A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost

Abstract

The present invention discloses a hybrid scheduling method based on maximum consumption of new energy and optimal power generation cost. The present invention first constructs the objective function of maximum consumption of new energy and minimum operation cost of integrated energy system, energy balance constraint condition, system energy supply equipment constraint condition, energy storage device constraint condition and rotating reserve constraint condition; then the above constraint condition is used as the constraint condition of non-dominated sorting genetic algorithm and multi-objective particle swarm algorithm, and the optimal solution of the above multi-objective function is obtained by integrating the genetic algorithm of non-dominated sorting and the multi-objective particle swarm algorithm; finally, in the iterative process, the population is sorted according to the crowding distance, and the whole population is divided into two parts according to the sorting result, wherein the best half of the population is optimized by the non-dominated sorting genetic algorithm, and the other half of the population is optimized by the multi-objective particle swarm algorithm, so that they converge around the optimal solution. The method provided by the present application can effectively promote the consumption of new energy and reduce the system operation cost.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid planning, and in particular to a hybrid scheduling method based on maximum consumption of new energy and optimal power generation cost. Background Art

[0002] With the increasing prominence of environmental problems and the continuous improvement of new energy power generation technology, large-scale development and utilization of new energy power generation is a new direction for power system dispatching. At present, my country's new energy is connected to the grid on a large scale, and the installed capacity of wind power and photovoltaic power are among the best in the world. Due to the large-scale grid connection of clean energy, it cannot be fully utilized, resulting in a large number of wind and solar power abandonment. Due to the uncertainty of wind and solar power generation, the realization of large-scale grid connection of wind and solar power has put forward higher requirements for the power system. In the development direction and positioning of renewable energy, it is mentioned that wind power is a "booster" for the transformation and development of new energy; solar power generation is an "accelerator" for the transformation and development of new energy; biomass energy, geothermal energy, and ocean energy are "new kinetic energy" for the transformation and development of new energy; in 2035, the installed capacity of wind power and photovoltaic power will exceed 50% of the total installed capacity, and between 2045 and 2050, the wind power and photovoltaic power generation will exceed 50% of the total power generation, and the carbon dioxide emissions of the energy system and the power system can reach their peak before 2030. In summary, new energy plays an increasingly important role. In this process, it is necessary to consider maximizing the absorption capacity of new energy and reducing the system cost as much as possible to achieve the best economic benefits. This is a complex multi-objective optimization problem. The present invention introduces a hybrid non-dominated sorting genetic algorithm NSGA-II and a multi-objective particle swarm algorithm MOPSO to find the optimal solution to the multi-objective model, thereby obtaining the optimal solution. Summary of the invention

[0003] The purpose of the present invention is to provide a hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost. The proposed method scientifically plans and accurately calculates the new energy consumption capacity, which is conducive to maximizing the new energy consumption capacity and minimizing the power generation cost. The technical solution is as follows:

[0004] A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost includes the following steps:

[0005] Step 1: Establish an optimization model based on the new energy consumption capacity and power generation cost, construct the maximum objective function of new energy consumption and the minimum objective function of the integrated energy system operation cost, energy balance constraints, system energy supply equipment constraints, energy storage device constraints and spinning reserve constraints;

[0006] Step 2, set the population size pop, the number of iterations gen, the number of objective functions M, the number of decision variables V, the inertia weight w, the individual learning coefficient c1, the global learning coefficient c2, and the velocity vector fori;

[0007] Step 3, using a non-dominated sorting genetic algorithm, iteratively solve the optimal value of the new energy consumption capacity and power generation cost of the best half of the population;

[0008] Step 4, through the multi-objective particle swarm algorithm, iteratively solve the optimal value of the new energy consumption capacity and power generation cost of the other half of the population;

[0009] Step 5, iteratively solving the optimal value of the new energy consumption capacity and power generation cost optimization model by hybrid non-dominated sorting genetic algorithm and multi-objective particle swarm algorithm;

[0010] The maximum objective function of new energy consumption described in step 1 is expressed as follows:

[0011]

[0012] Among them, the objective function f1(x) represents the maximum consumption of new energy, including the maximum combined output of wind and solar power; and They represent the output power of wind power and photovoltaic power at time t respectively, and T represents the number of hours in the system scheduling cycle.

[0013] The objective function for minimizing the operating cost of the integrated energy system described in step 1 is expressed as follows:

[0014]

[0015] Among them, the objective function f2(x) represents the minimum operating cost of the integrated energy system, including the fuel cost of the cogeneration unit, the P2G operating cost, the operating cost of the micro gas turbine, the wind abandonment penalty cost, the solar abandonment penalty cost, the environmental cost and the carbon trading cost;

[0016] The total fuel cost of the combined heat and power unit is expressed as follows:

[0017]

[0018] Where C1 is the total fuel cost of the cogeneration unit; C pi , C ci and C hi Respectively represent the i th A pure power unit with valve point effect, the i th CHP units and th The fuel cost of a pure heat unit; P i is the i th The power generation of the combined heat and power unit, Oi is the i th The power generated by a combined heat and power unit, H i is the i th The heat generated by a combined heat and power unit, T i is the i th The heat generated by the heating unit only; the coefficient a i ,b i ,c i Indicates the i th The fuel cost factor for each generator, factor d i and e i is the coefficient used to incorporate the valve point effect of generator i, α i ,β i ,γ i ,δ i ,ε i ,ζ i is the i th The coefficient of each CHP unit, η i ,θ i ,λ i is the i th Cost factor for each thermoelectric unit;

[0019] The operating cost of P2G equipment is expressed as follows:

[0020]

[0021] Among them, C2 is the operating cost of the P2G equipment, c1 is the operating cost coefficient, and d1 is the CO2 cost coefficient required by the equipment; the operating cost of the micro gas turbine is expressed as follows:

[0022]

[0023] Among them, C3 is the operating cost of the micro gas turbine, and a3 is the operating cost coefficient;

[0024] The penalty cost caused by wind curtailment is expressed as follows:

[0025]

[0026] Among them, C4 is the penalty cost for wind abandonment, and a4 is the penalty cost coefficient;

[0027] The penalty cost caused by the abandonment of light is expressed as follows:

[0028]

[0029] Among them, C5 is the penalty cost of abandoned light, and a5 is the penalty cost coefficient;

[0030] Environmental cost, expressed as follows:

[0031]

[0032] Among them, C6 is the environmental cost, a6 is the environmental cost coefficient;

[0033] Carbon trading costs are expressed as follows:

[0034] C7=λ(E p -E L )

[0035] Among them, λ is the carbon trading price in the market; E p is the actual carbon emissions; E L It is a free carbon emission allowance.

[0036] The energy balance constraints described in step 1 include electric energy balance constraints, thermal energy balance constraints, and cooling energy balance constraints; the electric energy balance constraint is expressed as follows:

[0037] P PV +P WT +P MT +P ES +P GP =P EB +P HP +P EC +P L (9)

[0038] Among them, P PV is the power output of the photovoltaic power generation system, P WT is the electric power output of the wind power generation system, P MT is the electric power output of the micro-turbine, P ES is the storage power of the energy storage system, P GP is the power supplied to the system from the large power grid, P EB is the power consumption of the electric boiler, P HP is the power consumption of the electric heat pump, P EC is the power consumption of the electric refrigerator, P L is the system electrical load;

[0039] Thermal energy balance constraint, the expression is as follows:

[0040] Q MT,h +Q GB +Q HP,h +Q EB +Q TES =P Lh (10)

[0041] Among them, Q MT,h is the thermal power output of the micro-turbine, Q GBis the thermal power output of the gas boiler, Q HP,h is the thermal power output of the heat pump, Q EB is the thermal power output of the electric boiler, Q TES is the thermal power stored in the thermal energy storage system, P Lh is the heat load of the system;

[0042] Cold energy balance constraint, the expression is as follows:

[0043] Q MT,c +Q EC +Q HP,c +Q CES =P Lc (11)

[0044] Among them, Q MT,c is the cooling power output of the micro-turbine, Q EC is the cooling power output by the electric refrigerator, Q HP,c is the cooling power output of the heat pump, Q CES is the cold power stored in the cold energy storage system, P Lc is the cooling load of the system.

[0045] The system energy supply equipment constraints described in step 1 include gas turbine output constraints, photovoltaic power generation output constraints, wind power generation output constraints, heat pump output constraints, electric boiler output constraints and electric refrigerator output constraints;

[0046] The gas turbine output constraint is expressed as follows:

[0047]

[0048] in, are the minimum power limit and maximum power of the unit output, P MT (t) is the power output of the gas turbine at time t;

[0049] Photovoltaic power generation output constraint, the expression is as follows:

[0050] P PV (t) = P PV,f (t) (13)

[0051] Among them, P PV (t) is the power of photovoltaic power generation, P PV,f (t) is the predicted power of photovoltaic power generation;

[0052] Wind power output constraint, the expression is as follows:

[0053] P WT (t) = P WT,f (t) (14)

[0054] Among them, P WT (t) is the power of photovoltaic power generation, P WT,f (t) is the predicted power of photovoltaic power generation;

[0055] The heat pump output constraint is expressed as follows:

[0056]

[0057]

[0058] Among them, P HP,h (t), P HP,c (t) are the output of the heat pump in heating and cooling states, They are the upper limits of the unit output of the electric heat pump in heating and cooling states respectively;

[0059] The output constraint of electric boiler is as follows:

[0060]

[0061] Among them, P EB (t) cooling power provided by the electric refrigerator, The upper limit of the cooling power of the electric chiller.

[0062] The energy storage device constraints described in step 1 include energy state range and charge and discharge power constraints;

[0063] Energy storage device constraints, the expression is as follows:

[0064]

[0065]

[0066]

[0067]

[0068] in, are the lower limit of the energy state of the battery and the upper limit of the energy state of the battery, Q EES (t) is the charge of the energy storage, are the upper limits of battery charging and discharging power, β EES,c (t), β EES,d (t) are the state variables of the energy storage charging and discharging.

[0069] The spinning reserve constraint conditions in step 1 include positive spinning reserve constraints and negative spinning reserve constraints;

[0070] Positive spinning reserve constraint, the expression is as follows:

[0071]

[0072]

[0073] Among them, P us i is the positive spinning reserve capacity provided by thermal power unit i, l U is the demand for positive spinning reserve due to load forecast error, S U The additional positive spinning reserve capacity required for the intervention of wind power, P D is the ultra-short-term load forecast value, P imax is the output upper limit of thermal power unit i, P i0 is the planned output of unit i on the day before, ΔP i is the real-time adjustment value of unit i, is the maximum upward regulation rate of unit i, and Δt is the regulation period;

[0074] Negative spinning reserve constraint, expressed as follows:

[0075]

[0076]

[0077] Among them, P dsi The negative spinning reserve capacity provided for thermal power unit i, l D is the demand for negative spinning reserve due to load forecast error, S D The additional negative spinning reserve capacity required for the intervention of wind power, P D is the ultra-short-term load forecast value, P i0 is the planned output of unit i on the day before, ΔP i is the real-time adjustment value of unit i, P im in is the lower limit of the output of thermal power unit i, is the maximum downward regulation rate of unit i, and Δt is the regulation period.

[0078] In step 3, the optimal values ​​of the new energy consumption capacity and power generation cost of the best half of the population are iteratively solved by the non-dominated sorting genetic algorithm, including the following steps:

[0079] Step 3.1, establish a population and determine the initial population number, set the number of population iterations, crossover, mutation percentage, mutation rate and mutation step; initialize the optimal target value and the corresponding objective function related parameters, the optimal target value is the value of the objective function of maximum new energy consumption capacity and minimum power generation cost;

[0080] Step 3.2, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x);

[0081] Step 3.3, determine whether the value of the individual population meets the constraint conditions of the new energy consumption index;

[0082] Step 3.4, when the value of the individual population does not meet the constraint condition of the new energy consumption index, the new energy consumption plan is adjusted based on the constraint processing mechanism, so that the value of the individual population after the adjustment meets the restriction of the constraint processing mechanism;

[0083] Step 3.5, when the values ​​of the population individuals meet the constraints of the new energy consumption index, the genetic algorithm based on non-dominated sorting performs non-dominated sorting on the population individuals, sorts each population according to the non-dominated frontier, calculates the crowding distance of each non-dominated level, and sorts the population according to the crowding distance.

[0084] Step 3.6, using the population to create a descendant population, after selection, crossover and mutation, applying the constraint processing mechanism, evaluating the new energy absorption capacity objective function f1(x) and the comprehensive energy system operation cost f2(x);

[0085] Step 3.7, merge the parent and offspring populations, perform non-dominated sorting, sort each population according to the non-dominated frontier, calculate the crowding distance of each non-dominated level, and sort the population according to the crowding distance.

[0086] Step 3.8, select a solution, record the crowding distance in ascending order, store the non-dominated solutions in a list, plot the non-dominated solutions in the list, increment the generation count, and the loop ends.

[0087] In step 4, the multi-objective particle swarm algorithm is used to iteratively solve the optimal value of the new energy consumption capacity and power generation cost of the other half of the population, including the following steps:

[0088] Step 4.1, generating a random particle swarm, initializing the optimal target value and the corresponding objective function related parameters, the optimal target value is the value of the objective function of maximizing the new energy consumption capacity and minimizing the power generation cost;

[0089] Step 4.2, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x);

[0090] Step 4.3, determine whether the particle group meets the constraints of the new energy consumption index;

[0091] Step 4.4, when the particle swarm does not meet the constraint conditions of the new energy consumption index, the new energy consumption plan is adjusted based on the constraint processing mechanism, so that the adjusted particle swarm meets the restrictions of the constraint processing mechanism;

[0092] Step 4.5, when the particle swarm meets the constraint conditions of the new energy consumption index, the fitness is calculated based on the multi-objective particle swarm algorithm to screen the non-inferior solution set. When a particle is not dominated by other particles, the particle is put into the non-inferior solution set, and a particle is randomly selected from the non-inferior solution set as the optimal particle of the swarm before the particle is updated;

[0093] Step 4.6, update the non-inferior solution set. When the new particle is not dominated by other particles and particles in the current non-inferior solution set, put the new particle into the non-inferior solution set, and randomly select a particle from the non-inferior solution set as the optimal particle of the group before each particle update;

[0094] Step 4.7, update the velocity and position of the particle, apply the constraint processing mechanism, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x), apply mutation and calculate the new solution;

[0095] Step 4.8, add the non-dominant particles to the repository, determine the dominance of the new repository members, keep only the non-dominated members in the repository, update the grid and grid index, delete members if the repository is full, draw the members in the external repository, modify the inertia, and the loop ends.

[0096] The genetic algorithm based on hybrid non-dominated sorting and multi-objective particle swarm algorithm described in step 5 seeks the optimal solution for maximizing the new energy consumption capacity and minimizing the power generation cost, including the following steps:

[0097] Step 5.1, setting the initial parameters of the non-dominated sorting genetic algorithm and the multi-objective particle swarm algorithm, initializing the optimal target value and the corresponding objective function related parameters, the optimal target value is the value of the objective function of maximizing the new energy consumption capacity and minimizing the power generation cost;

[0098] Step 5.2, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x);

[0099] Step 5.3, determine whether the value of the individual population meets the constraint conditions of the new energy consumption index;

[0100] Step 5.4, when the value of the individual population does not meet the constraint condition of the new energy consumption index, the new energy consumption plan is adjusted based on the constraint processing mechanism, so that the adjusted particle group meets the restriction of the constraint processing mechanism;

[0101] Step 5.5, when the values ​​of the individuals in the population meet the constraints of the new energy consumption index, perform non-dominated sorting, sort each population according to the non-dominated frontier, calculate the crowding distance of each non-dominated level, and sort the population according to the crowding distance. According to the sorting results, the population is divided into two parts;

[0102] Step 5.6, use the upper half of the population to create a descendant population, and after selection, crossover and mutation, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x) through the constraint processing mechanism; merge the parent and descendant populations, perform non-dominated sorting, sort each population according to the non-dominated frontier, calculate the crowding distance of each non-dominated level, and sort the population according to the crowding distance, select a solution, record the crowding distance in ascending order, store the non-dominated solution in a list, and plot the non-dominated solution in the list;

[0103] Step 5.7, the positions and costs of particles are initialized from the lower half of the population, the positions of particles are stored as their individual optimal positions, the dominance of particles is determined, the external repository representative is initialized, a grid is created and the grid index is found, the leader particle is selected from the external repository, the particle position and velocity are updated, the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x) are evaluated through the constraint processing mechanism, mutation is applied and the new solution is calculated;

[0104] In step 5.8, the non-dominated particles are added to the repository, the dominance of the new repository members is determined, only the non-dominated members are retained in the repository, the grid and grid index are updated and the inertia is modified to create a new set of particles with a size of half the total population, which is then filled with the non-dominated solutions in the repository, and finally the population of the genetic algorithm of the non-dominated sorting and the new set of particles of the multi-objective particle swarm algorithm are combined, the count is incrementally generated, and the loop ends.

[0105] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention introduces a hybrid non-dominated sorting genetic algorithm and a multi-objective particle swarm algorithm into the optimization problem of the new energy absorption capacity and the power generation cost, and uses the energy balance constraint, the system energy supply equipment constraint, the energy storage device constraint and the rotating reserve constraint as the constraint of the hybrid non-dominated sorting genetic algorithm and the multi-objective particle swarm algorithm, and performs scientific calculation and global planning on the system, which is conducive to more accurately calculating the maximum absorption capacity of the system for new energy and the minimum power generation cost. The hybrid NSGA-II-MOPSO algorithm solves a large number of model solving problems under equality constraints and inequality constraints. By setting and maintaining the global Pareto set, the number of Pareto optimal solutions searched per unit time is significantly increased, and the distribution range and uniformity of the solutions are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Figure 1 A flow chart of the calculation method for the maximum absorption capacity of new energy and the minimum power generation cost;

[0107] Figure 2It is a flowchart of the specific implementation method of the genetic algorithm based on non-dominated sorting;

[0108] Figure 3 It is a flowchart of the specific implementation method based on the hybrid NSGA-II-MOPSO algorithm;

[0109] Figure 4 It is a graph of non-dominated solutions stored in the external repository and list F1 at the end of each iteration based on the hybrid NSGA-II-MOPSO algorithm;

[0110] Figure 5 The overall flow chart of solving the multi-objective optimization model based on the hybrid NSGA-II-MOPSO algorithm. DETAILED DESCRIPTION

[0111] In order to make the technical solution and advantages of the present invention clearer, the implementation scheme will be described in further detail below. The present invention aims to minimize the power generation cost of the integrated energy system and maximize the promotion of new energy consumption, and proposes a multi-objective optimization method for the integrated system based on a hybrid non-dominated sorting genetic algorithm and a multi-objective particle swarm algorithm.

[0112] Example 1

[0113] like Figure 1 As shown, a hybrid scheduling method based on maximum consumption of new energy and optimal power generation cost includes the following steps:

[0114] Step 1: Establish an optimization model based on the new energy consumption capacity and power generation cost, construct the maximum objective function of new energy consumption and the minimum objective function of the integrated energy system operation cost, energy balance constraints, system energy supply equipment constraints, energy storage device constraints and spinning reserve constraints;

[0115] Step 2, set the population size pop, the number of iterations gen, the number of objective functions M, the number of decision variables V, the inertia weight w, the individual learning coefficient c1, the global learning coefficient c2, and the velocity vector fori;

[0116] Step 3, using a non-dominated sorting genetic algorithm, iteratively solve the optimal value of the new energy consumption capacity and power generation cost of the best half of the population;

[0117] Step 4, through the multi-objective particle swarm algorithm, iteratively solve the optimal value of the new energy consumption capacity and power generation cost of the other half of the population;

[0118] Step 5, iteratively solving the optimal value of the new energy consumption capacity and power generation cost optimization model by hybrid non-dominated sorting genetic algorithm and multi-objective particle swarm algorithm;

[0119] The objective function can be expressed as:

[0120]

[0121] Among them, f1(x) represents the maximum consumption of new energy, which is modified to a minimization function by adding a minus sign; f2(x) represents the minimum operating cost of the integrated energy system; and They represent the output power of wind power and photovoltaic power at time t respectively; T represents the number of hours in the system scheduling cycle; C1 is the total fuel cost of the cogeneration unit; C2 is the operating cost of the P2G equipment; C3 is the operating cost of the micro gas turbine; C4 is the penalty cost for wind abandonment; C5 is the penalty cost for photovoltaic abandonment; and C6 is the environmental cost.

[0122] Example 2

[0123] like Figure 2 As shown, the operation steps of the non-dominated sorting genetic algorithm NSGA-II are as follows:

[0124] Step a, firstly, perform random simulation operations to generate an initial population P with a size value of N, followed by the generation of a child population Q, and then combine the population P with the child population Q to finally generate a population M with a size value of 2N;

[0125] Step b, combine the parent and child populations, perform non-dominated sorting on them, and calculate the crowding of all individuals in the non-dominated layer, and then select appropriate individuals based on the non-dominated relationship and individual crowding to form a new parent population P1;

[0126] Step c, generate a new offspring population Q through a genetic algorithm, and combine the new parent population P1 with the new offspring population Q1 to form a new population M1. Repeat the above steps until the conditions are met and the cycle ends.

[0127] Example 3

[0128] The operation steps of the multi-objective particle swarm optimization algorithm MOPSO are as follows:

[0129] Step a, for each particle in the particle swarm, initialize its velocity and position;

[0130] Step b, calculating the target vector value of each particle and adding the non-dominated solution therein to the external memory;

[0131] Step c, determine the initial global best position gbest and individual best position pbest of the particle;

[0132] Step d, update the speed and position of the particle, and take certain measures to ensure that the particle flies within the search space;

[0133] Step e, calculate the target vector of each particle and adjust the individual best position pbest of the particle;

[0134] Step f, update the external memory and select the global best position gbest for each particle;

[0135] Step g, determine whether the algorithm termination condition is met, if so, the algorithm stops running; otherwise, return to step c.

[0136] Example 4

[0137] like Figure 3 As shown, the operation steps of the hybrid NSGA-II-MOPSO algorithm are as follows:

[0138] Step a, input the forecast data of wind power, photovoltaic and electric heating loads, initialize the parameters, and generate the initial population that meets the requirements of this paper;

[0139] Step b, update the initial particle velocity V i and position X i , the expression is as follows:

[0140]

[0141] Step c, calculate the fitness function value of the particles and sort their population, and divide the sorted population into M and N sub-populations evenly;

[0142] Step d, copy the subpopulation M with better fitness value to the next generation, and perform crossover and mutation with subpopulation N, α, β∈[0,1], the expression is as follows:

[0143]

[0144] Step e, using the current fitness of the particle and the individual extreme fitness obtained in step c, and the fitness size of the individual extreme value and the global extreme value, to update the individual extreme value and the global extreme value;

[0145] Step f, determine whether the maximum number of iterations or the global optimum is reached. If the requirements are met, stop the iteration; otherwise, continue with steps b to e.

[0146] Example 5

[0147] The present invention solves the problem through Matlab simulation. After determining the objective function and constraint conditions, the parameters are set in combination with the system. The parameters of the non-dominated sorting genetic algorithm are: the maximum number of iterations is 120 times, the population size N=200, and the crossover probability P c =0.7, mutation probability P m=0.4; the parameters based on the NSGA-II-MOPSO algorithm are: inertia factor ω = 0.5, learning factor c1 = 4, c2 = 1, the number of grids in each dimension n G =7, leadership selection pressure beta=2.

[0148] Specifically, the simulation results are as follows: Figure 4 The figure shows the non-dominated solutions stored in the external repository and in the list F1 at the end of each iteration of the hybrid NSGA-II-MOPSO algorithm. The size of list F1 is 100 and the size of the repository is 50. In this test function, the search space is so large that almost 90 iterations are required to fill the list. The size of the external repository is set to 50, from Figure 4 As can be seen in the graphs shown, the external repository used by MOPSO is never full and the maximum number of non-dominated particles found from the lower half of the population at iterations is only at iterations 13, 52, and 118. At iteration 52, one non-dominated solution found by NSGA-II from the upper half of the population dominates the 13 non-dominated solutions found by MOPSO from the lower half of the population, so list F1 has only one solution dominated by a non-dominated solution. The maximum number of non-dominated particles stored in list F1 fluctuates until it reaches 100. Once it exceeds 100, these non-dominated particles are sorted and only 100 non-dominated solutions are retained.

[0149] Compared with NSGA-II, MOPSO and TVAC-PSO, the hybrid NSGA-II-MOPSO algorithm proposed in the present invention has performance improvements of 0.168%, 0.237% and 0.746% in saving power generation costs, respectively. Compared with NSGA-II, MOPSO and TVAC-PSO, the hybrid method has performance improvements of 0.149%, 0.062% and 2.38% in reducing emission levels, respectively. Compared with NSGA-II, MOPSO and TVAC-PSO, the power generation cost obtained by the proposed method is reduced by 17 US dollars / hour, 24 US dollars / hour and 76 US dollars / hour, respectively. Compared with NSGA-II, MOPSO and TVAC-PSO, the emission levels obtained by the proposed method are 0.072t / h, 0.03t / h and 1.176t / h lower, respectively. Since wind power and photovoltaic power generation do not emit gas, they greatly promote the consumption of new energy. The hybrid algorithm can produce well-spread and diverse Pareto solutions, resulting in the minimum generation cost and the maximum consumption of new energy. The reason why the hybrid algorithm performs better is mainly due to the local search and effective constraint processing mechanism performed by MOPSO in the hybrid algorithm.

[0150] This paper proposes a hybrid scheduling method based on the maximum consumption of new energy and the optimal power generation cost. The hybrid NSGA-II-MOPSO algorithm solves the model solving problems under a large number of equality constraints and inequality constraints. By setting and maintaining the global Pareto set, the number of Pareto optimal solutions searched per unit time is significantly increased, and the distribution range and uniformity of the solutions are improved. The effectiveness of the proposed method is verified by examples.

[0151] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0152] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0153] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost, characterized in that: The steps include: Step 1: Establish an optimization model based on the new energy consumption capacity and power generation cost, construct the maximum objective function of new energy consumption and the minimum objective function of the integrated energy system operation cost, energy balance constraints, system energy supply equipment constraints, energy storage device constraints and spinning reserve constraints; Step 2, set the population size pop, the number of iterations gen, the number of objective functions M, the number of decision variables V, the inertia weight w, the individual learning coefficient c1, the global learning coefficient c2, and the velocity vector fori; Step 3, using a non-dominated sorting genetic algorithm, iteratively solve the optimal value of the new energy consumption capacity and power generation cost of the best half of the population; Step 4, through the multi-objective particle swarm algorithm, iteratively solve the optimal value of the new energy consumption capacity and power generation cost of the other half of the population; Step 5, iteratively solving the optimal value of the new energy consumption capacity and power generation cost optimization model by hybrid non-dominated sorting genetic algorithm and multi-objective particle swarm algorithm; The optimal solution for maximizing the new energy consumption capacity and minimizing the power generation cost by using the hybrid non-dominated sorting genetic algorithm and multi-objective particle swarm algorithm described in step 5 includes: Step 5.1, setting the initial parameters of the non-dominated sorting genetic algorithm and the multi-objective particle swarm algorithm, initializing the optimal target value and the corresponding objective function related parameters, the optimal target value is the value of the objective function of maximizing the new energy consumption capacity and minimizing the power generation cost; Step 5.2, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x); Step 5.3, determine whether the value of the individual population meets the constraint conditions of the new energy consumption index; Step 5.4, when the value of the individual population does not meet the constraint condition of the new energy consumption index, the new energy consumption plan is adjusted based on the constraint processing mechanism, so that the adjusted particle group meets the restriction of the constraint processing mechanism; Step 5.5, when the values ​​of the individuals in the population meet the constraints of the new energy consumption index, perform non-dominated sorting, sort each population according to the non-dominated frontier, calculate the crowding distance of each non-dominated level, and sort the population according to the crowding distance. According to the sorting results, the population is divided into two parts; Step 5.6, use the upper half of the population to create a descendant population, and after selection, crossover and mutation, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x) through the constraint processing mechanism; merge the parent and descendant populations, perform non-dominated sorting, sort each population according to the non-dominated frontier, calculate the crowding distance of each non-dominated level, and sort the population according to the crowding distance, select a solution, record the crowding distance in ascending order, store the non-dominated solution in a list, and plot the non-dominated solution in the list; Step 5.7, the positions and costs of particles are initialized from the lower half of the population, the positions of particles are stored as their individual optimal positions, the dominance of particles is determined, the external repository representative is initialized, a grid is created and the grid index is found, the leader particle is selected from the external repository, the particle position and velocity are updated, the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x) are evaluated through the constraint processing mechanism, mutation is applied and the new solution is calculated; In step 5.8, the non-dominated particles are added to the repository, the dominance of the new repository members is determined, only the non-dominated members are retained in the repository, the grid and grid index are updated and the inertia is modified to create a new set of particles with a size of half the total population, which is then filled with the non-dominated solutions in the repository, and finally the population of the genetic algorithm of the non-dominated sorting and the new set of particles of the multi-objective particle swarm algorithm are combined, the count is incrementally generated, and the loop ends.

2. A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost according to claim 1, characterized in that: The maximum objective function of new energy consumption described in step 1 is expressed as follows: Among them, the objective function f1(x) represents the maximum consumption of new energy, including the maximum combined output of wind and solar power; and They represent the output power of wind power and photovoltaic power at time t respectively, and T represents the number of hours in the system scheduling cycle.

3. A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost according to claim 1, characterized in that: The objective function for minimizing the operating cost of the integrated energy system described in step 1 is expressed as follows: Among them, the objective function f2(x) represents the minimum operating cost of the integrated energy system, including the fuel cost of the cogeneration unit, the P2G operating cost, the operating cost of the micro gas turbine, the wind abandonment penalty cost, the solar abandonment penalty cost, the environmental cost and the carbon trading cost; The total fuel cost of the combined heat and power unit is expressed as follows: Where C1 is the total fuel cost of the cogeneration unit; C pi , C ci and C hi Respectively represent the i th A pure power unit with valve point effect, the i th CHP units and th The fuel cost of a pure heat unit; P i is the i th The power generation of the combined heat and power unit, O i is the i th The power generated by a combined heat and power unit, H i is the i th The heat generated by a combined heat and power unit, T i is the i th The heat generated by the heating unit only; the coefficient a i ,b i ,c i Indicates the i th The fuel cost factor for each generator, factor d i and e i is the coefficient used to incorporate the valve point effect of generator i, α i ,β i ,γ i ,δ i ,ε i ,ζ i is the i th The coefficient of each CHP unit, η i ,θ i ,λ i is the i th Cost factor for each thermal power unit; The operating cost of P2G equipment is expressed as follows: Among them, C2 is the operating cost of the P2G equipment, c1 is the operating cost coefficient, and d1 is the CO2 cost coefficient required by the equipment; The operating cost of a micro gas turbine is expressed as follows: Among them, C3 is the operating cost of the micro gas turbine, and a3 is the operating cost coefficient; The penalty cost caused by wind curtailment is expressed as follows: Among them, C4 is the penalty cost for wind abandonment, and a4 is the penalty cost coefficient; The penalty cost caused by the phenomenon of abandoned light is expressed as follows: Among them, C5 is the penalty cost of abandoned light, and a5 is the penalty cost coefficient; Environmental cost, expressed as follows: Among them, C6 is the environmental cost, a6 is the environmental cost coefficient; Carbon trading costs are expressed as follows: C7=λ(E p -E L ) Among them, λ is the carbon trading price in the market; E p is the actual carbon emissions; E L It is a free carbon emission allowance.

4. A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost according to claim 1, characterized in that: The energy balance constraints described in step 1 include electric energy balance constraints, thermal energy balance constraints, and cooling energy balance constraints; The power balance constraint is expressed as follows: P PV +P WT +P MT +P ES +P GP =P EB +P HP +P EC +P L (9) Among them, P PV is the power output of the photovoltaic power generation system, P WT is the electric power output of the wind power generation system, P MT is the electric power output of the micro-turbine, P ES is the storage power of the energy storage system, P GP is the power supplied to the system from the large power grid, P EB is the power consumption of the electric boiler, P HP is the power consumption of the electric heat pump, P EC is the power consumption of the electric refrigerator, P L is the system electrical load; Thermal energy balance constraint, the expression is as follows: Q MT,h +Q GB +Q HP,h +Q EB +Q TES =P Lh (10) Among them, Q MT,h is the thermal power output of the micro-turbine, Q GB is the thermal power output of the gas boiler, Q HP,h is the thermal power output of the heat pump, Q EB is the thermal power output of the electric boiler, Q TES is the thermal power stored in the thermal energy storage system, P Lh is the heat load of the system; Cold energy balance constraint, the expression is as follows: Q MT,c +Q EC +Q HP,c +Q CES =P Lc (11) Among them, Q MT,c is the cooling power output of the micro-turbine, Q EC is the cooling power output by the electric refrigerator, Q HP,c is the cooling power output of the heat pump, Q CES is the cold power stored in the cold energy storage system, P Lc is the cooling load of the system.

5. A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost according to claim 1, characterized in that: The system energy supply equipment constraints described in step 1 include gas turbine output constraints, photovoltaic power generation output constraints, wind power generation output constraints, heat pump output constraints, electric boiler output constraints and electric refrigerator output constraints; The gas turbine output constraint is expressed as follows: in, are the minimum power limit and maximum power of the unit output, P MT (t) is the power output of the gas turbine at time t; Photovoltaic power generation output constraint, the expression is as follows: P PV (t)=P PV,f (t) (13) Among them, P PV (t) is the power of photovoltaic power generation, P PV,f (t) is the predicted power of photovoltaic power generation; Wind power generation output constraint, the expression is as follows: P WT (t)=P WT,f (t) (14) Among them, P WT (t) is the power of photovoltaic power generation, P WT,f (t) is the predicted power of photovoltaic power generation; The heat pump output constraint is expressed as follows: Among them, P HP,h (t), P HP,c (t) are the output of the heat pump in heating and cooling states, They are the upper limits of the unit output of the electric heat pump in heating and cooling states respectively; The output constraint of electric boiler is as follows: Among them, P EB (t) cooling power provided by the electric refrigerator, The upper limit of the cooling power of the electric chiller.

6. A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost according to claim 1, characterized in that: The energy storage device constraints described in step 1 include energy state range and charge and discharge power constraints; Energy storage device constraints, the expression is as follows: in, are the lower limit of the energy state of the battery and the upper limit of the energy state of the battery, Q EES (t) is the charge of the energy storage, are the upper limits of battery charging and discharging power, β EES,c (t), β EES,d (t) are the state variables of the energy storage charging and discharging.

7. A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost according to claim 1, characterized in that: The spinning reserve constraint conditions in step 1 include positive spinning reserve constraints and negative spinning reserve constraints; Positive spinning reserve constraint, the expression is as follows: Among them, P usi The positive spinning reserve capacity provided for thermal power unit i, l U is the demand for positive spinning reserve due to load forecast error, S U The additional positive spinning reserve capacity required for the intervention of wind power, P D is the ultra-short-term load forecast value, P imax is the output upper limit of thermal power unit i, P i0 is the planned output of unit i on the day before, ΔP i is the real-time adjustment value of unit i, is the maximum upward regulation rate of unit i, and Δt is the regulation period; Negative spinning reserve constraint, expressed as follows: Among them, P dsi The negative spinning reserve capacity provided for thermal power unit i, l D is the demand for negative spinning reserve due to load forecast error, S D The additional negative spinning reserve capacity required for the intervention of wind power, P D is the ultra-short-term load forecast value, P i0 is the planned output of unit i on the day before, ΔP i is the real-time adjustment value of unit i, P imin is the lower limit of the output of thermal power unit i, is the maximum downward regulation rate of unit i, and Δt is the regulation period.

8. A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost according to claim 1, characterized in that: In step 3, the optimal values ​​of the new energy consumption capacity and power generation cost of the best half of the population are iteratively solved by the non-dominated sorting genetic algorithm, including the following steps: Step 3.1, establish a population and determine the initial population number, set the number of population iterations, crossover, mutation percentage, mutation rate and mutation step; initialize the optimal target value and the corresponding objective function related parameters, the optimal target value is the value of the objective function of maximum new energy consumption capacity and minimum power generation cost; Step 3.2, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x); Step 3.3, determine whether the value of the individual population meets the constraint conditions of the new energy consumption index; Step 3.4, when the value of the individual population does not meet the constraint condition of the new energy consumption index, the new energy consumption plan is adjusted based on the constraint processing mechanism, so that the value of the individual population after the adjustment meets the restriction of the constraint processing mechanism; Step 3.5, when the values ​​of the population individuals meet the constraint conditions of the new energy consumption index, the genetic algorithm based on non-dominated sorting performs non-dominated sorting on the population individuals, sorts each population according to the non-dominated frontier, calculates the crowding distance of each non-dominated level, and sorts the population according to the crowding distance; Step 3.6, using the population to create a descendant population, after selection, crossover and mutation, applying the constraint processing mechanism, evaluating the new energy absorption capacity objective function f1(x) and the comprehensive energy system operation cost f2(x); Step 3.7, merge the parent and offspring populations, perform non-dominated sorting, sort each population according to the non-dominated frontier, calculate the crowding distance of each non-dominated level, and sort the population according to the crowding distance. Step 3.8, select a solution, record the crowding distance in ascending order, store the non-dominated solutions in a list, plot the non-dominated solutions in the list, increment the generation count, and the loop ends.

9. A hybrid dispatching method based on maximum consumption of new energy and optimal power generation cost according to claim 1, characterized in that: In step 4, the multi-objective particle swarm algorithm is used to iteratively solve the optimal value of the new energy consumption capacity and power generation cost of the other half of the population, including the following steps: Step 4.1, generating a random particle swarm, initializing the optimal target value and the corresponding objective function related parameters, the optimal target value is the value of the objective function of maximizing the new energy consumption capacity and minimizing the power generation cost; Step 4.2, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x); Step 4.3, determine whether the particle group meets the constraints of the new energy consumption index; Step 4.4, when the particle swarm does not meet the constraint conditions of the new energy consumption index, the new energy consumption plan is adjusted based on the constraint processing mechanism, so that the adjusted particle swarm meets the restrictions of the constraint processing mechanism; Step 4.5, when the particle swarm meets the constraint conditions of the new energy consumption index, the fitness is calculated based on the multi-objective particle swarm algorithm to screen the non-inferior solution set. When a particle is not dominated by other particles, the particle is put into the non-inferior solution set, and a particle is randomly selected from the non-inferior solution set as the optimal particle of the swarm before the particle is updated; Step 4.6, update the non-inferior solution set. When the new particle is not dominated by other particles and particles in the current non-inferior solution set, put the new particle into the non-inferior solution set, and randomly select a particle from the non-inferior solution set as the optimal particle of the group before each particle update; Step 4.7, update the velocity and position of the particle, apply the constraint processing mechanism, evaluate the new energy absorption capacity objective function f1(x) and the integrated energy system operation cost f2(x), apply mutation and calculate the new solution; Step 4.8, add the non-dominant particles to the repository, determine the dominance of the new repository members, keep only the non-dominated members in the repository, update the grid and grid index, delete members if the repository is full, draw the members in the external repository, modify the inertia, and the loop ends.

Images