Comprehensive energy system low-carbon economic configuration method considering electric vehicle access

Abstract

The invention relates to a low-carbon comprehensive energy system configuration method considering electric vehicle access, which comprises the following steps of: accessing an EV charging load model into an energy system IES model to obtain a comprehensive energy system; constructing an upper-layer capacity configuration model by taking the lowest cost of the integrated energy system as a target; constructing a lower-layer optimization operation model by taking the minimum daily operation cost of the integrated energy system as a target, and setting constraint conditions of the lower-layer optimization operation model; the constraint conditions comprise an electric energy balance constraint, a heat energy balance constraint, a cold energy balance constraint, a gas energy balance constraint, a unit climbing constraint, an energy storage constraint and an EV demand response constraint; and substituting an upper layer decision variable of the upper layer capacity configuration model into the lower layer optimization operation model by adopting a Pedal optimization algorithm, and calling a CPLEX solver to solve the lower layer optimization operation model to obtain an optimal operation scheme and daily operation cost. According to the invention, the low-carbon economic configuration of the IES accessed to the EV is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated energy system optimization and scheduling, and in particular to a low-carbon economic configuration method for an integrated energy system taking into account the access of electric vehicles. Background Art

[0002] IES, as a multi-energy coupling system with efficient energy utilization and promoting green and sustainable social development, will become an important energy application scenario in the future. At the same time, the low carbon nature of EV enables it to develop rapidly and gradually become an energy-consuming device that cannot be ignored on the user side. Therefore, it is of certain practical significance to incorporate EV into the planning and operation of IES. EV has the dual attributes of being a load or energy storage. Formulating a reasonable response mechanism and guiding its charging and discharging behavior can balance the peak-to-valley difference of the system and achieve peak-to-valley filling of the load curve. In addition, the low carbon nature of the system is also a problem worthy of attention. The addition of a step-by-step carbon trading mechanism is conducive to achieving low-carbon emissions of the system. The complexity of IES multi-energy flow coupling brings challenges to the solution of its model. Selecting a high-performance algorithm for solution is conducive to improving the solution efficiency and solution accuracy of the model. Traditional swarm intelligence algorithms, such as genetic algorithms, particle swarm algorithms, ant colony algorithms, etc., all have the problem of being easily trapped in local optimality, and low solution efficiency and solution accuracy. At present, there is no invention that applies the gazelle algorithm to the complex model of solving the integrated energy system. Therefore, the present invention selects the gazelle, a high-performance swarm intelligence algorithm, to solve the model. Summary of the invention

[0003] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present invention is to simulate the charging load data of EV based on the Monte Carlo algorithm. On this basis, through reasonable guidance, the charging and discharging behavior of EV is optimized, and the low-carbon emissions of the system are taken into account, and finally the low-carbon economic configuration of IES considering the access of EV is realized. Specifically, it includes: the construction of IES model considering EV access; EV charging load model and simulation; V2G (vehicle to grid, V2G) response model under dual incentive mechanism; ladder carbon trading mechanism model; establishment of two-level programming model; solution of two-level programming model.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A low-carbon economic configuration method for an integrated energy system considering electric vehicle access includes:

[0006] Obtaining an electric vehicle (EV) charging load model, connecting the EV charging load model to an energy system IES model, and obtaining a comprehensive energy system;

[0007] An upper capacity configuration model is constructed with the goal of minimizing the cost of the integrated energy system; the cost of the integrated energy system includes: system equipment investment cost and unit cycle operation cost;

[0008] With the goal of minimizing the daily operating cost of the integrated energy system, a lower-level optimization operation model is constructed. The daily operating cost of the integrated energy system includes: energy purchase cost, equipment operation and maintenance cost, EV demand response cost and carbon trading cost, and constraints of the lower-level optimization operation model are set; the constraints include: introducing electric energy balance constraints, thermal energy balance constraints, cooling energy balance constraints, gas energy balance constraints, unit ramping constraints, energy storage constraints and EV demand response constraints for user responsiveness under different incentive levels;

[0009] The Gazelle optimization algorithm is used to substitute the upper-level decision variables of the upper-level capacity configuration model into the lower-level optimization operation model, and the CPLEX solver is called to solve the lower-level optimization operation model to obtain the optimal operation plan and daily operation cost.

[0010] Optionally, constructing the upper layer capacity configuration model includes:

[0011] minF=F T +F Y ;

[0012]

[0013] Where, F is the total system cost; F T is the investment cost of the equipment in the system; F Y is the system operation cost per unit period; N is the number of equipment types; C i,c is the capacity of the i-th equipment; C i,p is the unit price of capacity of device i; r is the discount rate; Y is the device life cycle; s represents the number of typical scenarios; num s is the number of days corresponding to the typical scenario; F Y,s is the daily operating cost in the scenario.

[0014] Optionally, constructing the lower layer optimization operation model includes:

[0015] min F Y,s =C P +C M +C C +C DR ;

[0016]

[0017] C DR =s c ·C ev,c +s d·C ev,d ;

[0018] Among them, F Y,s is the typical daily operating cost; C P C is the energy purchase cost; M is the equipment operation and maintenance cost; C DR is the EV demand response cost; C C is the carbon trading cost; p t is the electricity price at time t; q t is the gas price at time t; is the gas purchase volume at time t; m i is the unit operation and maintenance cost of equipment i; P i,t is the power of the i-th type of equipment at time t; C ev,c is the charge response capacity; C ev,d is the discharge response capacity.

[0019] Optionally, setting the constraint conditions of the lower layer optimization operation model includes:

[0020] Setting the power balance constraint includes:

[0021] Introducing user responsiveness under different incentive levels, adjusting the charging and discharging power of the EV charging load model, and establishing the power balance constraint:

[0022]

[0023] in, are the charging and discharging power of EES at time t; P t PV , P t WT are the output power of PV and WT at time t respectively; P is the charging and discharging power of a single EV at time t; t EL is the electrical load at time t; is the power consumption of EB at time t; is the power consumption of EC at time t; is the electrical output power of CHP at time t; ω(i) t is the response status of EV at the current moment;

[0024] Setting the thermal energy balance constraint includes:

[0025]

[0026] in, are the charging and discharging powers of HES respectively; P EB,t is the heat generation power of EB at time t; P tHL is the heat load at time t; is the AC heat consumption power at time t, P GB,t is the thermal output power of GB at time t, is the thermal output power of CHP at time t;

[0027] Setting the cooling energy balance constraint includes:

[0028]

[0029] Among them, P AC,t , P EC,t are the cooling power of AC and EC at time t respectively; P is the CES charging and cooling power at time t; t CL is the cooling load at time t.

[0030] Optionally, adjusting the charging and discharging power of the EV charging load model includes:

[0031] The user responsiveness is used to describe the fuzzy mapping relationship between the user's response intention, response state and responsiveness, and the charging and discharging power of the EV charging load model is adjusted according to the fuzzy mapping relationship;

[0032] Describing the fuzzy mapping relationship includes:

[0033]

[0034] Where ω(i) is the response state of user i, w(i) = 0 represents no response, ω(i) = 1 represents response; σ(i) is the user responsiveness; σ T is the user responsiveness threshold;

[0035] Acquiring the user responsiveness includes:

[0036]

[0037] Among them, k r is the slope of the response changing with the stimulus level; x jl,max is the excitation level under response saturation; Δσ(i) is the maximum deviation of the response.

[0038] Optionally, setting the constraint conditions of the lower layer optimization operation model also includes:

[0039] Setting the gas energy balance constraint includes:

[0040]

[0041] in, is the gas purchase volume at time t; Pt QL is the gas load at time t; is the gas consumption power of CHP at time t; is the gas consumption power of GB at time t;

[0042] Setting the unit climbing constraint includes:

[0043]

[0044] in, is the output power of device i at time t; is the uphill and downhill power limit of device i;

[0045] Setting the energy storage constraint includes:

[0046]

[0047] Where: is the energy storage power of energy storage device i at time t; is the charging and discharging power of energy storage device i at time t; η i,ch , η i,dis is the charging and discharging efficiency of energy storage device i at time t; ε i is the self-discharge loss coefficient of energy storage device i; are the upper and lower limits of the capacity of energy storage device i;

[0048] Setting the EV demand response constraint includes:

[0049]

[0050] C ev,min ≤C ev,t ≤C ev,max

[0051]

[0052] Among them, C ev,t is the battery capacity of EV at time t; is the charging and discharging efficiency of the EV; is the charging and discharging power of EV at time t; T ch , T dis C is the charging and discharging time of EV; ev,max , C ev,min The upper and lower limits of EV battery capacity; is the maximum charging and discharging power of EV; Δt is the interval between charging and discharging; C ev,d is the discharge response capacity; C ev,c The charging response capacity.

[0053] Optionally, obtaining the optimal operation plan and daily operation cost includes:

[0054] Step 1: Initialize the gazelle population;

[0055] Step 2: The predicted load, wind and solar output, and initial configuration values ​​of equipment capacity are used as upper-level decision variables, and the upper-level decision variables of the upper-level capacity configuration model are substituted into the lower-level optimization operation model, and the lower-level optimization operation model is constrained. According to the equipment capacity parameters, the CPLEX solver is called to solve the lower-level optimization operation model to obtain the optimal operation plan and individual daily operation cost of the population;

[0056] Step 3, judging whether the individual is less than the maximum number of iterations, if the individual is less than the maximum number of iterations, re-calling the CPLEX solver to solve the lower optimization operation model, if the individual is greater than the maximum number of iterations, calculating the objective function value and constructing an elite matrix;

[0057] Step 4, determine whether the random number r is less than the target value. If the random number r is less than the target value, perform development operations in combination with the elite matrix. If the random number r is greater than the target value, update the gazelle position, top gazelle and elite matrix.

[0058] Step 5: Use the Predator Success Rate (PSR) s Update the gazelle position again and determine whether the updated gazelle position meets the termination condition. If so, output the optimal solution. If not, repeat step 2 until the termination condition is met.

[0059] Optionally, development operations in conjunction with the elite matrix include:

[0060] g i+1 =g i +s·R·R B ·(E i -R B ·g i )

[0061] Among them, g i+1 is the next iteration solution; g i is the current iterative solution; s is the gazelle's predation speed; R B is the random number vector of Brownian motion; R is the uniform random number vector in [0,1]; E i For the elite matrix.

[0062] Optionally, updating the gazelle position comprises:

[0063] Determine whether the number of iterations is an odd number. If the number of iterations is an odd number, use the first exploration formula to update the gazelle position:

[0064] g i+1 =g i +S·μ·R·R L ·(E i -R L ·g i );

[0065] Among them, g i+1 is the next iteration solution; g i is the current iterative solution; μ is a constant; S is the maximum speed that the gazelle can reach;

[0066] If the number of iterations is an even number, the second exploration formula is used to update the gazelle position:

[0067] g i+1 =g i +S·μ·CF·R·R B ·(E i -R L ·g i );

[0068] Where S is the maximum speed that the gazelle can reach; R is a uniform random number vector in [0,1]; R B is the random number vector of Brownian motion; R L is a random number vector based on Levy distribution.

[0069] Optionally, using the predator success rate PSR s Updated Gazelle positions again include:

[0070]

[0071] Among them, g i+1 is the next iteration solution; g i is the current iterative solution; UB, LB are the upper and lower bounds of the population; U is a binary vector; r is a random number in [0,1]; g r1 ,g r2 is the random index of the gazelle matrix; R is a uniform random number vector in [0,1]; LB is the lower bound of the population.

[0072] The beneficial effects of the present invention are:

[0073] The integrated energy system in the present invention has the characteristics of improving the utilization rate of renewable energy and promoting the efficient utilization of traditional energy. It will become an important form of energy supply in the future. As a green and low-carbon means of transportation, the ownership of EV has been rapidly developed. Incorporating EV into the optimal configuration and operation of the integrated energy system has certain practical significance. The present invention uses the Monte Carlo simulation method to simulate the travel behavior of EV very well. On this basis, a dual incentive response mechanism that can fully consider the user's response willingness is introduced to optimize the charging and discharging behavior of EV. At the same time, a step-by-step carbon trading mechanism is introduced to achieve low-carbon emissions of the system. In view of the model complexity of the multi-energy flow coupling of the integrated energy system, the gazelle algorithm, a high-performance swarm intelligence algorithm, is selected to improve the solution efficiency and solution accuracy for complex models. The present invention is verified by simulating the constructed model. The simulation results show that the present invention can well achieve the low-carbon economic configuration of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0075] Figure 1 A schematic diagram of a comprehensive energy system model considering electric vehicle access according to an embodiment of the present invention;

[0076] Figure 2 This is a flow chart of the Gazelle algorithm solution model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0079] This embodiment discloses a low-carbon economic configuration method for an integrated energy system considering the access of electric vehicles. First, an IES model considering the access of EV charging load is built, and then a Monte Carlo algorithm is used to simulate the disordered charging load of EV. On this basis, a V2G response model under a dual incentive mechanism is established. In view of the problems of low solution efficiency and solution accuracy of traditional algorithms such as particle swarm optimization and genetic algorithm, a high-performance gazelle optimization algorithm is used to solve the upper model, and the CPLEX solver is called to solve the lower optimization operation model, thereby obtaining the optimal configuration plan and daily operation cost.

[0080] This embodiment provides a low-carbon economic configuration method for an integrated energy system, the main contents of which are as follows:

[0081] 1. Build an IES model considering EV access:

[0082] The energy equipment included in the IES built in this embodiment is as follows: small wind turbine (WT), photovoltaic (PV), combined heating and power (CHP), gas boiler (GB), absorption cooler (AC), electric cooler (EC), electric boiler (EB), battery (electric energy storage, EES), cold energy storage (CES) and heat energy storage (HES).

[0083] In the IES constructed in this embodiment, energy is first supplied by the small WT and PV units established in the park, and the remaining energy demand is met by purchasing energy from the power grid and gas grid through IES. IES supplies energy for four types of loads on the user side: electricity, cooling, heat, and gas, and incorporates EV into the system operation. The cooling load is supplied by AC, EC, and CES equipment, the heating load is supplied by CHP, GB, EB, and HES equipment, the electric load is mainly supplied by PV, WT, CHP, EES, and the power grid, and the gas load is supplied by the system purchasing energy from the gas grid. EV has the dual flexibility of load and energy storage. It can be connected to IES as a deterministic load for charging, or it can be used as an energy storage device to reverse power the system. Consider the IES model for EV access as follows: Figure 1 shown.

[0084] 2.EV charging load model and simulation:

[0085] According to the survey statistics of the US Department of Transportation on household cars, the time it takes for users to go home and the daily mileage follow the normal distribution probability density function, and the travel behavior of EVs can be replaced by the travel behavior of fuel vehicles. Monte Carlo simulation is a simulation statistical method, and its actual application scenario is to simulate unavailable data. This method is used to model the EV group load in this embodiment.

[0086] 3. Build a V2G response model under the dual incentive mechanism:

[0087] When EVs participate in demand response, they are not only constrained by objective response capabilities, but also affected by the subjective will of users. This embodiment considers that the EV type is a private car, which has a longer stay time to participate in system scheduling after a day of travel. Some studies only guide EV charging and discharging through price-based demand response (DR), but price-based DR has a limited guiding effect on EV charging and discharging activities. It is necessary to combine incentive-based DR to optimize EV charging and discharging behavior.

[0088]

[0089] Where: x jl is the incentive level; α and β are the attractiveness coefficients of charging and discharging subsidies; s c Subsidy for charging; d To provide subsidies for discharge; The cost of charging loss; Discharge loss cost.

[0090] Charging subsidies can guide users' charging response, and the subsidy prices are set as follows:

[0091]

[0092] Where: This is the saturation charging subsidy. When this value is exceeded, the economy of the system will decrease.

[0093] The discharge subsidy can guide the user's discharge response and take into account the compensatory charging behavior caused by EV discharge. The subsidy price is set as follows:

[0094]

[0095] Where: This is a saturation discharge subsidy. When this value is exceeded, the economy of the system will decrease. is the compensatory charging cost; ω is the charging electricity price.

[0096] This embodiment uses responsiveness to describe the user's response willingness. The fuzzy mapping relationship between the user's response state and responsiveness is as follows:

[0097]

[0098] Where: ω(i) is the response state of user i, w(i) = 0 represents no response, ω(i) = 1 represents response; σ(i) is the user responsiveness; σ T is the user responsiveness threshold.

[0099] The user responsiveness under different incentive levels can be expressed as:

[0100]

[0101] Where: k r is the slope of the response changing with the stimulus level; x jl,max is the excitation level under response saturation; Δσ(i) is the maximum deviation of the response.

[0102] The changing rule of Δσ(i) is shown as follows:

[0103]

[0104] Where: k1 and k2 are the proportional coefficients of Δσ(i) changes before and after the incentive price produces a strong guiding effect, respectively, IP For x jl,max is the maximum excitation level.

[0105] The user's response is usually down (i)≤σ(i)≤σ up (i). In practical applications, historical data can be modeled and analyzed based on their distribution patterns. This embodiment uses uniform distribution to simply describe it:

[0106] σ(i)~U(σ down (i),σ up (i))(8)

[0107] 4. Build a ladder carbon trading mechanism model:

[0108] At present, my country mainly allocates free initial carbon quotas to market entities with emission reduction obligations through government regulatory departments, and market entities participate in the carbon trading market to freely trade carbon quotas based on their actual carbon emissions. The baseline method is used to determine the system carbon quota. In this example, carbon emissions mainly come from electricity purchases, CHP equipment and gas boilers.

[0109] The allocation of free carbon emission allowances under carbon trading is as follows:

[0110]

[0111]

[0112] E=Ee +E CHP +E GB (12)

[0113] Among them, E e 、E CHP 、E GB The free purchase of electricity, CHP equipment and gas boilers

[0114] Carbon emission quota, E is the total amount of free carbon emissions, q1 is the free carbon emissions per unit of electricity, is the amount of electricity purchased by the system at time t; θ is the free carbon emission per unit heat; ξ is the conversion coefficient of the power generation of the CHP equipment to the heat generation, is the thermal and electrical output power of CHP at time t, P GB,t is the thermal output power of GB at time t, T is, and Δt is the operating time of CHP and GB.

[0115] The carbon trading cost calculation model is as follows:

[0116] C C =ρ c (E′-E) (13)

[0117]

[0118] Where: E′ is the actual total carbon emissions, which is calculated in the same way as E, except that the carbon emission coefficient values ​​used in equations (9), (10), and (11) are different; C C is the carbon trading cost of the system; ρ c is a stepped carbon price; is the carbon trading base price; l is the carbon emission range; m is the carbon price growth coefficient; u is the interval number.

[0119] 5. Build IES two-layer planning model:

[0120] 5.1 Upper layer capacity configuration model:

[0121] 5.1.1 Objective function:

[0122] The upper capacity configuration stage aims to minimize the total system cost, which mainly includes the system equipment investment cost and the operating cost per unit period, namely:

[0123] minF=F T +F Y (15)

[0124]

[0125]

[0126] Where: F is the total system cost; F T is the investment cost of the equipment in the system; F Y is the system operation cost per unit period; N is the number of equipment types; C i,c is the capacity of the i-th equipment; C i,p is the capacity unit price of device i; r is the discount rate, set to 5%; Y is the device life cycle; s represents the number of typical scenarios; num s is the number of days corresponding to the typical scenario; F Y,s is the daily operating cost under scenario s.

[0127] 5.2 Lower layer optimization operation model:

[0128] 5.2.1 Objective function:

[0129] Based on the initial capacity of the equipment, the PIES lower-level optimization operation model aims to minimize the daily operating cost:

[0130] min F Y,s =C P +C M +C C +C DR (18)

[0131]

[0132] C DR =s c ·C ev,c +s d ·C ev,d (twenty one)

[0133] Where: F Y,s is the typical daily operating cost; C P C is the energy purchase cost; M is the equipment operation and maintenance cost; C DR is the EV demand response cost; p t is the electricity price at time t; q t is the gas price at time t; is the gas purchase volume at time t; m i is the unit operation and maintenance cost of equipment i; P i,t is the power of the i-th type of equipment at time t; C ev,c is the charge response capacity; C ev,d is the discharge response capacity.

[0134] 5.2.2 Constraints:

[0135] 1) Power balance constraints:

[0136]

[0137] Where: are the charging and discharging power of EES at time t; P t PV , P t WT are the output power of PV and WT at time t respectively; is the charging and discharging power of a single EV at time t, ω(i) t is the user's response state at time t, w(i) = 0 represents no response, ω(i) = 1 represents response, and the user's responsiveness under different incentive levels can be obtained by combining equations (5), (6), and (7). By changing the incentive level, the user's responsiveness can be adjusted, thereby changing the charging and discharging power of the EV load; P t EL is the electrical load at time t; is the power consumption of EB at time t; is the power consumption of EC at time t.

[0138] 2) Thermal energy balance constraints:

[0139]

[0140] Where: are the charging and discharging powers of HES respectively; P EB,t is the heat generation power of EB at time t; P t HL is the heat load at time t; is the AC heat consumption power at time t.

[0141] 3) Cold energy balance constraint:

[0142]

[0143] Where: P AC,t , P EC,t are the cooling power of AC and EC at time t respectively; P is the CES charging and cooling power at time t; t CL is the cooling load at time t.

[0144] 4) Gas energy balance constraints:

[0145]

[0146] Where: P t QL is the gas load at time t; is the gas consumption power of CHP at time t; is the gas consumption power of GB at time t.

[0147] 5) Unit climbing constraints:

[0148]

[0149] Where: is the output power of device i at time t; is the uphill and downhill power limit of device i.

[0150] 6) Energy storage constraints:

[0151]

[0152] Where: is the energy storage power of energy storage device i at time t; is the charging and discharging power of energy storage device i at time t; η i,ch , η i,dis is the charging and discharging efficiency of energy storage device i at time t; ε i is the self-discharge loss coefficient of energy storage device i; are the upper and lower limits of the capacity of energy storage device i.

[0153] 7) EV demand response constraints:

[0154] Before the vehicles participating in the system optimization dispatch travel on the next day, the EV battery power must meet the user's expectations. The EV meets the battery capacity limit constraints and the maximum charging and discharging power constraints.

[0155]

[0156] C ev,min ≤C ev,t ≤C ev,max (31)

[0157]

[0158] Where: C ev,t is the battery capacity of EV at time t; is the charging and discharging efficiency of the EV; is the charging and discharging power of EV at time t; T ch , T dis C is the charging and discharging time of EV; ev,max , C ev,min The upper and lower limits of EV battery capacity; It is the maximum charging and discharging power of EV.

[0159] 6. Solve the two-level programming model:

[0160] For the two-layer optimization configuration model proposed in this embodiment, the upper model is solved by using the gazelle optimization algorithm, and the lower model is solved by calling the CPLEX solver by MATLAB.

[0161] 6.1 Gazelle Optimization Algorithm:

[0162] For the study of IES capacity configuration, a large number of literatures have adopted traditional algorithms such as particle swarm and genetics for optimization and solution. This embodiment uses the gazelle optimization algorithm proposed in 2023 to solve the established two-layer configuration model. The gazelle optimization algorithm (GOA) is a novel, simple, and powerful global random optimization algorithm. GOA is explored and developed based on the core idea of ​​gazelle escaping predators. Its principle is as follows:

[0163] 1) Initialization:

[0164] GOA uses randomly initialized gazelles as search individuals, and the search individuals are n×d matrices of candidate solutions, expressed as follows:

[0165]

[0166] Where: X is the position vector matrix of the candidate population; x i,j is the vector position randomly generated in the jth dimension of the ith population; i = 1, 2, ..., n; j = 1, 2, ..., d, where n is the population size; d is the dimension. In each iteration, each individual x i,j Generate a candidate solution. Name the optimal solution the top gazelle and construct an elite matrix E i , in order to carry out the next search, the expression is as follows:

[0167]

[0168] Where: x i ' ,j is the position of the gazelle at the top. GOA assumes that both predators and gazelles are searching individuals and that when a predator is seen stalking a gazelle, they move in the same direction.

[0169] 2) Development phase:

[0170] This stage is the foraging stage, and the gazelle performs Brownian motion. Its expression is as follows:

[0171] g i+1 =g i +s·R·R B ·(E i -R B ·g i ) (36)

[0172] Where: g i+1 is the next iteration solution; g i is the current iterative solution; s is the gazelle's predation speed; R B is the random number vector of Brownian motion; R is the uniform random number vector in [0,1]; E i For the elite matrix.

[0173] 3) Exploration phase:

[0174] Assume that each iteration has a sudden change in direction. When the number of iterations is odd, the gazelle moves in one direction; when the number of iterations is even, the gazelle moves in another direction. The gazelle reacts first, while the predator reacts slower, so it is assumed that it first runs in Brownian motion and then moves in Levy flight. When the gazelle finds the predator, its behavior is as follows:

[0175] g i+1 =g i +S·μ·R·R L ·(E i -R L ·g i ) (37)

[0176] Where: μ is a constant; S is the maximum speed that the gazelle can reach; R L is a random number vector based on Levy distribution. The behavior of the predator chasing the gazelle is as follows:

[0177] g i+1 =g i +S·μ·CF·R·R B ·(E i -R L ·g i ) (38)

[0178]

[0179] Where: iter is the current number of iterations; max_iter is the maximum number of iterations.

[0180] Set PSR s Represents the success rate of the predator, which can be used to s To prevent GOA from falling into the local optimal value, the position update formula is as follows:

[0181]

[0182] Where: UB, LB are the upper and lower bounds of the population; U is a binary vector; r is a random number in [0,1]; g r1 ,g r2is the random index of the Gazelle matrix.

[0183] Furthermore, obtaining the optimal operation plan and daily operation cost includes:

[0184] Step 1: Initialize the gazelle population;

[0185] Step 2: Substitute the upper-level decision variables of the upper-level capacity configuration model into the lower-level optimization operation model, constrain the lower-level optimization operation model, and call the CPLEX solver to solve the lower-level optimization operation model according to the equipment capacity parameters to obtain the optimal operation plan and individual daily operation cost of the population;

[0186] Step 3: Determine whether the individual number is less than the maximum number of iterations. If so, call the CPLEX solver again to solve the lower optimization operation model. If the individual number is greater than the maximum number of iterations, calculate the objective function value and construct the elite matrix.

[0187] Step 4: Determine whether r is less than the target value. If r is less than the target value, perform development operations in combination with the elite matrix. If r is greater than the target value, update the gazelle position, top gazelle and elite matrix.

[0188] The formula for updating the top gazelle is:

[0189] The gazelle algorithm uses randomly initialized gazelles as search individuals, and the search individuals are n×d matrices of candidate solutions, expressed as follows:

[0190]

[0191] Where: X is the position vector matrix of the candidate population; x i,j is the vector position randomly generated in the jth dimension of the ith population; i = 1, 2, ..., n; j = 1, 2, ..., d, where n is the population size; d is the dimension. In each iteration, each individual x i,j Generate a candidate solution. Name the optimal solution the top gazelle and construct an elite matrix E i , in order to carry out the next search, the expression is as follows:

[0192]

[0193] Where: x i ' ,j For the location of the top gazelle;

[0194] Step 5: Use the Predator Success Rate (PSR) s Update the gazelle position again and determine whether the updated gazelle position meets the termination condition. If so, output the optimal solution. If not, repeat step 2 until the termination condition is met.

[0195] 6.2 Two-layer model solution process, such as Figure 2 :

[0196] For the two-layer model of this embodiment, a two-layer optimization algorithm with GOA and CPLEX solvers in the upper and lower layers is used for solving the problem. The specific steps are as follows:

[0197] 1) Input the wind and solar power output and load forecast values ​​for each typical day, electricity and gas prices, and utility parameters of energy coupling equipment;

[0198] 2) Initialize the parameters of the gazelle algorithm, set the number of algorithm populations, the maximum number of iterations, and the upper and lower limits of the search space of the decision variables;

[0199] 3) Population initialization: randomly initialize the gazelle population;

[0200] 4) Determine whether the termination condition is met. If so, output the optimal solution and operation plan; if not, continue to execute step 5);

[0201] 5) Substitute the upper-level decision variables into the lower-level ones to constrain the lower-level optimization model;

[0202] 6) Call the CPLEX solver to solve the lower-level model and obtain the optimal operation plan and daily operation cost;

[0203] 7) Calculate the objective function value and construct the elite matrix for the next search;

[0204] 8) Determine whether r is less than 0.5. If so, perform development operations. If not, determine whether the number of iterations is an odd number. If so, perform exploration operations using formula (37). If so, perform exploration operations using formula (38).

[0205] 9) Update the gazelle position, top gazelle and elite matrix, and use PSRs and formula (39) to update the gazelle position again to prevent it from falling into the local optimal solution;

[0206] 10) Go to step 4).

[0207] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A low-carbon economic configuration method for an integrated energy system considering electric vehicle access, characterized in that: include: Obtaining an electric vehicle (EV) charging load model, connecting the EV charging load model to an energy system IES model, and obtaining a comprehensive energy system; Taking the lowest cost of the integrated energy system as the goal, constructing an upper capacity configuration model; The cost of the comprehensive energy system includes: system equipment investment cost and unit cycle operating cost; With the goal of minimizing the daily operating cost of the integrated energy system, a lower-level optimization operation model is constructed. The daily operating cost of the integrated energy system includes: energy purchase cost, equipment operation and maintenance cost, EV demand response cost and carbon trading cost, and constraints of the lower-level optimization operation model are set; the constraints include: introducing electric energy balance constraints, thermal energy balance constraints, cooling energy balance constraints, gas energy balance constraints, unit ramping constraints, energy storage constraints and EV demand response constraints for user responsiveness under different incentive levels; The Gazelle optimization algorithm is used to substitute the upper-level decision variables of the upper-level capacity configuration model into the lower-level optimization operation model, and the CPLEX solver is called to solve the lower-level optimization operation model to obtain the optimal operation plan and daily operation cost.

2. The low-carbon economic configuration method of the integrated energy system considering electric vehicle access according to claim 1 is characterized in that: Constructing the upper layer capacity configuration model includes: minF=F T +F Y ; Where, F is the total system cost; F T is the investment cost of the equipment in the system; F Y is the system operation cost per unit period; N is the number of equipment types; C i,c is the capacity of the i-th equipment; C i,p is the capacity unit price of device i; r is the discount rate; Y is the device life cycle; s represents the number of typical scenarios; num s is the number of days corresponding to the typical scenario; F Y,s is the daily operating cost in the scenario.

3. The low-carbon economic configuration method of the integrated energy system considering electric vehicle access according to claim 1 is characterized in that: Constructing the lower layer optimization operation model includes: min F Y,s =C P +C M +C C +C DR ; C DR =s c ·C ev,c +s d ·C ev,d ; Among them, F Y,s is the typical daily operating cost; C P C is the energy purchase cost; M is the equipment operation and maintenance cost; C DR is the EV demand response cost; C C is the carbon trading cost; p t is the electricity price at time t; q t is the gas price at time t; is the gas purchase volume at time t; m i is the unit operation and maintenance cost of equipment i; P i,t is the power of the i-th type of equipment at time t; C ev,c is the charge response capacity; C ev,d is the discharge response capacity.

4. The low-carbon economic configuration method of the integrated energy system considering electric vehicle access according to claim 1 is characterized in that: The constraint conditions for setting the lower-level optimization operation model include: Setting the power balance constraint includes: Introducing user responsiveness under different incentive levels, adjusting the charging and discharging power of the EV charging load model, and establishing the power balance constraint: in, are the charging and discharging power of EES at time t respectively; are the output power of PV and WT at time t respectively; is the charging and discharging power of a single EV at time t; is the electrical load at time t; is the power consumption of EB at time t; is the power consumption of EC at time t; is the electrical output power of CHP at time t; ω(i) t is the response status of EV at the current moment; Setting the thermal energy balance constraint includes: in, are the charging and discharging powers of HES respectively; P EB,t is the heat generation power of EB at time t; is the heat load at time t; is the AC heat consumption power at time t, P GB,t is the thermal output power of GB at time t, is the thermal output power of CHP at time t; Setting the cooling energy balance constraint includes: Among them, P AC,t , P EC,t are the cooling power of AC and EC at time t respectively; is the CES charging and cooling power at time t; is the cooling load at time t.

5. The low-carbon economic configuration method of the integrated energy system considering electric vehicle access according to claim 4 is characterized in that: Adjusting the charging and discharging power of the EV charging load model includes: The user responsiveness is used to describe the fuzzy mapping relationship between the user's response intention, response state and responsiveness, and the charging and discharging power of the EV charging load model is adjusted according to the fuzzy mapping relationship; Describing the fuzzy mapping relationship includes: Where ω(i) is the response state of user i, w(i) = 0 represents no response, ω(i) = 1 represents response; σ(i) is the user responsiveness; σ T is the user responsiveness threshold; Acquiring the user responsiveness includes: Among them, k r is the slope of the response changing with the stimulus level; x jl,max is the excitation level under response saturation; Δσ(i) is the maximum deviation of the response.

6. The low-carbon economic configuration method of the integrated energy system considering electric vehicle access according to claim 1 is characterized in that: The constraint conditions for setting the lower layer optimization operation model also include: Setting the gas energy balance constraint includes: in, is the gas purchase volume at time t; is the gas load at time t; is the gas consumption power of CHP at time t; is the gas consumption power of GB at time t; Setting the unit climbing constraint includes: in, is the output power of device i at time t; is the uphill and downhill power limit of device i; Setting the energy storage constraint includes: Where: is the energy storage power of energy storage device i at time t; is the charging and discharging power of energy storage device i at time t; η i,ch , η i,dis is the charging and discharging efficiency of energy storage device i at time t; ε i is the self-discharge loss coefficient of energy storage device i; are the upper and lower limits of the capacity of energy storage device i; Setting the EV demand response constraint includes: C ev,min ≤C ev,t ≤C ev,max Among them, C ev,t is the battery capacity of EV at time t; is the charging and discharging efficiency of the EV; is the charging and discharging power of EV at time t; T ch 、T dis C is the charging and discharging time of EV; ev,max , C ev,min The upper and lower limits of EV battery capacity; is the maximum charging and discharging power of EV; Δt is the interval between charging and discharging; C ev,d is the discharge response capacity; C ev,c The charging response capacity.

7. The low-carbon economic configuration method of the integrated energy system considering electric vehicle access according to claim 1 is characterized in that: Obtaining the optimal operation plan and daily operation cost includes: Step 1, initialize the gazelle population; Step 2: The predicted load, wind and solar output, and initial configuration values ​​of equipment capacity are used as upper-level decision variables, and the upper-level decision variables of the upper-level capacity configuration model are substituted into the lower-level optimization operation model, and the lower-level optimization operation model is constrained. According to the equipment capacity parameters, the CPLEX solver is called to solve the lower-level optimization operation model to obtain the optimal operation plan and individual daily operation cost of the population; Step 3, judging whether the individual is less than the maximum number of iterations, if the individual is less than the maximum number of iterations, re-calling the CPLEX solver to solve the lower optimization operation model, if the individual is greater than the maximum number of iterations, calculating the objective function value and constructing an elite matrix; Step 4, determine whether the random number r is less than the target value. If the random number r is less than the target value, perform development operations in combination with the elite matrix. If the random number r is greater than the target value, update the gazelle position, top gazelle and elite matrix. Step 5: Use the Predator Success Rate (PSR) s Update the gazelle position again and determine whether the updated gazelle position meets the termination condition. If so, output the optimal solution. If not, repeat step 2 until the termination condition is met.

8. The low-carbon economic configuration method of the integrated energy system considering electric vehicle access according to claim 7 is characterized in that: Development operations in conjunction with the Elite Matrix include: g i+1 =g i +s·R·R B ·(E i -R B ·g i ) Among them, g i+1 is the next iteration solution; g i is the current iterative solution; s is the gazelle's predation speed; R B is the random number vector of Brownian motion; R is the uniform random number vector in [0,1]; E i For the elite matrix.

9. The low-carbon economic configuration method of the integrated energy system considering electric vehicle access according to claim 7 is characterized in that: Updating the Gazelle position involves: Determine whether the number of iterations is an odd number. If the number of iterations is an odd number, use the first exploration formula to update the gazelle position: g i+1 =g i +S·μ·R·R L ·(E i -R L ·g i ); Among them, g i+1 is the next iteration solution; g i is the current iterative solution; μ is a constant; S is the maximum speed that the gazelle can reach; If the number of iterations is an even number, the second exploration formula is used to update the gazelle position: g i+1 =g i +S·μ·CF·R·R B ·(E i -R L ·g i ): Where S is the maximum speed that the gazelle can reach; R is a uniform random number vector in [0,1]; R B is the random number vector of Brownian motion; R L is a random number vector based on Levy distribution.

10. The low-carbon economic configuration method of the integrated energy system considering electric vehicle access according to claim 1, characterized in that: Using the Predator Success Rate (PSR) s Updated Gazelle positions again include: Among them, g i+1 is the next iteration solution; g i is the current iterative solution; UB, LB are the upper and lower bounds of the population; U is a binary vector; r is a random number in [0,1]; g r1 ,g r2 is the random index of the gazelle matrix; R is a uniform random number vector in [0,1]; LB is the lower bound of the population.

Images