k For energy purchase costs, C CO2,k For carbon trading costs, C waste,k For waste disposal subsidies, C sell,k Revenue from energy sales.

[0450] Scrapping cost C scrap This refers to the disposal costs incurred when equipment in a system needs to be scrapped due to reaching the end of its service life, technological obsolescence, or failure, including dismantling costs and environmental remediation costs. For the sake of simplicity, this article defines it as 3% of the initial investment cost.

[0451] Loss cost C loss This refers to the economic cost incurred due to energy losses during the conversion, transmission, and storage processes in the system. For the sake of simplicity, this paper defines it as 5% of the energy purchase cost.

[0452] The constraints of the configuration model mainly consider the limitations of equipment installation capacity:

[0453] 0≤S i ≤S i,max ;

[0454] In the formula: S i S represents the installation capacity of the i-th type of equipment. i,max This represents the maximum capacity that the i-th type of device can be installed in.

[0455] A lower-level scheduling and operation model is constructed based on the long-term operation model of hydrogen energy storage, with the objective function being to minimize the system's annual operating cost C, which is the C in the upper-level objective function. op The details are as follows:

[0456] Operation and maintenance cost C om This refers to the total maintenance cost of all devices within the scheduling cycle, expressed as:

[0457]

[0458] In the formula: λ i P represents the unit capacity operation and maintenance cost of device i; i(r, t) represents the output power of the ith device at the rth operating day and the tth time period; X represents the set of devices in the system, including photovoltaic units, waste power generation, biogas fermentation, electrolytic cells, hydrogen fuel cells, cogeneration units, electric boilers, energy storage batteries, heat storage tanks, and hydrogen storage tanks.

[0459] The above formula can be simplified by the mapping function h(r) between the typical day k and the operating day r as follows:

[0460]

[0461] In the formula, p k is the probability of the kth typical day occurring in the whole year; N r is the number of operating days in the whole year.

[0462] The purchase cost C buy is the cost of purchasing power from the upper-level power grid, and the expression is as follows:

[0463]

[0464] In the formula, P buy (r, t) and G buy (r, t) are the purchased electric power and natural gas power at the rth operating day and the tth time period, respectively. It can be simplified as follows:

[0465]

[0466] The waste treatment subsidy C waste is the subsidy obtained by the system for treating rural waste, and the expression is as follows:

[0467]

[0468] In the formula, W gar (r, t) is the amount of waste treated at the rth operating day and the tth time period; MBD(r, t) is the mass of organic waste input into the biogas tank at the rth operating day and the tth time period. It can be simplified as follows:

[0469]

[0470] The sale income C sell is the income generated by the system from the sale of hydrogen, and the expression is as follows:

[0471]

[0472] In the formula, m sell (r, t) is the amount of hydrogen sold by the system at the rth operating day and the tth time period. It can be simplified as follows:

[0473]

[0474] The constraints include energy balance constraints, system energy purchase constraints, and equipment output constraints. The energy balance constraints include:

[0475] 1) Electrical power balance constraint:

[0476]

[0477] where: P L (r, t) is the electrical load of the system at r operating day t.

[0478] 2) Thermal power balance constraint:

[0479]

[0480] where: H L (r, t) is the thermal load of the system at r operating day t.

[0481] 3) Natural gas balance constraint:

[0482] G buy (r, t) + G BD (r, t) = G CHP (r, t) + G L (r, t);

[0483] where: G L (r, t) is the gas load of the system at r operating day t.

[0484] 4) Hydrogen energy balance constraint:

[0485] m EL (r, t) + m H2S,dis (r, t) = m FC (r, t) + m sell (r, t) + m H2S,ch (r, t) + m L (r, t);

[0486] where: m L (r, t) is the hydrogen load of the system at r operating day t.

[0487] The system energy purchase constraints include:

[0488] P buy,min ≤ P buy (t) ≤ P buy,max ;

[0489] G buy,min ≤ G buy (r, t) ≤ G buy,max ;

[0490] The equipment output constraints include the individual equipment constraints above.

[0491] The established double-layer configuration model is solved in this example, and the specific steps for obtaining the configuration scheme include:

[0492] Step 1: Model initialization. Initialize the decision variable x of the main problem 0 , set the initial lower bound LB = E T x 0 , and the upper bound UB = +∞. Set the convergence threshold ε = 0.1%, and set the iteration count i = 0.

[0493] Step 2: Solve the main problem. In the i-th iteration, solve the main problem, i.e., the upper-layer configuration problem, to obtain the current device capacity x i , update the lower bound LB = E T x i , and pass it to the subproblem.

[0494] Step 3: Solve the subproblem. Given the device capacity x i , solve the feasible subproblem to obtain Solve the convex feasible subproblem to obtain

[0495] Step 4: Generate Benders cut plane. If the optimal value of the problem optimization objective is not equal to 0, i.e., the subproblem is infeasible, generate a feasibility cut to constrain the search space of the main problem, so that subsequent iterations avoid infeasible solutions, update the iteration count i = i + 1, and return to step 2.

[0496] If the optimal value of the problem optimization objective is equal to 0, i.e., the subproblem is feasible, solve the optimal subproblem to obtain and Generate an optimality cut, add it to the constraints of the main problem, and update the upper bound UB according to the formula.

[0497] Step 5: Convergence determination. If the upper and lower bounds satisfy the convergence condition |UB-LB| < ε, stop the iteration and output the optimal solution. If it does not converge, update the iteration count i = i + 1, and return to step 2 to continue solving the main problem.

[0498] Step 6: Output the result. Finally, output the optimal device capacity configuration x * .

[0499] Please refer to the structure schematic diagram of the computer device provided by the embodiments of the application shown in Figure 6 The computer device 400 provided by the embodiments of the application includes a processor 410 and a memory 420, the memory 420 stores a computer program executable by the processor 410, and the computer program is executed by the processor 410 to perform the method as above.

[0500] The embodiment of the present application further provides a storage medium 430, wherein the storage medium 430 stores a computer program, and the computer program is executed by the processor 410 to perform the method described above.

[0501] The storage medium 430 can be implemented by any type of volatile or nonvolatile storage devices or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0502] In the description of the present application, the terms "first", "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined as "first", "second" can explicitly or implicitly include one or more of the features. The meaning of "plurality" is two or more, unless otherwise specifically limited.

[0503] In the present application, unless otherwise specifically defined and limited, the terms "mounting", "connection", "connecting", "fixing" and the like should be understood in a broad sense, for example, can be fixed connection, can also be detachable connection, or integral; can be mechanical connection, can also be electrical connection; can be directly connected, can also be indirectly connected through an intermediate medium, can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0504] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0505] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0506] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0507] It should be understood that portions of the present application can be implemented with hardware, software, firmware or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented with software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any of the following technologies, known in the art, or their combinations can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0508] Those skilled in the art of the present technology can understand that all or part of the steps carried out by the above-mentioned embodiments can be completed by a program instructing the relevant hardware, and the program can be stored in a computer readable storage medium. When the program is executed, it includes one of the steps of the method embodiment or a combination thereof.

[0509] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it should be understood that the above-mentioned embodiments are exemplary and cannot be understood as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above-mentioned embodiments within the scope of the present application.

Claims

1. A method for optimal configuration of a comprehensive energy system considering long-term characteristics of hydrogen storage energy, characterized in that, The method comprises the following steps: constructing a multi-energy coupling mathematical model of a rural integrated energy system including hydrogen energy storage; generating a typical day containing multi-dimensional data based on a typical day generation method based on spectral joint clustering, constructing a hydrogen energy storage long-period operation model, and decomposing the hydrogen storage amount at any time within the annual operation period into an initial hydrogen storage amount and a dynamic superposition of a typical day scenario; based on the multi-energy coupling mathematical model and the hydrogen energy storage long-period operation model, constructing a double-layer optimization configuration model including capacity configuration and scheduling operation, the double-layer optimization configuration model including an upper-layer capacity configuration model and a lower-layer scheduling operation model, the upper-layer capacity configuration model taking the minimum equal annual value cost in the whole life cycle as an optimization target and taking the capacity of each device as a decision variable, and the lower-layer scheduling operation model taking the minimum annual operation cost as an optimization target and taking the output of each device and the energy storage operation strategy as a decision variable; solving the double-layer optimization configuration model by using a Benders decomposition method, splitting the double-layer optimization configuration model into a main problem and a sub-problem, and using a cut plane method to iteratively approach an optimal solution to obtain the capacity configuration and the scheduling operation strategy. 2.The method of claim 1, wherein, The multi-energy coupling mathematical model comprises: a mathematical model of an energy supply device: a power generation model of a distributed photovoltaic unit, a waste power supply model, and a natural gas power model of a biogas tank; a mathematical model of an energy conversion device: a hydrogen production model of an electrolytic cell, a hydrogen fuel cell power model, and a combined heat and power unit output model; a mathematical model of an energy storage device: a charge and discharge model of an energy storage battery, a charge and discharge heat model of a heat storage tank, and a charge and discharge hydrogen model of a hydrogen storage tank. 3.The method of claim 2, wherein, The mathematical model of the energy supply device comprises: a power generation model of a distributed photovoltaic unit: where: P PV (t) is the actual power generated by the photovoltaic plant at time t; P STC is the rated output power of the photovoltaic plant; G ING (t) is the solar radiation intensity at time t; G STC is the solar radiation intensity in standard conditions; k PV is the temperature coefficient; T c (t) is the surface temperature of the photovoltaic panel at time t; T r is the reference temperature; a waste power supply model: H WI (t) = W gar (t) a WI β WI q WI η WI ; H WG (t) = W gar (t) a WG q WG η WG ; H gar (t) = H WI (t) + H WG (t); P gar (t) = (H WI (t) + H WG (t)) η WHB (1 - η aux ); In the formula: H WI (t) and H WG (t) are the heat power generated by the incineration and gasification of household garbage respectively in the t period; H gar (t) is the heat power generated by the garbage in the t period; P gar (t) is the power generated by the garbage in the t period; W gar (t) is the amount of garbage treated in the t period; α WI and α WG are the proportions of incinerable garbage and gasifiable garbage in the household garbage respectively; β WI is the proportion of the remaining incinerable garbage after the leachate is discharged from the storage pool; q WI and q WG are the low-grade heat values of the incinerable garbage and the gasifiable garbage respectively; η WI , η WG , η WHB , η aux are the incinerator efficiency, the pyrolysis gasifier efficiency, the steam turbine power generation efficiency, and the auxiliary power rate respectively; a natural gas power model of a biogas tank: G BD (t) = V BD (t) η B2G q CH4 ; In the formula: G BD (t) is the natural gas power obtained after biogas purification treatment; η B2G is the purification coefficient of biogas to natural gas; q CH4 is the natural gas heat value, V BD (t) is the gas production model of the biogas tank; V BD (t) = a BD M BD (t) + b BD H BD (t); H BD (t) = (T z (t+1) - T z (t))V B c m ρ m + H loss (t); H loss (t) = c m M BD (t) (T z (t) - T e ) + k o A o (T z (t) - T e ); wherein: a BD is the gas production coefficient per unit of organic matter; β BD is the gas production coefficient per unit of thermal energy; H BD (t) is the thermal power supplied to the biogas plant during the period t; H loss (t) is the thermal power dissipated by the biogas plant during the period t; V B is the volume of the biogas plant; c m is the thermal capacity of the feed; p m is the density of the feed; M BD (t) is the mass of organic waste introduced into the biogas plant during the period t; A o is the area of the heat dissipation inside the plant; k o is the heat transfer coefficient; T e is the external temperature of the biogas plant. 4.The method of claim 2, wherein, The mathematical model of the energy conversion device comprises: a hydrogen production model of an electrolytic cell: wherein: m EL (t) is the mass of hydrogen produced by the electrolyzer during the time period t; η EL,H2 is the efficiency of the electrolyzer plant for producing hydrogen; P EL (t) is the power consumed by the electrolyzer during the time period t; q H is the heat value of the hydrogen a heat recovery model of an electrolytic cell: H EL (t) = η EL,h (1 - η EL,H2 )P EL (t); wherein: H EL (t) is the thermal power recovered by the electrolytic cell for the time period t; η EL,h is the heat exchange efficiency of the heat exchange device; a hydrogen fuel cell power model: P FC (t) = η FC,e m FC (t)q H ; where: P FC (t) is the hydrogen fuel cell power generation at time t; η FC,e is the hydrogen fuel cell power generation efficiency; m FC (t) is the hydrogen fuel cell hydrogen consumption at time t; a heat recovery model of a hydrogen fuel cell: H FC (t) = η FC,h m FC (t)q H ; where: H FC (t) is the hydrogen fuel cell heat production power at time t; η FC,h is the hydrogen fuel cell heat production efficiency; a combined heat and power unit output model: In the formula: G CHP (t), P CHP (t), H CHP (t) are input gas power, output electric and heat power of the CHP unit in the t period, respectively; η CHP , α CHP are electric conversion efficiency and heat energy efficiency ratio of the CHP unit, respectively. 5.The method of claim 2, wherein, The mathematical model of the energy storage device comprises: a charge and discharge model of an energy storage battery: wherein: S ES (t) represents the amount of electricity stored by the energy storage battery at time t; σ ES represents the self-discharge rate of the energy storage battery; η ES represents the charge-discharge efficiency of the energy storage battery; P ES,ch (t) and P ES,dis (t) represent the charge and discharge power of the energy storage battery at time t, respectively; Δt represents the time span from t to t+1. a charge and discharge heat model of a heat storage tank: wherein: S HS (t) represents the thermal energy stored by the thermal storage tank at time t; σ HS represents the self-heating rate of the thermal storage tank; η HS represents the charging and discharging efficiency of the thermal storage tank; H HS,ch (t) and H HS,dis (t) respectively represent the charging and discharging power of the thermal storage tank at time t; a charge and discharge hydrogen model of a hydrogen storage tank: wherein: S H2S (t) represents the hydrogen mass stored in the hydrogen storage tank at time t; σ H2S represents the hydrogen storage self-degradation rate; η H2S represents the hydrogen charging and discharging efficiency of the hydrogen storage tank; m H2S,ch (t) and m H2S,dis (t) and m (t) respectively represent the hydrogen charging amount and the hydrogen discharging amount of the hydrogen storage tank at time t. 6.The method of claim 1, wherein, The typical day generation method based on spectral joint clustering generates a typical day containing multi-dimensional data, and comprises the following steps: Let the original data matrix be where m represents days, n represents variable dimension; construct row similarity matrix Describes the multivariate time series similarity between days and days: wherein: x i is the multivariate time series feature vector on day i, x j is the multivariate time series feature vector on day j, and σ is a Gaussian kernel bandwidth parameter that controls the rate of similarity decay. Constructing a column similarity matrix Describing associations between variable-hourly joint features: where: k and l are column indices corresponding to different variable-hour combinations; x :,k is the kth column of the data matrix, representing data for all days at a certain variable-hour combination; x :,l is the lth column of the data matrix, corresponding to another different variable-hour combination; calculating a normalized Laplacian matrix for rows and columns respectively: wherein: L r , L c are the row and column normalized Laplacian matrices, respectively; D r , D c are the row and column degree matrices, respectively; To double normalization matrix X T L r X+XL c X T Perform singular value decomposition, select the first s eigenvectors to form a low-dimensional embedding space; K-means clustering is performed on rows and columns respectively in low dimensional space to get row clusters C r and column clusters C c The objective function is to minimize the residual error between row and column clusters: where: μ Cr,Cc represents the row and column cluster interaction center; X(i,j) directly represents the value of the i-th time point, j-th variable-hour feature in the original data matrix; For each cluster of rows C k , compute its multivariate time series mean containing all days as typical day: In the formula: represents the mean value of the hth hour, the vth variable in the kth typical day, representing the common mode of the cluster; x (d,h,v) represents the specific value of the dth day, the hth hour, the vth variable in the original data; calculating a typical day probability according to the proportion of intra-cluster samples: based on the selected typical day, extracting corresponding photovoltaic output, electrical load demand, thermal load demand, gas load demand, and hydrogen load demand data, and outputting s typical day data; each typical day completely retains 24-hour photovoltaic output data and the coupling characteristics of electrical, thermal, and hydrogen multi-energy loads within the day. 7.The method of claim 6, wherein, The construction of the hydrogen energy storage long-period operation model comprises: the relationship between the hydrogen storage amounts at two consecutive time points within a typical day is: ΔS H2S (k, 0) = 0; wherein: ΔS H2S (k, t) represents the hydrogen mass stored in the hydrogen storage tank at time t of k typical days, in kg, and the initial time within a day is set as 0, which is irrelevant to the actual hydrogen storage amount of the hydrogen storage tank; m H2S,ch (k, t) and m H2S,dis (k, t) respectively represent the hydrogen charging amount and the hydrogen discharging amount of the hydrogen storage tank at time t of k typical days. the initial time point hydrogen storage quality of the hydrogen storage tank between two consecutive operation days needs to meet: S H2S,0 (r+1) = (1 - σ H2S ×24) S H2S,0 (r) + ΔS H2S (k = h(r), 24); In the formula: S H2S,0 (r+1) represents the initial hydrogen storage mass of the hydrogen storage tank on the r+1 operating day; h(r) is a mapping function between the typical day k and the operating day r; in order to ensure that the hydrogen storage amount of the hydrogen storage tank is the same within a year, the following needs to be met: S H2S,0 (1) = (1 - σ H2S × 24) S H2S,0 (N r )+ ΔS H2S (k = h(N r ), 24); In the formula: N r is the number of operating days throughout the year; the actual hydrogen storage quality of the hydrogen storage tank at any time within a year can be decomposed into an initial hydrogen storage amount and a dynamic superposition of a typical day scenario, and is expressed as: S H2S (r,t) = S H2S,0 (r) + ΔS H2S (k = h(r), t); In the formula, S H2S (r, t) represents the real hydrogen mass stored by the hydrogen storage tank at time t on the rth operating day, which is the sum of the initial hydrogen storage amount on the rth operating day and the hydrogen mass stored at time t on the rth operating day. the hydrogen gas stored in the hydrogen storage tank needs to meet the upper and lower limit constraints: α H2S,min S H2S,cap ≤S H2S (r,t)≤α H2S,max S H2S,cap ; the constraints are processed as follows: where ΔS H2S,max (k) and ΔS H2S,min (k) represent the maximum and minimum values of the hydrogen storage mass of the hydrogen storage tank on the k typical day, respectively, in kg. 8.The method of claim 1, wherein, The upper layer capacity configuration model has a target function minC of minimizing the life cycle equivalent annual cost as an optimization target: minC = C inv +C op +C scrap +C loss ; where C inv is the equal annual investment cost, C op is the operating cost, C scrap is the retirement cost, and C loss is the attrition cost. where: y represents the useful life of the equipment; r is the discount rate; μ i is the unit capacity investment cost for the ith equipment; S i Xi represents the configuration capacity of the ith device; X represents a set of devices in the system that perform capacity configuration, including waste power generation, biogas fermentation, electrolytic tank, hydrogen fuel cell, combined heat and power unit, energy storage battery, heat storage tank, and hydrogen storage tank; where s is the total number of typical days in a year; p k is the probability of k typical days occurring in a year; N r is the number of operating days in a year; C om,k is the operation and maintenance cost of a typical day, C buy,k is the cost of purchasing energy, C CO2,k is the cost of carbon trading, C waste,k is the waste disposal subsidy, C sell,k is the income from selling energy. 9.The method of claim 8, wherein, The upper layer capacity configuration model further includes the following constraints: Device installation capacity limit constraints: 0 < S i ≤ S i,max ; In the formula, S i S represents the installation capacity of the ith device i,max S represents the upper limit of the capacity that the ith device can allow to install 10.The method of claim 8, wherein, The lower layer scheduling operation model takes the minimum annual operation cost as the optimization objective of the objective function minC op is: minC op = C om + C buy + C CO2 - C waste - C sell ; The objective function includes the annual operation cost C of the system om , the cost of purchasing energy C buy , the cost of carbon trading C CO2 , the subsidy for waste treatment C waste , and the income from selling energy C sell ; Wherein: where λ i is the unit capacity operation and maintenance cost of device i; P i (k, t) represents the output power of the ith device at typical day k period t; X represents the set of devices in the system, including photovoltaic units, waste power generation, biogas fermentation, electrolytic cells, hydrogen fuel cells, combined heat and power units, electric boilers, energy storage batteries, heat storage tanks, and hydrogen storage tanks; where: P buy (k,t) and G buy (k,t) are the typical daily exogenous electricity and natural gas power at time period t of day k, respectively. wherein: W gar (k, t) is the amount of waste treated in period t of typical day k; M BD (r, t) is the mass of organic waste fed to the biogas plant in period t of typical day k; where: m sell (k, t) is the system hydrogen sales volume for typical day k, time period t. 11.The method of claim 10, wherein, The lower layer scheduling operation model further includes the following constraints: Electric power balance constraints: P buy (r, t) + P PV (r, t) + P gar (r, t) + P FC (r, t) + P CHP (r, t) + P ES,dis (r, t) + P = P EL (r, t) + P EB (r, t) + P ES,ch (r, t) + P L (r, t) wherein: P L (r,t) is the electrical load of the system at the rth day and the tth period, P buy (r,t) is the purchased electrical power at the rth day and the tth period, P PV (r,t) is the actual power generation of the photovoltaic unit at the rth day and the tth period; P gar (r,t) is the power generation of the waste power plant at the rth day and the tth period; P FC (r,t) is the power generation of the hydrogen fuel cell at the rth day and the tth period; P CHP (r,t) is the input gas power of the CHP unit at the rth day and the tth period, P ES,ch (r,t) and P ES,dis (r,t) represent the charging and discharging power of the energy storage battery at the rth day and the tth period, respectively; P EL (r,t) is the power consumption of the electrolytic cell at the rth day and the tth period; Thermal power balance constraints: H gar (r, t) + H CHP (r, t) + H EL (r, t) + H FC (r, t) + H EB (r, t) + H HS,dis (r, t); = H BD (r, t) + H HS,ch (r, t) + H L (r, t) wherein H L (r, t) is the heat load of the system at time period t of day r; H gar (r, t) is the heat generation power of the garbage at time period t of day r; H FC (r, t) is the heat generation power of the hydrogen fuel cell at time period t of day r; H CHP (r, t) is the heat power of the CHP unit at time period t of day r, H HS,ch (r, t) and H HS,dis (r, t) respectively represent the heat storage and release power of the heat storage tank at time period t of day r; H EL (r, t) is the heat power recovered by the electrolytic tank at time period t of day r; H BD (r, t) is the heat power supplied to the biogas tank at time period t of day r; Natural gas balance constraints: G buy (r,t)+G BD (r,t)=G CHP (r,t)+G L (r,t); In the formula, G L (r, t) is the gas load of the system at r operation day t period, G buy (r, t) is the purchased natural gas power at r operation day t period; G BD (r, t) is the natural gas power obtained after purification treatment of biogas at r operation day t period; G CHP (r, t) is the input gas power of the CHP unit at r operation day t period; Hydrogen energy balance constraints: m EL (r, t) + m H2S,dis (r, t) = m FC (r, t) + m sell (r, t) + m H2S,ch (r, t) + m L (r, t); wherein: m L (r, t) is the hydrogen load of the system at time period t of day r; m EL (r, t) is the hydrogen mass produced by the electrolyzer at time period t of day r; m H2S,ch (r, t) and m H2S,dis (r, t) respectively represent the hydrogen charging and discharging mass of the hydrogen storage tank at time period t of day r; m FC (r, t) is the hydrogen mass consumed by the hydrogen fuel cell at time period t of day r; m sell (r, t) is the hydrogen mass sold by the system at time period t of day r; System energy purchase constraints: P buy,min ≤P buy (t)≤P buy,max ; G buy,min ≤G buy (r,t)≤G buy,max ; Each device operation constraint, which is also mapped to the constraint of a typical day k. 12.The method of claim 1, wherein, The Benders decomposition method is used to solve the double-layer optimization configuration model, the double-layer optimization configuration model is split into a master problem and a sub-problem, and a cutting plane method is used to iteratively approximate the optimal solution, including the following steps: Step 1: Initialize decision variables x of the master problem 0 , set initial lower bound LB = E T x 0 , upper bound UB = +∞; set convergence threshold ε, set iteration count i = 0; Step 2: In the i-th iteration, solve the master problem, i.e., the upper level configuration problem to obtain the current device capacities x i , update the lower bound LB = E T x i , and pass it to the subproblem; Step 3: Solve the feasible subproblem for z i In the given case, solve the feasible subproblem to obtain z k i , solve the convexified feasible subproblem to obtain λ k i ; Step 4: If the optimal value of the optimization target of the problem is not equal to 0, i.e., the sub-problem is infeasible, a feasible cutting plane is generated to constrain the search space of the master problem, so that subsequent iterations avoid infeasible solutions, and the iteration number i is updated as i=i+1 and returns to step 2; If the optimal value of the problem optimization objective equals 0, i.e. the subproblem is feasible, solve the optimal subproblem to get y k i and μ k i generate an optimality cut plane, add it to the master problem constraints, and update the upper bound UB according to the equation Step 5: If the upper and lower bounds satisfy the convergence condition |UB-LB|<ε, stop iteration and output the optimal solution; if not converged, update the iteration number i=i+1 and return to step 2 to continue solving the master problem; Step 6: Final output optimal device capacity configuration x * .

13. A system for optimizing the configuration of a comprehensive energy system taking into account the medium and long-term characteristics of hydrogen storage, characterized by The system comprises the method according to any one of claims 1 to 12. A mathematical model construction unit is configured to construct a multi-energy coupling mathematical model of a rural integrated energy system with hydrogen energy storage; A typical day generation unit is configured to generate a typical day containing multi-dimensional data based on a spectral joint clustering typical day generation method, and to construct a hydrogen energy storage long-period operation model, which decomposes the hydrogen storage amount at any time within the annual operation period into an initial hydrogen storage amount and a dynamic superposition of a typical day scenario; A double-layer model construction unit is configured to construct a double-layer optimization configuration model containing capacity configuration and scheduling operation based on the multi-energy coupling mathematical model and the hydrogen energy storage long-period operation model, the double-layer optimization configuration model including an upper layer capacity configuration model and a lower layer scheduling operation model, the upper layer capacity configuration model having a life cycle equivalent annual cost minimization as an optimization target and taking device capacities as decision variables, and the lower layer scheduling operation model having an annual operation cost minimization as an optimization target and taking device outputs and energy storage operation strategies as decision variables; A solving unit is configured to solve the double-layer optimization configuration model using a Benders decomposition method, split the double-layer optimization configuration model into a master problem and a sub-problem, and use a cutting plane method to iteratively approximate the optimal solution to obtain capacity configuration and scheduling operation strategies.

14. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method according to any one of claims 1 to 12.

15. A storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement the method according to any one of claims 1 to 12.

Images