Alternating current and direct current hybrid micro-grid optimization scheduling method

Abstract

The invention discloses an alternating current and direct current hybrid micro-grid optimization scheduling method, and relates to the field of micro-grids. The method comprises the following steps: constructing a mathematical model of an AC / DC hybrid micro-grid system; according to the mathematical model, establishing an operation cost model of the AC / DC hybrid microgrid system; the operation cost model comprises operation cost and environment processing cost; the operation cost comprises the operation and maintenance cost of a controllable distributed power supply, the operation and maintenance cost of an uncontrollable distributed power supply, the cost of an energy storage unit, the cost of interaction with a large power grid and the cost of electric energy interaction of an electric vehicle; acquiring historical operation data of the AC / DC hybrid micro-grid system at each moment in a preset time period; according to historical operation data, an improved dung beetle optimization algorithm is adopted to solve the operation cost model, and an optimal scheduling result is obtained. According to the invention, comprehensive optimization of multiple dimensions of power generation, power storage and interaction of the AC / DC micro-grid system can be realized.

Description

Technical Field

[0001] The present application relates to the field of microgrids, and in particular to an AC / DC hybrid microgrid optimization scheduling method. Background Art

[0002] With the continuous innovation of clean energy technology, microgrid technology is gradually receiving more and more attention. In order to solve the problem of clean energy security, the operation optimization research of microgrid has become one of the important means to optimize the energy industry. Microgrids usually include controllable traditional gas and oil power generation units and uncontrollable photovoltaic and wind clean power generation units. Although the clean power generation units on the DC side have advantages over traditional gas and oil power generation in terms of environmental protection and practicality, there is still a considerable amount of AC load in the line, so AC and DC hybrid mutual transfer and optimization planning are essential.

[0003] However, due to the solar radiation and meteorological conditions caused by natural factors, as well as operation and maintenance factors such as equipment loss and shading, the supply of uncontrollable energy such as solar energy and wind energy in AC / DC microgrid systems is full of randomness. At the same time, since AC / DC microgrids involve multi-energy hybrid operation, including the conversion and interconnection of multiple energy sources, the complexity of control is also a major problem of AC / DC microgrid systems. At the same time, the fuel cost, operation efficiency, maintenance cost and pollutant emission of traditional controllable energy such as gas turbines and diesel generators are also a major handle for the operation of AC / DC hybrid microgrid systems. In addition, with the application of new energy, electric vehicles have become a key part of microgrids that cannot be ignored. Therefore, the optimization problem of AC / DC microgrid systems is the key that needs to be solved urgently. Summary of the invention

[0004] The purpose of this application is to provide an AC / DC hybrid microgrid optimization scheduling method, which can achieve comprehensive optimization of power generation, power storage, and interaction in multiple dimensions of the AC / DC microgrid system.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] The present application provides an AC / DC hybrid microgrid optimization scheduling method, comprising: constructing a mathematical model of an AC / DC hybrid microgrid system; the AC / DC hybrid microgrid system comprises: a controllable distributed power source, an uncontrollable distributed power source, an energy storage unit, a large power grid interaction unit and an electric vehicle; establishing an operation cost model of the AC / DC hybrid microgrid system according to the mathematical model; the operation cost model comprises: operation cost and environmental treatment cost; the operation cost comprises the operation and maintenance cost of the controllable distributed power source, the operation and maintenance cost of the uncontrollable distributed power source, the cost of the energy storage unit, the cost of interaction with the large power grid and the electric vehicle power interaction cost; obtaining historical operation data of the AC / DC hybrid microgrid system at each moment in a preset time period; solving the operation cost model using a dung beetle optimization algorithm based on the historical operation data to obtain an optimal scheduling result; the dung beetle optimization algorithm uses a tent chaotic map to initialize the population, the individual selection mechanism adopts a non-dominated sorting mechanism, and an elite reverse learning method is used to optimize the rolling ball behavior; the optimal scheduling result comprises the optimal output power of the controllable distributed power source, the output power of the uncontrollable distributed power source, the output power of the energy storage unit, the power of interaction with the large power grid and the power of the electric vehicle at each moment.

[0007] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0008] The present application provides an AC / DC hybrid microgrid optimization scheduling method, wherein the operating cost model includes operating costs and environmental treatment costs, wherein the operating costs include the operation and maintenance costs of controllable distributed power sources, the operation and maintenance costs of uncontrollable distributed power sources, the costs of energy storage units, the costs of interaction with the large power grid, and the costs of electric vehicle power interaction, thereby achieving comprehensive optimization of pollution and operation and maintenance of controllable distributed power sources, optimization of operation and maintenance of uncontrollable distributed power sources, and combining energy storage elements, large power grid interaction, and electric vehicle power interaction, thus filling the research gap and achieving comprehensive optimization effects in multiple dimensions of power generation, storage, and interaction.

[0009] The present application provides an AC / DC hybrid microgrid optimization scheduling method, which uses Tent chaotic mapping to solve the problem of uneven population distribution in the initialization of the original dung beetle optimization algorithm; uses a non-dominated sorting mechanism to enhance the selection process of the dung beetle optimization algorithm; uses elite reverse learning to enhance the exploration ability and diversity of the population, and optimizes the distribution of the number of different types of dung beetles, thereby achieving flexible adjustment of the search strategy at different stages, thereby improving the balance between global search performance and local search performance, and solving the problems of slow convergence speed, low global search efficiency and easy falling into local optimal solutions of the dung beetle optimization algorithm in the field of economic optimization scheduling of AC / DC hybrid microgrids. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of the present application 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0011] Figure 1 A schematic diagram of a flow chart of an AC / DC hybrid microgrid optimization scheduling method in an embodiment of the present application;

[0012] Figure 2 A schematic diagram of the structure of an AC / DC hybrid microgrid system provided in another embodiment of the present application;

[0013] Figure 3 A tree diagram showing the composition relationship of an objective function provided by another embodiment of the present application;

[0014] Figure 4 A schematic diagram of a dung beetle optimization algorithm process provided by another embodiment of the present application;

[0015] Figure 5 A comparison diagram of convergence curve optimization provided in one embodiment of the present application. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0017] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0018] In AC / DC hybrid microgrid systems, particle swarm optimization, genetic algorithms and other optimization algorithms are widely used. In recent years, some new systems have also adopted the gray wolf algorithm, which has achieved certain results. As for the dung beetle optimization algorithm, a new optimization algorithm, when applied to the field of AC / DC hybrid microgrid system optimization, there are also considerable gaps in its convergence optimization speed, optimization accuracy, whether it is easy to fall into local optimality, and how to improve the algorithm in a targeted manner when these problems arise. In view of this, Figure 1 As shown, the embodiment of the present application provides an AC / DC hybrid microgrid optimization scheduling method, including the following steps 101 to 104. Among them:

[0019] Step 101: construct a mathematical model of an AC / DC hybrid microgrid system; the AC / DC hybrid microgrid system includes: a controllable distributed power source, an uncontrollable distributed power source, an energy storage unit, a large power grid interaction unit and an electric vehicle.

[0020] Step 102: Based on the mathematical model, an operation cost model of the AC / DC hybrid microgrid system is established; the operation cost model includes: operation cost and environmental treatment cost; the operation cost includes the operation and maintenance cost of the controllable distributed power source, the operation and maintenance cost of the uncontrollable distributed power source, the cost of the energy storage unit, the cost of interaction with the large power grid and the electric vehicle power interaction cost.

[0021] Step 103: Obtain historical operation data of the AC / DC hybrid microgrid system at each moment in a preset time period.

[0022] Step 104: According to the historical operation data, the operation cost model is solved by using a dung beetle optimization algorithm to obtain an optimal scheduling result; the dung beetle optimization algorithm uses a tent chaotic map to initialize the population, the individual selection mechanism adopts a non-dominated sorting mechanism, and an elite reverse learning method is used to optimize the rolling behavior; the optimal scheduling result includes the output power of the optimal controllable distributed power source, the output power of the uncontrollable distributed power source, the output power of the energy storage unit, the power of interaction with the large power grid, and the power of the electric vehicle at each moment.

[0023] By implementing the above steps 101 to 104, the operating cost model includes operating costs and environmental treatment costs. The operating costs include the operation and maintenance costs of controllable distributed power sources, the operation and maintenance costs of uncontrollable distributed power sources, the costs of energy storage units, the costs of interaction with the large power grid, and the costs of electric vehicle power interaction. This achieves comprehensive optimization of pollution and operation and maintenance of controllable distributed power sources, optimization of operation and maintenance of uncontrollable distributed power sources, and combines energy storage elements, large power grid interaction, and electric vehicle power interaction, filling the research gap and achieving comprehensive optimization effects in multiple dimensions of power generation, storage, and interaction.

[0024] The dung beetle optimization algorithm used in this application is an improved dung beetle optimization algorithm based on the use of Tent chaotic mapping, non-dominated sorting mechanism, and elite reverse learning.

[0025] In another exemplary embodiment of the present application, the AC / DC hybrid microgrid system includes: a controllable distributed power source, an uncontrollable distributed power source, an energy storage unit, a large power grid interaction unit and an electric vehicle. When the controllable distributed power source includes a diesel generator, the mathematical model of the AC / DC hybrid microgrid system in the above step 101 includes a mathematical model of the diesel generator, and the mathematical model of the diesel generator characterizes the relationship between the fuel and power of the diesel generator. The mathematical model of the diesel generator is:

[0026] Q(t)=α1P(t)+βP0

[0027] Where Q(t) is the fuel of the diesel generator at time t, P(t) is the power of the diesel generator at time t, P0 is the rated power of the diesel generator, and α1 and β are both power coefficients.

[0028] Uncontrollable distributed power sources include photovoltaic power generation units and wind power generation units. The mathematical model of the AC / DC hybrid microgrid system also includes the mathematical model of the photovoltaic power generation unit and the mathematical model of the wind power generation unit. The mathematical model of the photovoltaic power generation unit is:

[0029]

[0030] Where P PV is the output power of the photovoltaic power generation unit, P m is the output power of the photovoltaic power generation unit under standard test conditions, I is the light intensity, and I STC is the solar radiation intensity of the photovoltaic power generation unit under standard test conditions, α2 is the temperature coefficient of the photovoltaic power generation unit, T' is the temperature of the photovoltaic power generation unit, T M It is the temperature of the photovoltaic power generation unit under standard test conditions.

[0031] The mathematical model of the wind power generation unit is:

[0032]

[0033] Where P WT is the output power of the wind power generation unit, P r is the rated power of the wind power generation unit, v is the wind speed, v ci is the cut-in wind speed, v m is the rated wind speed, v co To cut out the wind speed, a', b', c', and d' are all wind speed parameters.

[0034] The mathematical model of the photovoltaic power generation unit mentioned above may also be called a photovoltaic power generation power expression, and the mathematical model of the wind power generation unit may also be called a wind power generation power expression.

[0035] The mathematical model of the AC / DC hybrid microgrid system also includes the mathematical model of the energy storage unit and the mathematical model of the electric vehicle. The mathematical model of the energy storage unit is:

[0036]

[0037] In the formula, SOC(t) is the charge capacity of the battery at time t, SOC(t-1) is the charge capacity of the battery at time t-1, and P bess (t) is the charging and discharging power of the battery at time t, η -is the efficiency during discharge, η + The mathematical model of the energy storage unit characterizes the state of charge of the energy storage unit during charging and discharging.

[0038] The mathematical model of an electric vehicle is:

[0039]

[0040] In the formula, f d (x) is the probability density function of the electric vehicle schedule, x is the daily mileage of the electric vehicle, μ d is the mean, σ d is the variance.

[0041] If the electric vehicle is directly connected to the power grid after returning home after the last trip of the day, the probability density distribution of the time of returning home is expressed as f s (x) is the probability density distribution of the time of returning home, μ s and σ s are the mean and variance of the time of returning home respectively; the probability density distribution of the time of leaving home is expressed as f e (x) is the probability density distribution of the time of leaving home, μ e and σ e are the mean and variance of the time of leaving home, respectively.

[0042] The electric vehicle schedule probability density function describes the probability distribution of electric vehicle usage at different time periods of the day and is a mathematical expression of electric vehicle usage behavior.

[0043] In another exemplary embodiment of the present application, the objective function of the operating cost model is:

[0044] f

[0045] F=C1+C2

[0046] Where C1 is the operating cost and C2 is the environmental treatment cost. The objective function is to minimize the total cost, including the operating cost and the environmental treatment cost.

[0047] In another exemplary embodiment of the present application, when the controllable distributed power source includes a micro gas turbine and a diesel generator, and the uncontrollable distributed power source includes a photovoltaic unit and a wind unit, the composition structure of the AC / DC hybrid microgrid system is as follows: Figure 2 As shown. Then the expression of the operating cost C1 is:

[0048] C1=C MT_om +C DE_om +a×C grid +C PV_om +C WT_om +CBESS +C EV

[0049] In the formula, C MT_om is the operation and maintenance cost of the micro gas turbine, C DE_om is the operation and maintenance cost of the diesel generator, C grid is the cost of interacting with the large power grid, C PV_om is the operation and maintenance cost of the photovoltaic unit, C WT_om is the operation and maintenance cost of the wind turbine, C BESS is the cost of the energy storage unit, C EV is the electric energy interaction cost of electric vehicles; a is the grid connection coefficient, which is 1 when connected to the grid and 0 when off-grid.

[0050] Operation and maintenance cost of micro gas turbine MT_om Divided into fuel cost C MT_o and maintenance cost C MT_m , the operation and maintenance cost of the micro gas turbine is C MT_om The expression is:

[0051] C MT_om =C MT_o +C MT_m

[0052] In the formula, C MT_o is the fuel cost of the microturbine, C is the natural gas price, Δt is the time interval, LHV is the lower heating value, P MT is the output power of the micro gas turbine, η MT is the efficiency of the micro gas turbine, T is the cycle; C MT_m is the maintenance cost of the micro gas turbine, K MT is the maintenance factor of the micro gas turbine.

[0053] Operation and maintenance cost of diesel generators C DE_om Divided into fuel cost C DE_o and maintenance cost C DE_m , the operation and maintenance cost of diesel generators C DE_om The expression is:

[0054] C DE_om =C DE_o +C DE_m

[0055] In the formula, C DE_o is the fuel cost of the diesel generator, P DE is the output power of the diesel generator, c1, c2, c3 are the combustion coefficients of the diesel generator; C DE_mThe maintenance cost of diesel generators is K DE_m is the maintenance factor of the diesel generator, P DE (t) is the output power of the diesel generator at time t.

[0056] When the grid connection coefficient a = 1, the cost of interacting with the large grid is C grid The expression is:

[0057]

[0058] Where P grid (t) is the power interacting with the large power grid at time t, G grid (t) is the electricity purchase price or electricity sales price at time t.

[0059] Since photovoltaic and wind power generation are clean energy, the operation and maintenance cost of photovoltaic and wind power generation units is C PV_om , C WT_om Only maintenance costs are included.

[0060] The operation and maintenance cost of the photovoltaic unit C PV_om The expression is:

[0061]

[0062] In the formula, K PV_m is the maintenance factor of the photovoltaic power generation unit, and PPV(t) is the output power of the photovoltaic power generation unit at time t.

[0063] Operation and maintenance cost of wind turbines C WT_om The expression is:

[0064]

[0065] In the formula, K WT_m is the maintenance factor of the wind power unit, P WT (t) is the output power of the wind power generation unit at time t.

[0066] The cost of the energy storage unit C BESS The expression is:

[0067]

[0068] Where P BESS is the output power of the energy storage unit, which is positive when charging and negative when discharging; C p is the discounted value of the total investment cost, t BESS is the annual operation time of the energy storage system, K BESS is the investment recovery factor of the energy storage system; G BESSis the total investment cost of the energy storage system, , δ is the depreciation rate of the energy storage unit, and n' is the designed service life.

[0069] Electric vehicle electricity interaction cost C EV The expression is:

[0070]

[0071] Where n is the number of electric vehicles, P EVc,i (t), P EVd,i (t) are the charging power and discharging power of the ith electric vehicle at time t, Δt c,i and Δt d,i are the time intervals for charging and discharging of the i-th electric car at time t.

[0072] In another exemplary embodiment of the present application, when the controllable distributed power source includes a micro gas turbine and a diesel generator, the expression of the environmental treatment cost C2 is:

[0073] C2=C MT_P +C DT_P

[0074] In the formula, C MT_P is the environmental treatment cost of the micro gas turbine, C DT_P Environmental treatment costs for diesel generators.

[0075] Environmental treatment cost of micro gas turbine C MT_P The expression is:

[0076]

[0077] Where D y H y is the control cost coefficient of the main pollutant y, P MT (t) is the output power of the micro gas turbine at time t, Δt is the time interval, and M is the amount of the main pollutant.

[0078] Environmental treatment cost of diesel generators C DT_P The expression is:

[0079]

[0080] Where P DE (t) is the output power of the diesel generator at time t.

[0081] The composition relationship of the objective function can be expressed as follows Figure 3 The tree diagram shown.

[0082] In another exemplary embodiment of the present application, when the controllable distributed power sources include micro gas turbines and diesel generators, and the uncontrollable distributed power sources include photovoltaic units and wind units, the constraints of the operating cost model include: power balance constraints, output constraints, climbing power constraints, grid-connected transmission power constraints, energy storage unit constraints, and electric vehicle charge state constraints.

[0083] The power balance constraint is:

[0084] P L +P EV +P BESS =P MT +P DE +P Grid +P PV +P WT

[0085] Where P L is the load power, P EV is the charging and discharging power of the electric vehicle, P BESS is the output power of the energy storage unit, P MT is the output power of the micro gas turbine, P DE is the output power of the diesel generator, P Grid is the power of interaction with the large power grid, P PV is the output power of the photovoltaic power generation unit, P WT is the output power of the wind power generation unit.

[0086] Output constraints include output constraints of micro gas turbines, output constraints of diesel generators, output constraints of wind power generation and output constraints of photovoltaic power generation.

[0087] The output constraint of the micro gas turbine is expressed as:

[0088]

[0089] Where P MT (t) is the output power of the micro gas turbine at time t, and are the minimum and maximum output power of the micro gas turbine at time t respectively.

[0090] The output constraint of the diesel generator is expressed as:

[0091]

[0092] Where P DE (t) is the output power of the diesel generator at time t, and are the minimum and maximum output power of the diesel generator at time t respectively.

[0093] The output constraint of wind power generation is expressed as:

[0094]

[0095] Where P WT (t) is the output power of the wind power generation unit at time t, and are the minimum and maximum output power of the wind power generation unit at time t respectively.

[0096] The output constraint of photovoltaic power generation is expressed as:

[0097]

[0098] Where P PV (t) is the output power of the photovoltaic power generation unit at time t, and are the minimum and maximum output power of the photovoltaic power generation unit at time t respectively.

[0099] The ramp power constraints include: the ramp power constraints of micro gas turbines and the ramp power constraints of diesel generators. For photovoltaic units and wind units, photovoltaic power generation depends on the intensity of solar radiation, while wind power generation depends on wind speed. The output power of these two renewable energy sources is essentially determined by natural conditions, rather than adjusted by active control. Therefore, there is no traditional ramp power constraint.

[0100] The ramp power constraint of the micro gas turbine is:

[0101] |P MT (t)-P MT (t-1)|≤r MT

[0102] Where P MT (t-1) is the output power of the micro gas turbine at time t-1, r MT is the upper limit of the ramp rate of the micro gas turbine.

[0103] The climbing power constraint of the diesel generator is:

[0104] |P DE (t)-P DE (t-1)|≤r DT

[0105] Where P DE (t-1) is the output power of the diesel generator at time t-1, r DT It is the upper limit of the ramp rate of the diesel generator.

[0106] When the grid-connected coefficient a=1, the grid-connected transmission power constraint is:

[0107]

[0108] Where P grid (t) is the power interacting with the large power grid at time t, and are the minimum and maximum values ​​of the interaction power with the large power grid at time t respectively.

[0109] The energy storage unit constraints are:

[0110]

[0111] Where P BESS (t) is the output power of the energy storage unit at time t, and are the lower and upper limits of the output power of the energy storage unit at time t; SOC(t) is the state of charge of the energy storage unit at time t, and are the lower and upper limits of the energy storage capacity at time t respectively.

[0112] The state of charge constraint of electric vehicles is:

[0113]

[0114] In the formula, SOC EV (t) is the state of charge of the electric vehicle at time t, and are the lower and upper limits of the electric vehicle battery capacity at time t respectively.

[0115] In another exemplary embodiment of the present application, the above step 103 obtains the historical operation data of the AC / DC hybrid microgrid system at each moment in a preset time period, specifically: at each hour of a day, the output power of wind power generation, the output power of photovoltaic power generation and the power load in the AC / DC hybrid microgrid system are obtained.

[0116] In another exemplary embodiment of the present application, the present application improves the original dung beetle algorithm and applies the improved multi-strategy to improve the dung beetle optimization algorithm. The original dung beetle algorithm can be summarized as the following several processes: ① Initialize the dung beetle population and the parameters of the dung beetle algorithm; ② Calculate the fitness value of all dung beetle positions according to the objective function; ③ Update the positions of all dung beetles; ④ Determine whether each updated dung beetle is out of the boundary; ⑤ Update the current optimal solution and its fitness value; ⑥ Repeat the above steps, and after reaching the maximum number of iterations, output the global optimal value and its optimal solution.

[0117] For step ①, the maximum number of iterations T, population size, upper bound Lb, lower bound Ub, and dimension D of the optimization problem of the algorithm should be set for initialization; the dung beetle population is divided into four types: rolling dung beetles, egg-laying dung beetles, foraging dung beetles, and thief dung beetles.

[0118] For steps ② and ③, since different dung beetles perform different duties, the four dung beetles update their positions in different ways. The rolling dung beetle uses the sun as a navigation to ensure that the dung ball rolls on a straight path. Natural factors such as light intensity and wind will affect the route of the rolling dung beetle. The rolling dung beetle updates its position in the following way: i' (t'+1)=x i' (t')+α*k*x i' (t'-1)+b*Δx, where t' represents the current iteration number, x i' (t') represents the position information of the i'th dung beetle at the t'th iteration. α is a natural coefficient, indicating whether it deviates from the original direction. It is assigned to -1 or 1 according to the probability method. k∈(0,0.2) represents the deflection coefficient. b∈(0,1) represents a constant. k and b are set to 0.1 and 0.3 respectively. Δx is used to simulate the change of light intensity, which is expressed as Δx=|x i' (t')-X w |, where X w Represents the global worst position. When the dung beetle encounters an obstacle and cannot move forward, it needs to adjust its direction by dancing. The formula for the rolling ball dung beetle to update its position by dancing is defined as follows: Where θ∈[0,π] represents the deflection angle. When θ is equal to 0, π / 2 or π, the position of the dung beetle will not be updated.

[0119] For step ④, the oviposition dung beetle adopts the boundary selection strategy to simulate the oviposition area of ​​female dung beetles. The oviposition area is defined as In the formula, represents the current local optimal position, Lb * and Ub * Respectively represent the lower and upper limits of the spawning area, R = 1-t / T max , T max represents the maximum number of iterations, Lb and Ub represent the lower and upper limits of the optimization problem, respectively. During the iteration process, the position of the dung beetle laying eggs changes dynamically, which is defined as In the formula, B i' (t') is the position information of the i'th egg-laying dung beetle at the t'th iteration, b1 and b2 represent two independent random vectors of size 1×D, and D represents the dimension of the optimization problem.

[0120] It is necessary to establish an optimal foraging area to guide the foraging dung beetles to find food and simulate their foraging behavior, where the optimal foraging area is defined as In the formula, represents the current local optimal position, Lb' and Ub' represent the lower and upper limits of the optimal foraging area, respectively, and the position update of the foraging dung beetle is represented by x i' (t'+1)=x i' (t')+C1(x i' (t')-Lb')+C2(x i' (t')-Ub'), where x i' (t') is the position information of the i'th dung beetle at the t'th iteration, C1 represents a random number following a normal distribution, and C2∈(0,1) represents a random vector.

[0121] For step ⑤, due to is the best foraging area, so the position update of the thief dung beetle can be expressed as in In the formula, is the position information of the i'th thief dung beetle at the t'th iteration, g represents a random vector of size 1×D that follows the normal distribution, and S represents a constant value.

[0122] The original dung beetle algorithm has room for improvement in population initialization, convergence ability, and local optimality. Therefore, this application will improve it from three directions: 1. Use Tent chaotic mapping to solve the problem of uneven population distribution in the initialization of the original algorithm; 2. Use a non-dominated sorting mechanism to enhance the selection process of the algorithm; 3. Use elite reverse learning to enhance the exploration ability and diversity of the population.

[0123] For improvement 1, under the condition of a limited number of initial populations, a more uniform distribution means a greater possibility of being closer to the global optimal solution. Although increasing the number of initial populations can expand the coverage of the solution space, it will also increase the burden on the optimization algorithm. Therefore, the quality of the initial population has a great impact on the convergence speed of the optimization algorithm and the accuracy of the solution. Increasing the diversity of the initial population can significantly improve the performance of the optimization algorithm. The dung beetle algorithm usually uses random generation to generate the initial population, but this method often leads to an uneven distribution of the initial population, thereby reducing the diversity of the initial population. This improved method uses the Tent chaotic map to improve and optimize the population initialization. The distribution function of the Tent map is more uniform, and the chaotic sequence it generates is more globally ergodic. It can be expressed as Where k represents the current population size, Z k is the kth population, with an initial value of 0.152, and β is the chaos parameter, with an initial value of 0.4.

[0124] In another exemplary embodiment of the present application, the traditional individual selection mechanism often lacks a clear evaluation and ranking of the superiority and inferiority relationship between individuals in the population, which may make it difficult for the algorithm to effectively distinguish and utilize the advantages of different individuals when dealing with multi-objective optimization problems, thereby affecting the overall performance and convergence speed of the algorithm. In order to solve this problem, for improvement 2, this method introduces a non-dominated sorting mechanism to improve the selection process of the dung beetle algorithm. The non-dominated sorting mechanism in the above step 104 specifically includes the following steps 201 to 207:

[0125] Step 201: input an initial population and multiple objective functions in advance.

[0126] Step 202: For each individual in the initial population, initialize the solution set dominated by each individual and the number of times it is dominated to 0.

[0127] Step 203: Compare any pair of individuals p and q in the initial population. If the evaluation result of individual p on all objective functions is not inferior to that of individual q, and the evaluation result of individual p on at least one objective function is better than that of individual q, then individual p dominates individual q, and individual q is added to the solution set of individual p, and the number of times individual q is dominated is increased by 1.

[0128] Step 204: Put individuals with a domination count of 0 into the first frontier, and assign a level of 1 to the individuals in the first frontier.

[0129] Step 205: Let the value of u be 2.

[0130] Step 206: Add the individuals with a dominance of 1 in the solution set of each individual in the first front to the u-th front, and set the individual level in the u-th front to u.

[0131] Step 207: Determine whether all individuals in the initial population have been assigned ranks. If not, increase the value of u by 1 and return to the step of "adding the individuals with 1 dominated times in the solution set of each individual in the first front to the u-th front, and setting the rank of the individuals in the u-th front to u"; if so, obtain the rank of each individual in the initial population, sort them by rank, and output the sorted initial population.

[0132] The non-dominated sorting process of steps 201 to 207 mainly includes (1) initialization; (2) non-dominated sorting; (3) crowding distance calculation; and (4) selection and generation of a new population.

[0133] The pre-input content includes the population P, the number of objective functions M (F=C1+C2 in this application means two objective functions), and the decision variable dimension D. On this basis, step (2) includes the following parts: ① initialization; ② dominance relationship calculation; ③ determination of the first frontier; ④ construction of other frontiers; (5) looping.

[0134] ① Initialization. For each individual p: initialize the solution set S dominated by p p and the number of times dominated n p =0.

[0135] ② Dominance relationship calculation. For each pair of individuals p and q in the population: compare their performance on the objective function; establish a dominance rule. If p is not inferior to q in all objectives and is superior to q in at least one objective, then p dominates q. The formula can be expressed as According to the dominance rule, if p dominates q, then q is added to the solution set S p , the formula is S p =S p ∪{q}, if q dominates p, then increase the number of times p is dominated by n p .f i (p) and f i (q) represents the evaluation result of individuals p and q on the i-th objective function (1,…,i…,M), which is an abstract mathematical concept. It is increased by 1, because the increase in the number of dominated times is one-time, which can be expressed by a formula similar to the above: Ifq dominates p, then n p =n p +1.

[0136] Among them, f i (p) and f i (q) represents the evaluation result of individuals p and q on the i-th objective function (1,…,i…,M), which is an abstract mathematical concept. It is increased by 1, because the increase in the number of dominated times is one-time, which can be expressed by a formula similar to the above: Ifq dominates p, then n p =n p +1.

[0137] ③ Determine the first frontier. If individual p is not dominated by any other individual, that is, n p =0, it is put into the first frontier F1 and assigned level 1; the first frontier individual set F1 contains all non-dominated individuals.

[0138] ④Construct other frontiers. For each frontier F i , from F i The dominating set S of each individual p in p Remove the number of dominated times one by one, that is, n q =n q -1; if q is dominated by n times q If it decreases to 0, add q to the next frontier F i+1, and set its rank to i+1; repeat this process until all individuals are sorted. After the above steps, output the sorted population P, including the non-dominated rank of each individual.

[0139] Specifically, level 1 (F1) contains all individuals that are not dominated by any other individuals, level 2 (F2) contains all individuals that are dominated only by individuals in level 1, and so on. i ) contains all individuals that are dominated only by individuals of rank less than i.

[0140] This is a cyclic process. i+1 After confirming, continue to i+1 Follow the same steps (consider it as a new F i ), we can get F i+2 , until all individuals are sorted.

[0141] The congestion calculation in step (3) includes the following parts: ① initialization; ② boundary individual processing; ③ distance calculation.

[0142] ① Initialization. For each frontier, initialize the crowding distance of each individual to 0.

[0143] ② Boundary individual processing: For each objective function, the individuals in each frontier are sorted by the objective value, and the crowding distance between the individual with the smallest objective value and the individual with the largest objective value in each frontier is set to infinity.

[0144] ③ Distance calculation. For each individual in the middle of the frontier, calculate its crowding distance, the formula is: Accumulate the distance of each objective function to obtain the total crowding distance of the individual, and output the population P containing the crowding distance information. Accumulate the crowding distance of all objective functions of each individual to obtain the total crowding distance of each individual, and output the population containing the total crowding distance.

[0145] Where, d k' is the crowding distance of individual k' in each frontier, d k'0 is the initial crowding distance of individual k' in each frontier, f k'-1 (m) is the target value of the mth objective function for each individual k'-1 in the frontier, f k'+1 (m) is the target value of the mth objective function for each individual k'+1 in the frontier, f min (m) is the minimum target value of the mth target function, f max (m) is the maximum target value of the mth objective function.

[0146] For step (4), selecting and generating a new population, the selection part is to give priority to individuals with lower non-dominated levels. If multiple individuals are at the same level, individuals with larger crowding distances are selected to ensure the diversity of the solution set; generating a new population is to generate the next generation of population through genetic operations such as crossover and mutation.

[0147] Step (5) is looped, and the loop content is to repeat the non-dominated sorting, crowding distance calculation, selection and generation of new populations until the preset termination condition (number of iterations, number of calculations) is reached.

[0148] In another exemplary embodiment of the present application, for improvement 3, the randomness of the dung beetle algorithm may hinder the convergence speed and accuracy of the algorithm in the later stages of iteration. In order to optimize this defect, the present application introduces an elite reverse learning strategy to improve the rolling behavior of the dung beetle algorithm. The elite reverse learning method can more effectively avoid falling into local optimal solutions, accelerate the convergence process of the algorithm, and improve the quality of the final solution by identifying and utilizing elite individuals in the population, that is, those solutions that perform well on the objective function, to guide the search direction of other individuals.

[0149] For improvement 3, the elite reverse learning process adopted by this method includes: ① selecting elite individuals; ② calculating the boundaries of elite individuals; ③ generating reverse solutions; ④ boundary checking; ⑤ evaluating and replacing reverse solutions.

[0150] ① Select elite individuals. In each iteration, the top 10% of individuals are selected from the current non-dominated solution set as elite individuals. These individuals usually have better fitness function values ​​(objective function values).

[0151] ② Calculate the boundaries of elite individuals. For the selected elite individuals, calculate the minimum value T on each variable dimension Lb and the maximum value T Ub , forming the boundary range of elite individuals. This step is to generate new solutions during reverse learning, which are inversely related to the positions of elite individuals.

[0152] ③ Generate reverse solution. According to the upper and lower boundaries of the elite individuals, generate the reverse solution of each elite individual. The formula is: X'=rand()×(T Lb +T Ub )-X, where rand() is a random number uniformly distributed between [0,1], which is used to increase the randomness of the reverse solution. X is the position vector of the current elite individual.

[0153] ④ Boundary check. In order to ensure that the generated reverse solution is within the feasible search space, it is necessary to perform boundary check on the reverse solution. If the reverse solution exceeds the boundary, its value is limited between the upper and lower bounds. That is, if it exceeds the upper boundary, the reverse solution is directly taken as the upper boundary, and the same applies to the lower boundary.

[0154] ⑤ Evaluation and replacement of reverse solutions. Calculate the objective function value of each reverse solution and compare it with the objective function value of the original elite individual. If the reverse solution is better than the original elite individual in all objectives (that is, the reverse solution dominates the original solution), replace the elite individual with the reverse solution.

[0155] In another exemplary embodiment of the present application, the dung beetle optimization algorithm process used in the present application is as follows: Figure 4 The specific implementation process of using the dung beetle optimization algorithm to solve the operating cost model is steps 301 to 307:

[0156] Step 301: Initialize the dung beetle population using Tent chaotic mapping to obtain an initial dung beetle population.

[0157] Step 302: Use a non-dominated sorting mechanism to perform non-dominated sorting on the initial dung beetle population to obtain a sorted dung beetle population.

[0158] Step 303: updating the positions of the dung beetles in the sorted dung beetle population to obtain a new dung beetle population.

[0159] Step 304: Update the fitness values ​​of the dung beetles in the new dung beetle population to obtain the current global optimal solution.

[0160] Step 305: Use a non-dominated sorting mechanism to perform non-dominated sorting on the new dung beetle population to obtain a sorted new dung beetle population.

[0161] Step 306: Adopting the elite reverse learning method to optimize the rolling behavior and reverse learning the elite individuals.

[0162] Step 307: Determine whether the maximum number of iterations has been reached. If not, replace the sorted dung beetle population with the sorted new dung beetle population, and return to the step of "updating the positions of the dung beetles in the sorted dung beetle population to obtain a new dung beetle population"; if so, output the current global optimal solution.

[0163] Figure 5 The paper demonstrates the comparison of the cost optimization of the Improved Dung Beetle Optimization (IDBO) algorithm used in this application with the Grey Wolf Optimization (GWO) algorithm commonly used in the field of microgrid optimization and the original Dung Beetle Optimizer (DBO), which verifies the superiority and uniqueness of the Dung Beetle Optimizer (DBO) algorithm used in this application compared with other algorithms.

[0164] This application clearly defines the constraints and optimization objective function of the cost problem in the AC / DC hybrid microgrid, and determines the components of each system. The optimization objective function is to minimize the cost.

[0165] Existing technologies generally focus on microgrid systems such as single photovoltaic wind power or single fuel engine, without fully integrating the term AC and DC. At the same time, the discussion on the interaction with large power grids such as power purchase and sales is not complete. In addition, existing research still has gaps in the research on electric vehicles connected to microgrids. At the same time, there is also a gap in the cross-application of the new optimization algorithm, the dung beetle optimization algorithm, in the field of microgrids.

[0166] The objective function and power constraints designed in this application cover the above areas, and can achieve comprehensive optimization of pollution and operation and maintenance of micro gas turbines and diesel generators. The operation and maintenance of photovoltaic units and wind units are optimized, and energy storage components and large power grid interactions are combined, and electric vehicles are connected to microgrids, filling the research gap and achieving comprehensive optimization effects in multiple dimensions of power generation, storage, and interaction.

[0167] This application uses an improved dung beetle optimization algorithm, which is applied to the economic optimization scheduling problem of AC / DC hybrid microgrids. Economic scheduling is performed by simulating the behavior of dung beetles, such as rolling, reproduction, local exploration, and global improvement. Through the improvement of the dung beetle optimization algorithm, the Tent chaotic map is used to solve the problem of uneven population distribution in the initialization of the original algorithm; the non-dominated sorting mechanism is used to enhance the selection process of the algorithm; elite reverse learning is used to enhance the exploration ability and diversity of the population, and the distribution of the number of different types of dung beetles is optimized, so that the search strategy can be flexibly adjusted at different stages, thereby improving the balance between global search performance and local search performance. This solves the problems of slow convergence speed, low global search efficiency, and easy to fall into local optimal solutions of the dung beetle optimization algorithm in the field of economic optimization scheduling of AC / DC hybrid microgrids.

[0168] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0169] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. An AC / DC hybrid microgrid optimization scheduling method, characterized in that: include: Construct mathematical model of AC / DC hybrid microgrid system; The AC / DC hybrid microgrid system includes: a controllable distributed power source, an uncontrollable distributed power source, an energy storage unit, a large power grid interaction unit and an electric vehicle; According to the mathematical model, an operation cost model of the AC / DC hybrid microgrid system is established; the operation cost model includes: operation cost and environmental treatment cost; the operation cost includes the operation and maintenance cost of the controllable distributed power source, the operation and maintenance cost of the uncontrollable distributed power source, the cost of the energy storage unit, the cost of interaction with the large power grid and the electric vehicle power interaction cost; Obtain historical operation data of the AC / DC hybrid microgrid system at each moment within a preset time period; According to the historical operation data, the dung beetle optimization algorithm is used to solve the operation cost model to obtain the optimal scheduling result; the dung beetle optimization algorithm uses Tent chaotic mapping to initialize the population, the individual selection mechanism adopts a non-dominated sorting mechanism, and an elite reverse learning method is used to optimize the rolling ball behavior; the optimal scheduling result includes the output power of the optimal controllable distributed power source, the output power of the uncontrollable distributed power source, the output power of the energy storage unit, the power of interaction with the large power grid and the power of the electric vehicle at each moment.

2. The AC / DC hybrid microgrid optimization scheduling method according to claim 1 is characterized in that: When the controllable distributed power source includes a diesel generator, the mathematical model of the diesel generator is: Q(t)=α1P(t)+βP0; In the formula, Q(t) is the fuel of the diesel generator at time t, P(t) is the power of the diesel generator at time t, P0 is the rated power of the diesel generator, and α1 and β are both power coefficients; The uncontrollable distributed power source includes a photovoltaic power generation unit and a wind power generation unit; The mathematical model of the photovoltaic power generation unit is: Where P PV is the output power of the photovoltaic power generation unit, P m is the output power of the photovoltaic power generation unit under standard test conditions, I is the light intensity, and I STC is the solar radiation intensity of the photovoltaic power generation unit under standard test conditions, α2 is the temperature coefficient of the photovoltaic power generation unit, T' is the temperature of the photovoltaic power generation unit, T M is the temperature of the photovoltaic power generation unit under standard test conditions; The mathematical model of the wind power generation unit is: Where P WT is the output power of the wind power generation unit, P r is the rated power of the wind power generation unit, v is the wind speed, v ci is the cut-in wind speed, v m is the rated wind speed, v co is the cut-out wind speed, a', b', c', d' are wind speed parameters; The mathematical model of the energy storage unit is: In the formula, SOC(t) is the charge capacity of the battery at time t, SCO(t-1) is the charge capacity of the battery at time t-1, and P bess (t) is the charging and discharging power of the battery at time t, η - is the efficiency during discharge, η + The efficiency of charging; The mathematical model of the electric vehicle is: In the formula, f d (x) is the probability density function of the electric vehicle schedule, x is the daily mileage of the electric vehicle, μ d is the mean, σ d is the variance; If the electric vehicle is directly connected to the power grid after returning home after the last trip of the day, the probability density distribution of the time of returning home is expressed as f s (x) is the probability density distribution of the time of returning home, μ s and σ s are the mean and variance of the time of returning home respectively; the probability density distribution of the time of leaving home is expressed as f e (x) is the probability density distribution of the time of leaving home, μ e and σ e are the mean and variance of the time of leaving home, respectively.

3. The AC / DC hybrid microgrid optimization scheduling method according to claim 1 is characterized in that: The objective function of the operating cost model is: minF; F = C1 + C2; In the formula, C1 is the operating cost and C2 is the environmental treatment cost.

4. The AC / DC hybrid microgrid optimization scheduling method according to claim 3 is characterized in that: When the controllable distributed power source includes a micro gas turbine and a diesel generator, and the uncontrollable distributed power source includes a photovoltaic unit and a wind unit, the expression of the operating cost C1 is: C1=C MT_om +C DE_om +a×C grid +C PV_om +C WT_om +C BESS +C EV ; In the formula, C MT_om is the operation and maintenance cost of the micro gas turbine, C DE_om is the operation and maintenance cost of the diesel generator, C grid is the cost of interacting with the large power grid, C PV_om is the operation and maintenance cost of the photovoltaic unit, C WT_om is the operation and maintenance cost of the wind turbine, C BESS is the cost of the energy storage unit, C EV is the electric energy interaction cost of electric vehicles; a is the grid connection coefficient, a is 1 when connected to the grid and 0 when off-grid; Micro gas turbine operation and maintenance cost C MT_om The expression is: C MT_om =C MT_0 +C MT_m ; In the formula, C MT_0 is the fuel cost of the microturbine, C is the natural gas price, Δt is the time interval, LHV is the lower heating value, P MT is the output power of the micro gas turbine, η MT is the efficiency of the micro gas turbine, T is the cycle; C MT_m is the maintenance cost of the micro gas turbine, K MT is the maintenance factor of the micro gas turbine; Operation and maintenance cost of diesel generators C DE_om The expression is: C DE_om =C DE_o +C DE_m ; In the formula, C DE_o is the fuel cost of the diesel generator, P DE is the output power of the diesel generator, c1, c2, c3 are the combustion coefficients of the diesel generator; C DE_m The maintenance cost of diesel generators is K DE_m is the maintenance factor of the diesel generator, P DE (t) is the output power of the diesel generator at time t; When the grid connection coefficient a = 1, the cost of interacting with the large grid is C grid The expression is: Where P grid (t) is the power interacting with the large power grid at time t, G grid (t) is the electricity purchase price or electricity sales price at time t; The operation and maintenance cost of the photovoltaic unit C PV_om The expression is: In the formula, K PV_m is the maintenance factor of the photovoltaic power generation unit, P PV (t) is the output power of the photovoltaic power generation unit at time t; Operation and maintenance cost of wind turbines C WT_om The expression is: In the formula, K WT_m is the maintenance factor of the wind power unit, P WT (t) is the output power of the wind power generation unit at time t; The cost of the energy storage unit C BESS The expression is: Where P BESS is the output power of the energy storage unit, C p is the discounted value of the total investment cost, t BESS is the annual operation time of the energy storage system, K BESS is the investment recovery factor of the energy storage system; G BESS is the total investment cost of the energy storage system, δ is the depreciation rate of the energy storage unit, and n' is the design service life; Electric vehicle energy interaction cost C EV The expression is: Where n is the number of electric vehicles, P EVc,i (t), P EVd,i (t) are the charging power and discharging power of the ith electric vehicle at time t, Δt c,i and Δt d,i are the time intervals for charging and discharging of the i-th electric car at time t.

5. The AC / DC hybrid microgrid optimization scheduling method according to claim 3 is characterized in that: When the controllable distributed power source includes a micro gas turbine and a diesel generator, the expression of the environmental treatment cost C2 is: C2=C MT_P +C DT_P ; In the formula, C MT_P is the environmental treatment cost of the micro gas turbine, C DT_P Environmental treatment costs for diesel generators; Environmental treatment cost of micro gas turbine C MT_P The expression is: Where D y H y is the control cost coefficient of the main pollutant y, P MT (t) is the output power of the micro gas turbine at time t, Δt is the time interval, and M is the amount of the main pollutant; Environmental treatment cost of diesel generators C DT_P The expression is: Where P DE (t) is the output power of the diesel generator at time t.

6. The AC / DC hybrid microgrid optimization scheduling method according to claim 3 is characterized in that: When the controllable distributed power source includes a micro gas turbine and a diesel generator, and the uncontrollable distributed power source includes a photovoltaic unit and a wind unit, the constraints of the operation cost model include: power balance constraint, output constraint, climbing power constraint, grid-connected transmission power constraint, energy storage unit constraint and electric vehicle charge state constraint; The power balance constraint is: P L +P EV +P BESS =P MT +P DE +P Grid +P PV +P WT ; Where P L is the load power, P EV is the charging and discharging power of the electric vehicle, P BESS is the output power of the energy storage unit, P MT is the output power of the micro gas turbine, P DE is the output power of the diesel generator, P Grid is the power of interaction with the large power grid, P PV is the output power of the photovoltaic power generation unit, P WT is the output power of the wind power generation unit; Output constraints include output constraints of micro gas turbines, output constraints of diesel generators, output constraints of wind power generation, and output constraints of photovoltaic power generation; The output constraint of the micro gas turbine is expressed as: Where P MT (t) is the output power of the micro gas turbine at time t, and are the minimum and maximum output power of the micro gas turbine at time t respectively; The output constraint of the diesel generator is expressed as: Where P DE (t) is the output power of the diesel generator at time t, and are the minimum and maximum output power of the diesel generator at time t respectively; The output constraint of wind power generation is expressed as: Where P WT (t) is the output power of the wind power generation unit at time t, and are the minimum and maximum output power of the wind power generation unit at time t, respectively; The output constraint of photovoltaic power generation is expressed as: Where P PV (t) is the output power of the photovoltaic power generation unit at time t, and are the minimum and maximum output power of the photovoltaic power generation unit at time t respectively; The ramp power constraints include: the ramp power constraints of the micro gas turbine and the ramp power constraints of the diesel generator; The ramp power constraint of the micro gas turbine is: |P MT (t)-P MT (t-1)|≤r MT ; Where P MT (t-1) is the output power of the micro gas turbine at time t-1, r MT is the upper limit of the ramp rate of the micro gas turbine; The climbing power constraint of the diesel generator is: |P DE (t)-P DE (t-1)|≤r DT ; Where P DE (t-1) is the output power of the diesel generator at time t-1, r DT The upper limit of the ramp rate of the diesel generator; When the grid-connected coefficient a=1, the grid-connected transmission power constraint is: Where P grid (t) is the power interacting with the large power grid at time t, and are the minimum and maximum values ​​of the interaction power with the large power grid at time t, respectively; The energy storage unit constraints are: Where P BESS (t) is the output power of the energy storage unit at time t, and are the lower and upper limits of the output power of the energy storage unit at time t; SOC(t) is the state of charge of the energy storage unit at time t, and are the lower and upper limits of the energy storage capacity at time t respectively; The state of charge constraint of electric vehicles is: In the formula, SOC EV (t) is the state of charge of the electric vehicle at time t, and are the lower and upper limits of the battery capacity of the electric vehicle at time t respectively.

7. The AC / DC hybrid microgrid optimization scheduling method according to claim 1, characterized in that: The distribution function of the Tent chaotic map is: In the formula, Z k+1 is the k+1th population, Z k is the kth population, and β is the chaos parameter.

8. The AC / DC hybrid microgrid optimization scheduling method according to claim 1, characterized in that: The non-dominated sorting mechanism specifically includes: Input the initial population and multiple objective functions in advance; For each individual in the initial population, initialize the solution set dominated by each individual and the number of times it is dominated to 0; Compare any pair of individuals p and q in the initial population. If the evaluation results of individual p on all objective functions are not inferior to those of individual q, and the evaluation results of individual p on at least one objective function are better than those of individual q, then individual p dominates individual q, add individual q to the solution set of individual p, and increase the number of times individual q is dominated by 1. Put the individuals with a domination time of 0 into the first frontier, and assign the individuals in the first frontier a level of 1; Let the value of u be 2; Add the individuals with the dominance number of 1 in the solution set of each individual in the first frontier to the u-th frontier, and set the individual level in the u-th frontier to u; Determine whether all individuals in the initial population have been assigned ranks. If not, increase the value of u by 1 and return to step "add the individuals with 1 dominated times in the solution set of each individual in the first frontier to the u-th frontier, and set the rank of the individuals in the u-th frontier to u"; if so, obtain the rank of each individual in the initial population, sort them according to the rank, and output the sorted initial population.

9. The AC / DC hybrid microgrid optimization scheduling method according to claim 8, characterized in that: The non-dominated sorting mechanism further includes: For each objective function, the individuals in each front are sorted by objective value, and the crowding distance between the individual with the smallest objective value and the individual with the largest objective value in each front is set to infinity; For each objective function, according to the formula Calculate the crowding distance of each individual in the frontier except for the individual whose crowding distance is set to infinity; where d k' is the crowding distance of individual k' in each frontier, d k'0 is the initial crowding distance of individual k' in each frontier, f k'-1 (m) is the target value of the mth objective function for each individual k'-1 in the frontier, f k'+1 (m) is the target value of the mth objective function for each individual k'+1 in the frontier, f min (m) is the minimum target value of the mth target function, f max (m) is the maximum target value of the mth target function; Accumulate the crowding distances of all objective functions of each individual to obtain the total crowding distance of each individual, and output the population containing the total crowding distance; Individuals with low ranks are preferentially selected to generate new populations; if multiple individuals are at the same rank, the individuals with larger crowding distance are selected.

10. The AC / DC hybrid microgrid optimization scheduling method according to claim 1, characterized in that: The operation cost model is solved by using the dung beetle optimization algorithm, which specifically includes: Initialize the dung beetle population using Tent chaotic mapping to obtain the initial dung beetle population; The non-dominated sorting mechanism is used to sort the initial dung beetle population, and the sorted dung beetle population is obtained; The positions of the dung beetles in the sorted dung beetle population are updated to obtain a new dung beetle population; Update the fitness values ​​of dung beetles in the new dung beetle population to obtain the current global optimal solution; The non-dominated sorting mechanism is used to sort the new dung beetle population, and the sorted new dung beetle population is obtained; Adopt elite reverse learning method to optimize rolling behavior and reverse learning of elite individuals; Determine whether the maximum number of iterations has been reached. If not, replace the sorted dung beetle population with the sorted new dung beetle population, and return to step "update the positions of the dung beetles in the sorted dung beetle population to obtain a new dung beetle population"; if so, output the current global optimal solution.

Images