Electric power system economic dispatching optimization method and system based on improved dragonfly algorithm

Abstract

The invention particularly relates to a power system economic dispatching optimization method and system based on an improved dragonfly algorithm, and the method comprises the steps: determining a target function and a constraint condition; an improved dragonfly algorithm is adopted to optimize the power system economic dispatching model; relevant parameters in the dragonfly algorithm are initialized; iteratively calculating fitness values of dragonfly population individuals; weight parameters of the five behaviors in the dragonfly population are updated, the five behaviors in the dragonfly population are calculated, and the search radius is updated; judging whether other similar objects exist around the current dragonfly or not, and updating the step length vector and the position vector; judging whether the maximum number of iterations is reached or not, and if yes, stopping iteration and outputting an optimal solution; and if the maximum number of iterations is not reached, continuing the iterative operation. According to the method, the problem of economic dispatching of the power system under multi-target and multi-constraint conditions can be solved, and maximum economization of dispatching of the power system under wind power integration is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power system economic dispatch, and in particular to a power system economic dispatch optimization method and system based on an improved dragonfly algorithm. Background Art

[0002] The consumption of fossil energy and the resulting environmental pollution have attracted widespread attention to clean energy represented by wind energy. However, the integration of wind power and the demand for the development of low-carbon electricity have brought new challenges to the reliable operation and economic dispatch of the power system. Formulating a reasonable power system dispatch strategy under wind power grid connection is of great significance in maintaining the safe and stable operation of the power system, promoting the utilization rate of renewable energy, and improving the economic efficiency of power system operation.

[0003] When traditional optimization algorithms deal with power system scheduling problems, as the number of iterations increases, the amount of calculation gradually increases, and the efficiency is low. Intelligent optimization algorithms, with their global search capabilities and adaptability, can more effectively cope with the changing and complex optimization needs in power systems. However, the integration of wind power and the introduction of carbon trading have increased the complexity of power system scheduling plans, the number of decision variables has increased, and the objective function and constraints have become more complex. Some intelligent optimization algorithms are prone to falling into local optimal solutions and cannot find global optimal solutions.

[0004] Therefore, it is necessary to provide a new power system economic dispatch optimization method and system based on the improved dragonfly algorithm. Summary of the invention

[0005] Based on the above-mentioned problems existing in the prior art, the purpose of the embodiments of the present invention is to provide an optimization method and system for economic dispatch of an electric power system based on an improved dragonfly algorithm, which can handle the economic dispatch problem of an electric power system under multi-objective and multi-constraint conditions, and further achieve the maximum economy of electric power system dispatch under wind power grid connection under the premise of considering stability, environmental protection and low carbon.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is: a method for optimizing economic dispatch of a power system based on an improved dragonfly algorithm, comprising:

[0007] S1, determine the objective function and constraints of the power system economic dispatch model;

[0008] S2, taking the dragonfly algorithm as the benchmark algorithm, the improved dragonfly algorithm is used to optimize the economic dispatch model of the power system;

[0009] S3, initializing the relevant parameters in the dragonfly algorithm, including: population size, maximum number of iterations, individual dimension, upper and lower limits of individual motion position range, dragonfly population position vector and step vector;

[0010] S4, iteratively calculate the fitness values ​​of the dragonfly population individuals and update the positions of food and natural enemies in the population, that is, the two individuals with the best and worst fitness;

[0011] S5, update the weight parameters of the five behaviors in the dragonfly population, calculate the five behaviors in the dragonfly population, and update the search radius;

[0012] S6, determining whether there are other dragonflies of the same species around the current dragonfly. If there are other dragonflies of the same species around the current dragonfly, the step vector and position vector are updated by changing the grouping behavior weight and inertia weight to obtain a new step vector formula and position vector update formula for the dragonfly population; if there are no other dragonflies of the same species around the current dragonfly, the position vector is updated by using the calculation formula introduced into the Levy flight by the dynamic learning factor; and then the updated position vector is verified based on the constraint conditions;

[0013] S7, determine whether the maximum number of iterations has been reached, if so, stop the iteration and output the optimal solution; if not, return to step S4 to continue the iterative operation.

[0014] Furthermore, the objective function of determining the economic dispatch model of the power system is the sum of the power generation cost of the thermal power unit, the power generation cost of the wind power unit and the carbon trading cost.

[0015] Furthermore, the calculation formula for the power generation cost of the thermal power unit is:

[0016]

[0017] Among them, F G is the power generation cost of the thermal power unit, t is the dispatch time, T is a dispatch cycle, i is the group number of the thermal power unit, N is the total number of thermal power units in the system, P i,t is the output of the i-th thermal power unit at time t, parameter a i 、b i 、c i is the coal consumption coefficient of the i-th thermal power unit, e i and h i is the valve point effect coefficient of the i-th thermal power unit, g i E is the pollutant gas emitted per unit of power generation of thermal power units, i is the cost coefficient for treating polluted gases.

[0018] Furthermore, the constraints of the power system economic dispatch model include: system power balance constraints, thermal power unit output power constraints, wind power unit output power constraints, thermal power unit ramp rate constraints, reserve capacity constraints, thermal power unit start and stop constraints, thermal power unit carbon emission quota constraints and carbon emission right purchase quota constraints.

[0019] Furthermore, the system power balance constraint is:

[0020]

[0021] Among them, t is the scheduling time, T is a scheduling cycle, i is the group number of the thermal power unit, N is the total number of thermal power units in the system, P i,t is the output of the i-th thermal power unit at time t, M is the number of wind turbines connected to the grid, j is the group number of the wind turbine, P j,t is the output power of the jth wind turbine at the dispatching time t, P load,t is the total load of the system at time t.

[0022] Furthermore, the improved dragonfly algorithm is used to optimize the economic dispatch model of the power system, including:

[0023] Step S21, changing the weight a of the grouping behavior in the dragonfly algorithm;

[0024] Step S22, changing the inertia weight ω in the dragonfly algorithm;

[0025] Step S23, introducing dynamic learning factors into Levy flight.

[0026] Furthermore, in S6, by changing the grouping behavior weight and the inertia weight, a new step vector formula of the dragonfly population is obtained:

[0027] ΔX k+1 =(sS i +a′A i +cC i +fF i +eE i )+ω′ΔX k

[0028] Where ΔX k+1 is the modified k+1-time step vector, s is the weight coefficient of the separation behavior, S i is the separation behavior position vector of the i-th dragonfly individual, a' is the changed grouping behavior weight, A i is the grouping behavior vector of the i-th dragonfly individual, c is the weight coefficient of the grouping behavior, C i is the aggregation behavior position vector of the i-th dragonfly individual, f is the weight coefficient of foraging behavior, F i is the foraging behavior position vector of the i-th dragonfly individual, e is the weight coefficient of the avoidance behavior, E i is the position vector of the enemy avoidance behavior of the i-th dragonfly individual, ω' is the changed inertia weight, ΔX k is the step vector of k times, k is the current number of iterations;

[0029] Furthermore, in S6, by changing the grouping behavior weight and the inertia weight, the new position vector update formula of the dragonfly population is obtained as follows:

[0030] X k+1 =X k +ΔX k+1

[0031] Among them, X k+1 is the position of the individual at k+1 times, X k is the position of the individual at k times, ΔX k+1 is the step vector for k+1 times.

[0032] Further, in S6, the calculation formula for introducing the dynamic learning factor into the Levy flight is:

[0033]

[0034] Among them, η is the adaptive learning factor, λ is the relevant control parameter, k max is the maximum number of iterations, k is the current number of iterations;

[0035] The formula for updating the individual position of the dragonfly at this time is:

[0036] X k+1 =X k +ηLe′vy(d)×X k

[0037] Where η is the adaptive learning factor, Le′vy(x) is the Levy flight, and X k+1 is the position of the individual at k+1 times, X k is the position of the individual at k times.

[0038] The power system economic dispatch optimization system based on the improved dragonfly algorithm is applied to the above-mentioned power system economic dispatch optimization method based on the improved dragonfly algorithm, and the system includes:

[0039] A preprocessing module is used to determine the objective function and constraints of the power system economic dispatch model;

[0040] Improved modeling module, used to optimize the economic dispatch model of power system by using the improved dragonfly algorithm with the dragonfly algorithm as the benchmark algorithm;

[0041] Initialization module, used to initialize the relevant parameters in the dragonfly algorithm, including: population size, maximum number of iterations, individual dimension, upper and lower limits of individual motion position range, dragonfly population position vector and step vector;

[0042] Iterative calculation module, used to iteratively calculate the fitness value of dragonfly population individuals and update the positions of food and natural enemies in the population, that is, the two individuals with the best and worst fitness;

[0043] The behavior calculation module is used to update the weight parameters of the five behaviors in the dragonfly population, calculate the five behaviors in the dragonfly population, and update the search radius;

[0044] The position update module is used to determine whether there are other dragonflies of the same species around the current dragonfly. If there are other dragonflies of the same species around the current dragonfly, the step vector and position vector update formula of the dragonfly population are obtained by changing the grouping behavior weight and inertia weight to update the step vector and position vector; if there are no other dragonflies of the same species around the current dragonfly, the calculation formula introduced into the Levy flight by the dynamic learning factor is used to update the position vector; then the updated position vector is verified based on the constraint conditions;

[0045] The output judgment module is used to judge whether the maximum number of iterations has been reached. If so, the iteration is stopped and the optimal solution is output; if not, the iteration is returned to continue the iterative operation.

[0046] The beneficial effects of the present invention are as follows: determining the objective function and constraint conditions of the economic dispatch model of the power system; taking the dragonfly algorithm as the benchmark algorithm, adopting the improved dragonfly algorithm to optimize the economic dispatch model of the power system; initializing the relevant parameters in the dragonfly algorithm, the relevant parameters include: population size, maximum number of iterations, individual dimension, upper and lower limits of the individual motion position range, dragonfly population position vector and step vector; performing iterations, calculating the fitness values ​​of the dragonfly population individuals, updating the positions of food and natural enemies in the population, that is, the two individuals with the best and worst fitness; updating the weight parameters of the five behaviors in the dragonfly population, and calculating the five behaviors in the dragonfly population. Update the search radius; determine whether there are other dragonflies of the same species around the current dragonfly. If there are other dragonflies of the same species around the current dragonfly, obtain the new step vector formula and position vector update formula of the dragonfly population by changing the grouping behavior weight and inertia weight to update the step vector and position vector; if there are no other dragonflies of the same species around the current dragonfly, use the calculation formula introduced by the dynamic learning factor into the Levy flight to update the position vector; then verify the updated position vector based on the constraint conditions; determine whether the maximum number of iterations has been reached. If so, stop the iteration and output the optimal solution; if the maximum number of iterations has not been reached, return to continue the iterative operation. The power system economic dispatch optimization method based on the improved dragonfly algorithm of the present invention changes the weight of the fixed-value queuing behavior and the inertia weight into a decreasing function that gradually decreases with the number of iterations, which speeds up the convergence speed of the algorithm while reducing the risk of the algorithm falling into the local optimum, and is more conducive to finding the global optimal solution; considering that the step size is too large during Levy flight and the local optimum is easily ignored, an adaptive learning factor is introduced into the dragonfly individual position update formula during Levy flight, so that when the position is updated, the dragonfly can adaptively determine the moving distance according to the advantages and disadvantages of its own position, thereby improving the quality of the algorithm's global search. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0048] In the figure:

[0049] Figure 1 A flowchart of a method for optimizing economic dispatch of a power system based on an improved dragonfly algorithm provided in the first embodiment of the present invention;

[0050] Figure 2 A convergence comparison diagram of the original dragonfly algorithm and the improved dragonfly algorithm is provided for the first embodiment of the present invention;

[0051] Figure 3 A schematic diagram of the output of each unit after optimization using the improved dragonfly algorithm is provided for the first embodiment of the present invention;

[0052] Figure 4A schematic diagram of modules of a power system economic dispatch optimization system based on an improved dragonfly algorithm provided in the second embodiment of the present invention;

[0053] Figure 5 It is a structural diagram of a network-side server provided according to a third embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] First embodiment:

[0056] The first embodiment of the present invention provides an optimization method for economic dispatch of an electric power system based on an improved dragonfly algorithm, comprising: determining the objective function and constraint conditions of an economic dispatch model of an electric power system; taking the dragonfly algorithm as a benchmark algorithm, and using the improved dragonfly algorithm to optimize the economic dispatch model of an electric power system; initializing relevant parameters in the dragonfly algorithm, the relevant parameters including: population size, maximum number of iterations, individual dimension, upper and lower limits of individual motion position range, dragonfly population position vector, and step vector; performing iterations, calculating fitness values ​​of individuals in the dragonfly population, and updating the positions of food and natural enemies in the population, i.e., the two individuals with the best and worst fitness; updating weight parameters of five behaviors in the dragonfly population; and number, and calculate the five behaviors in the dragonfly population, and update the search radius; determine whether there are other dragonflies of the same kind around the current dragonfly. If there are other dragonflies of the same kind around the current dragonfly, by changing the grouping behavior weight and inertia weight, the new step vector formula and position vector update formula of the dragonfly population are obtained to update the step vector and position vector; if there are no other dragonflies of the same kind around the current dragonfly, the calculation formula introduced into the Levy flight by the dynamic learning factor is used to update the position vector; then the updated position vector is verified based on the constraint conditions; determine whether the maximum number of iterations has been reached. If so, stop the iteration and output the optimal solution; if the maximum number of iterations has not been reached, return to continue the iterative operation. The power system economic dispatch optimization method based on the improved dragonfly algorithm of the present invention changes the weight of the fixed-value queuing behavior and the inertia weight into a decreasing function that gradually decreases with the number of iterations, which speeds up the convergence speed of the algorithm while reducing the risk of the algorithm falling into the local optimum, and is more conducive to finding the global optimal solution; considering that the step size is too large during Levy flight and the local optimum is easily ignored, an adaptive learning factor is introduced into the dragonfly individual position update formula during Levy flight, so that when the position is updated, the dragonfly can adaptively determine the moving distance according to the advantages and disadvantages of its own position, thereby improving the quality of the algorithm's global search.

[0057] The following is a detailed description of the implementation details of the power system economic dispatch optimization method based on the improved dragonfly algorithm in this embodiment. The following content is only for the convenience of understanding the implementation details, which is not necessary for the implementation of this solution. The specific process of this embodiment is as follows: Figure 1 As shown, this implementation is applied to an economic dispatch optimization system of a power system based on an improved dragonfly algorithm.

[0058] Step S1, determining the objective function and constraints of the power system economic dispatch model.

[0059] Specifically, the objective function of determining the economic dispatch model of the power system is the sum of the power generation cost of the thermal power unit, the power generation cost of the wind power unit and the carbon trading cost.

[0060] Step S11, calculating the power generation cost of the thermal power unit in the objective function.

[0061] The power generation cost of thermal power units is generally the coal consumption cost of each unit during the dispatch cycle. The calculation formula for coal consumption cost is:

[0062]

[0063] Among them, F G is the coal consumption cost, t is the dispatch time, T is a dispatch cycle, i is the group number of the thermal power unit, N is the total number of thermal power units in the system, P i,t is the output of the i-th thermal power unit at time t; f(P i,t ) is the coal consumption cost function of thermal power units, parameter a i 、b i 、c i is the coal consumption coefficient of the i-th thermal power unit.

[0064] In actual operation, the wire drawing phenomenon that occurs when the intake valve of the steam turbine unit is suddenly opened during operation will superimpose a pulsation effect on the traditional fuel cost curve, namely the "valve point effect", which will significantly affect the fuel consumption rate and operating efficiency of the unit. The coal consumption cost function formula after considering the valve point effect is:

[0065]

[0066] Among them, f(P i,t ) is the coal consumption cost function after considering the valve point effect, P i,t is the output of the i-th thermal power unit at time t, parameter a i 、b i 、c i is the coal consumption coefficient of the i-th thermal power unit, e i and h iis the valve point effect coefficient of the i-th thermal power unit, is the lower limit of the output power of the i-th thermal power unit.

[0067] Thermal power units generate electricity by burning fossil fuels, and the emission of pollutants will pollute the environment. In addition to considering the cost of coal consumption and the valve point effect in actual operation, the cost of pollutant treatment should also be included in the cost of thermal power generation. In summary, the formula for calculating the cost of thermal power generation is:

[0068]

[0069] Among them, F G is the power generation cost of the thermal power unit, t is the dispatch time, T is a dispatch cycle, i is the group number of the thermal power unit, N is the total number of thermal power units in the system, P i,t is the output of the i-th thermal power unit at time t, parameter a i 、b i 、c i is the coal consumption coefficient of the i-th thermal power unit, e i and h i is the valve point effect coefficient of the i-th thermal power unit, g i E is the pollutant gas emitted per unit of power generation of thermal power units, i is the cost coefficient for treating polluted gases.

[0070] Step S12, calculating the power generation cost of the wind turbine in the objective function.

[0071] The operation and maintenance cost of a wind farm is linearly related to the output power of the wind farm. The calculation formula for the operation and maintenance cost of a wind turbine is:

[0072]

[0073] Among them, F w is the operation and maintenance cost of the wind turbine, t is the scheduling time, T is a scheduling cycle, M is the number of wind turbines connected to the grid, j is the group number of the wind turbine, C w is the power generation cost coefficient of the wind turbine, P j,t is the output power of the jth wind turbine at scheduling time t.

[0074] In practical applications, if the actual output power of a wind turbine is less than the predicted wind power, the grid reserve needs to be used to compensate for the lack of wind power, resulting in an overestimated cost of wind power. Conversely, if the actual output power of a wind turbine is greater than the predicted wind power, additional underestimated costs will be incurred. The formula for calculating the overestimated cost of wind power is as follows:

[0075]

[0076] Among them, F w(+) is the overestimation cost of wind power, t is the dispatch time, T is a dispatch cycle, M is the number of wind turbines connected to the grid, j is the group number of the wind turbine, C w(+) is the overestimation cost coefficient of wind power, P' j,t , P j,t is the output power of the jth wind turbine at the dispatching time t, P' j,t is the predicted value of the output power of the j-th wind turbine at the scheduling time t.

[0077] The underestimated cost of wind power can be calculated as follows:

[0078]

[0079] Among them, F w(-) is the underestimated cost of wind power, t is the dispatch time, T is a dispatch cycle, M is the number of wind turbines connected to the grid, j is the group number of the wind turbine, C w(-) is the underestimated cost coefficient of wind power, P j,t is the output power of the jth wind turbine at the dispatching time t, P' j,t is the predicted value of the output power of the j-th wind turbine at the scheduling time t.

[0080] In summary, the power generation cost of a wind turbine should be the sum of three parts. The power generation cost calculation formula of a wind turbine is:

[0081] F W =F w +F w(+) +F w(-)

[0082] Among them, F W is the power generation cost of the wind turbine, F w is the operation and maintenance cost of wind turbines, F w(+) is the overestimated cost of wind power, F w(-) The underestimated cost of wind power.

[0083] Step S13, calculating the carbon trading cost in the objective function.

[0084] The present invention adopts the baseline method as the initial allocation method of carbon quotas, and the carbon emissions calculation formula of thermal power units is:

[0085]

[0086] Among them, E Gp is the carbon emission of the thermal power unit, t is the scheduling time, T is a scheduling cycle, i is the group number of the thermal power unit, N is the total number of thermal power units in the system, δ iis the carbon emission intensity per unit of electricity of the i-th thermal power unit, P j,t is the output power of the jth wind turbine at scheduling time t, and ΔT is the scheduling time length.

[0087] The calculation formula for the carbon emission quota of thermal power units is:

[0088]

[0089] Among them, E Gq is the carbon emission quota of the thermal power unit, t is the scheduling time, T is a scheduling cycle, i is the group number of the thermal power unit, N is the total number of thermal power units in the system, η is the carbon emission benchmark per unit electricity, P j,t is the output power of the jth wind turbine at scheduling time t, and ΔT is the scheduling time length.

[0090] The formula for calculating carbon trading costs is:

[0091]

[0092] Among them, F g is the carbon trading cost, K C is the carbon trading price, E Gp is the carbon emission of thermal power units, E Gq It is the carbon emission quota of thermal power units.

[0093] Taking all the above factors into consideration, the objective function of the power system economic dispatch model in the present invention is the sum of the power generation cost of thermal power units, the power generation cost of wind power units and the carbon trading cost. The objective function is expressed as:

[0094] minF=F G +F W +F g

[0095] Among them, F G is the power generation cost of thermal power units, P W is the power generation cost of the wind turbine, F g is the carbon trading cost. The purpose of the power system economic dispatch model is to minimize the sum of the objective functions.

[0096] The constraints of the power system economic dispatch model include: system power balance constraints, thermal power unit output power constraints, wind power unit output power constraints, thermal power unit ramp rate constraints, reserve capacity constraints, thermal power unit start and stop constraints, thermal power unit carbon emission quota constraints and purchase carbon emission quota constraints.

[0097] The system power balance constraint is:

[0098]

[0099] Among them, t is the scheduling time, T is a scheduling cycle, i is the group number of the thermal power unit, N is the total number of thermal power units in the system, P i,t is the output of the i-th thermal power unit at time t, M is the number of wind turbines connected to the grid, j is the group number of the wind turbine, P j,t is the output power of the jth wind turbine at the dispatching time t, p load,t is the total load of the system at time t.

[0100] The output power constraint of thermal power units is:

[0101]

[0102] Among them, P i,tmin is the minimum power of the i-th thermal power unit, P i,t is the output of the i-th thermal power unit at time t, P i,tmax is the maximum power of the i-th thermal power unit, and N is the total number of thermal power units in the system.

[0103] The output power constraint of the wind turbine is:

[0104] 0≤P j,t ≤P j,tmax ,(j=1,2,…,M)

[0105] Among them, P j,t is the output power of the jth wind turbine at the dispatching time t, p j,tmax is the maximum power of the jth wind turbine, and M is the number of wind turbines connected to the grid.

[0106] In the actual operation of the power grid, the commissioning and shut-down of thermal power units need to be completed in a relatively short period of time, so there are higher requirements for the ramp rate of conventional thermal power units. The ramp rate constraint of thermal power units is:

[0107] P i,t -P i,t-1 ≤r iu ×T 60

[0108] P i,t-1 -P i,t ≤r id ×T 60

[0109] Among them, P i,t is the output of the i-th thermal power unit at time t, P i,t-1 is the output power of the i-th thermal power unit at time t-1, r iu is the maximum rate of increase of the output power of the i-th thermal power unit, T 60 is the scheduling time period, usually 1h, rid is the maximum decrease rate of the output power of the i-th thermal power unit.

[0110] The reserve constraint is:

[0111]

[0112] Where i is the group number of the thermal power unit, N is the total number of thermal power units in the system, and d i is the reserve capacity factor of thermal power units, is the maximum output of the i-th thermal power unit, is the minimum output of the i-th thermal power unit, P i,t is the output of the i-th thermal power unit at time t, P j,t is the output power of the jth wind turbine at the dispatching time t, S t,up is the positive spare capacity, S t,down Negative spare capacity.

[0113] The start and stop constraints of thermal power units are:

[0114]

[0115] in, is the cumulative startup time of the i-th thermal power unit at the scheduling time t, is the minimum continuous on time, is the cumulative shutdown time of the i-th thermal power unit at the scheduling time t, Minimum continuous downtime.

[0116] The carbon emission quota constraints of thermal power units and the purchase of carbon emission quota constraints are:

[0117]

[0118] E Gp -E Gq ≤E qH

[0119] Among them, E Gq The carbon emission quota for thermal power units. is the maximum carbon emission quota, E Gp is the carbon emission of thermal power units, E qH The maximum carbon emission quota that can be purchased.

[0120] Step S2, taking the dragonfly algorithm as the benchmark algorithm, an improved dragonfly algorithm is used to optimize the economic dispatch model of the power system.

[0121] Specifically, the existing dragonfly algorithm is a new type of swarm intelligence algorithm developed by simulating the behavior of dragonfly groups in nature. The algorithm draws on the five behaviors of dragonfly populations: separation behavior, grouping behavior, aggregation behavior, predation behavior, and enemy avoidance behavior. The five behaviors of dragonfly populations can represent the movement of dragonflies in the population and can be used to update the position of individual dragonflies during the algorithm's search and optimization process.

[0122] The separation behavior formula is:

[0123]

[0124] Among them, S i is the separation behavior position vector of the i-th dragonfly individual, X is the current position of the dragonfly, and X j is the position of the jth dragonfly around the current dragonfly, and N is the number of dragonflies around it.

[0125] The formula for group behavior is:

[0126]

[0127] Among them, A i is the grouping behavior vector of the i-th dragonfly individual, V j is the flying speed of the jth dragonfly individual in the surrounding area, and N is the number of dragonflies in the surrounding area.

[0128] The aggregation behavior formula is:

[0129]

[0130] Among them, C i is the aggregation behavior position vector of the i-th dragonfly individual, X j is the position of the jth dragonfly around the current dragonfly, X is the position of the current individual, and N is the number of dragonflies around it.

[0131] The formula for predation behavior is:

[0132] F i =X + -X

[0133] Among them, F i is the foraging behavior position vector of the i-th dragonfly individual, X is the current individual location, and X + is the location of the food; that is, the location of the individual with the best fitness value in the current population.

[0134] The formula for avoiding enemy behavior is:

[0135]

[0136] Among them, E iis the position vector of the enemy avoidance behavior of the i-th dragonfly individual, X is the current individual position; X - is the location of the natural enemy; that is, the location of the individual with the worst fitness value in the current population.

[0137] In order to update the position of individual dragonflies in real time in the search space, the step vector ΔX and the position vector X are introduced for updating the position of individual dragonflies. The position vector represents the relative position of the dragonfly in the population, and the step vector represents the distance moved by the individual to the next iteration. The step vector formula is as follows:

[0138] ΔX k+1 =(sS i +aA i +cC i +fF i +eE i )+ωΔX k

[0139] Where ΔX k+1 is the k+1th step vector, s is the weight coefficient of the separation behavior, S i is the position vector of the separation behavior of the i-th dragonfly individual, a is the weight coefficient of the grouping behavior, A i is the grouping behavior vector of the i-th dragonfly individual, c is the weight coefficient of the grouping behavior, C i is the aggregation behavior position vector of the i-th dragonfly individual, f is the weight coefficient of foraging behavior, F i is the foraging behavior position vector of the i-th dragonfly individual, e is the weight coefficient of the avoidance behavior, E i is the position vector of the ith dragonfly’s avoidance behavior, ω is the inertia weight, ΔX k is the step vector of k times, where k is the current number of iterations.

[0140] The update formula of the position vector is:

[0141] X k+1 =X k +ΔX k+1

[0142] Among them, X k+1 is the position of the individual at k+1 times, X k is the position of the individual at k times, ΔX k+1 is the step vector for k+1 times.

[0143] When there are no other dragonflies around the individual, the above formulas are no longer applicable. The random walk of Levy flight mixes short-distance search and long-distance search, which can increase the diversity and randomness of the population and expand the search range. After adopting the Levy flight strategy, the dragonfly position update formula is:

[0144] X k+1 =X k +Le′vy(d)×X k

[0145]

[0146]

[0147] Among them, k is the current iteration number, X k+1 is the position of the individual at k+1 times, X k is the position of the individual k times, d is the dimension of the position vector, r1 and r2 are two random numbers in [0,1], β is a constant, β=1.5 in the present invention, and Le′vy(x) is the Levy flight.

[0148] Although the dragonfly algorithm has excellent global search capabilities, it also has the disadvantages of slow convergence speed and low convergence accuracy when solving some complex optimization problems. When the dragonfly individual updates its position, the team weight a that affects the individual dragonfly's movement speed and the inertia weight ω that determines the step vector are both fixed values. The size of the weight will affect the strength of the dragonfly algorithm's global or local search capabilities. At the same time, when the dragonfly algorithm adopts the Levy flight strategy, the individual movement step is random. When the step is too large, it is not conducive to the algorithm's local optimization. Therefore, it is necessary to improve the original dragonfly algorithm.

[0149] The improved dragonfly algorithm is used to optimize the economic dispatch model of the power system, including:

[0150] Step S21, changing the weight a of the grouping behavior in the dragonfly algorithm.

[0151] Specifically, compared with other optimization problems, the optimization scheduling of power systems has a larger algorithm search space and a more complex objective function. The dragonfly algorithm will show a slower convergence speed in the initial stage, which will cause the algorithm to easily fall into the local optimal solution. As the number of iterations increases, when the dragonfly algorithm converges too fast, it is not conducive to the local optimization of the algorithm. Therefore, in the present invention, the teaming behavior weight a is changed to a weight value that gradually decreases with the number of iterations. The changed teaming behavior weight a' is:

[0152]

[0153] Among them, a' is the changed teaming behavior weight, a s is the initial weight introduced, a e is the introduced termination weight, k max is the maximum number of iterations, and k is the current number of iterations.

[0154] Step S22, changing the inertia weight ω in the dragonfly algorithm.

[0155] The inertia weight ω determines the moving step size of the individual during iteration, and its change is more sensitive to the algorithm's search accuracy. When the weight is large, the algorithm has a stronger global search capability but a weaker local search capability; when the inertia weight is small, the local search capability will be improved and the global search will be relatively weak. In the present invention, the fixed value inertia weight ω is changed to a nonlinear decreasing function that gradually decreases with the number of iterations, so that the dragonfly algorithm can search for the global optimal value. The changed inertia weight ω' is:

[0156]

[0157] Among them, ω' is the changed inertia weight, k max is the maximum number of iterations, k is the current number of iterations, ω max is the initial inertia weight value, ω min is the final inertia weight value.

[0158] According to the changed grouping behavior weight and inertia weight, the new step vector formula of the dragonfly population is:

[0159] ΔX k+1 =(sS i +a′A i +cC i +fF i +eE i )+ω′ΔX k

[0160] Where ΔX k+1 is the modified k+1-time step vector, s is the weight coefficient of the separation behavior, S i is the separation behavior position vector of the i-th dragonfly individual, a' is the changed grouping behavior weight, A i is the grouping behavior vector of the i-th dragonfly individual, c is the weight coefficient of the grouping behavior, C i is the aggregation behavior position vector of the i-th dragonfly individual, f is the weight coefficient of foraging behavior, F i is the foraging behavior position vector of the i-th dragonfly individual, e is the weight coefficient of the avoidance behavior, E i is the position vector of the enemy avoidance behavior of the i-th dragonfly individual, ω' is the changed inertia weight, ΔX k is the step vector of k times, where k is the current number of iterations.

[0161] Step S23, introducing dynamic learning factors into Levy flight.

[0162] When there are no dragonflies of the same species around the individual, the global optimization is performed by Lévy flight. However, the moving step length of Lévy flight is randomly determined. When the step length is large, this will cause the algorithm to easily ignore the local optimal value. To address this problem, the present invention introduces the adaptive learning factor η into the Lévy flight, and the calculation formula is as follows:

[0163] η=2 λ-1

[0164]

[0165] Among them, η is the adaptive learning factor, λ is the relevant control parameter, k max is the maximum number of iterations, and k is the current number of iterations.

[0166] The formula for updating the individual position of the dragonfly at this time is:

[0167] X k+1 =X k +ηLe′vy(d)×X k

[0168] Where η is the adaptive learning factor, Le′vy(x) is the Levy flight, and X k+1 is the position of the individual at k+1 times, X k is the position of the individual at k times.

[0169] Due to the addition of the learning factor, the moving step size during Levy flight can change adaptively with the number of iterations. The larger step size in the early stage is conducive to the algorithm to quickly jump out of the local optimal solution. In the later stage of the algorithm, the search is performed with a smaller step size, which improves the global search quality.

[0170] Step S3, initializing relevant parameters in the dragonfly algorithm, including: population size, maximum number of iterations, individual dimension, upper and lower limits of individual motion position range, dragonfly population position vector and step vector.

[0171] Specifically, the output P of the thermal power unit i,t As a decision variable, first set the initial dragonfly position to:

[0172]

[0173] Among them, N p is the population size, T is the scheduling period, the dragonfly individual dimension is N×T, x i is a scheduling scheme in x.

[0174] The output of thermal power units at each moment is randomly generated by the calculation formula, which is:

[0175] P i,t =Pi,tmin +r×(P i,tmax -P i,tmin )

[0176] Among them, P i,t is the output of the i-th thermal power unit at time t, P i,tmin is the minimum power of the i-th thermal power unit, P i,tmax is the maximum power of the i-th thermal power unit, and r is a random number between the interval [0,1].

[0177] The upper and lower limits of the individual motion position range are the constraints in step S1.

[0178] Step S4, iteratively calculate the fitness values ​​of the dragonfly population individuals, and update the positions of food and natural enemies in the population, that is, the two individuals with the best and worst fitness.

[0179] Specifically, the fitness value is a key indicator to measure the quality of an individual, which reflects the degree of fit of the solution represented by the individual with the goals in the economic dispatch problem of the power system, such as cost minimization and benefit maximization.

[0180] Step S5, updating the weight parameters of the five behaviors in the dragonfly population, calculating the five behaviors in the dragonfly population, and updating the search radius.

[0181] Specifically, different behaviors collectively affect the movement direction and method of individuals in the population; at the same time, the search radius is updated. The search radius determines how large a range of space an individual can explore around it. Appropriate radius adjustment helps to find a balance between global search and local search, avoiding falling into the local optimal solution too early or the problem of low search efficiency.

[0182] Step S6, determine whether there are other dragonflies of the same species around the current dragonfly. If there are other dragonflies of the same species around the current dragonfly, the step vector and position vector are updated by changing the grouping behavior weight and inertia weight to obtain a new step vector formula and position vector update formula for the dragonfly population; if there are no other dragonflies of the same species around the current dragonfly, the position vector is updated by using the calculation formula introduced into the Levy flight using the dynamic learning factor; and then the updated position vector is verified based on the constraint conditions.

[0183] Furthermore, by changing the grouping behavior weight and inertia weight, the new step vector formula of the dragonfly population is obtained as follows:

[0184] ΔX k+1 =(sS i +a′A i +cC i +fF i +eE i )+ω′ΔX k

[0185] Where ΔX k+1 is the modified k+1-time step vector, s is the weight coefficient of the separation behavior, S i is the separation behavior position vector of the i-th dragonfly individual, a' is the changed grouping behavior weight, A i is the grouping behavior vector of the i-th dragonfly individual, c is the weight coefficient of the grouping behavior, C i is the aggregation behavior position vector of the i-th dragonfly individual, f is the weight coefficient of foraging behavior, F i is the foraging behavior position vector of the i-th dragonfly individual, e is the weight coefficient of the avoidance behavior, E i is the position vector of the enemy avoidance behavior of the i-th dragonfly individual, ω' is the changed inertia weight, ΔX k is the step vector of k times, where k is the current number of iterations.

[0186] Furthermore, by changing the grouping behavior weight and inertia weight, the new position vector update formula of the dragonfly population is obtained as follows:

[0187] X k+1 =X k +ΔX k+1

[0188] Among them, X k+1 is the position of the individual at k+1 times, X k is the position of the individual at k times, ΔX k+1 is the step vector for k+1 times.

[0189] Furthermore, the calculation formula introduced into Levy flight using the dynamic learning factor is:

[0190] η=2 λ-1

[0191]

[0192] Among them, η is the adaptive learning factor, λ is the relevant control parameter, k max is the maximum number of iterations, and k is the current number of iterations.

[0193] The formula for updating the individual position of the dragonfly at this time is:

[0194] X k+1 =X k +ηLe′vy(d)×X k

[0195] Where η is the adaptive learning factor, Le′vy(x) is the Levy flight, and X k+1 is the position of the individual at k+1 times, X kis the position of the individual at k times.

[0196] Step S7, determine whether the maximum number of iterations has been reached, if so, stop the iteration and output the optimal solution; if not, return to step S4 to continue the iterative operation.

[0197] As an example, all experiments in the present invention are performed on a personal computer with a processor of 12th Gen Intel(R) Core(TM) i9-12900KF, a graphics card model of NVIDIA GeForce RTX 3080Ti GPU, and a random access memory (RAM) of 32.0GB. The embodiments of the present invention use a standard IEEE-39 node standard test system to perform simulation experiments in MATLAB 2022b, and the test cycle is one day, divided into 24 hours. This system contains a total of 10 thermal power units and a wind farm, and the wind farm is incorporated into the system through 8 nodes.

[0198] In this embodiment, the initial parameters of the original dragonfly algorithm are set as follows: the population size is set to 50; the maximum number of iterations is 100; s is set to 0.1; a is set to 0.1; c is set to 0.7; f is set to 1; e is set to 1; and ω is set to 0.7. s Set to 0.9; a e Set the value of ω to 0.3; max The value of is set to 1.2; ω min The value of is 0.5; other parameters remain unchanged.

[0199] In order to verify the effectiveness and superiority of the improved dragonfly optimization algorithm and the original dragonfly optimization algorithm in optimizing the economic dispatch of power systems, this patent conducted simulation experiments based on the original dragonfly optimization algorithm and the improved dragonfly optimization algorithm. The convergence curves of the two algorithms are shown in Figure 2. Figure 2 shown.

[0200] The optimal system operating cost was obtained when the dragonfly optimization algorithm before improvement was iterated 50 times, while the optimal system operating cost was obtained when the improved dragonfly algorithm was iterated 30 times. This shows that the convergence speed of the algorithm can be improved by changing the weight of the grouping behavior. At the same time, the dragonfly algorithm before improvement is more likely to fall into the local optimum during the optimization process than the improved dragonfly algorithm. This shows that by changing the inertia weight and introducing the dynamic learning factor in Levy flight, the algorithm can avoid falling into the local optimum while fully optimizing globally. Table 1 below shows the system operating cost after optimizing the power system dispatching strategy using the dragonfly algorithm before and after improvement. In one dispatching cycle, the system operating cost optimized by the improved dragonfly algorithm is $1,370,342, which is $6,721 lower than that of the dragonfly algorithm before improvement. This shows that the improved dragonfly algorithm can better improve the economy of the power system. Figure 3 The figure shows the output of each unit after optimization using the improved dragonfly algorithm. By reasonably allocating the output of thermal power units and wind power units, the emission of carbon dioxide is reduced as much as possible, and the economy of the power system under wind power grid connection is maximized.

[0201] Table 1 Comparison of algorithm optimization scheduling results before and after improvement

[0202]

[0203] The first embodiment of the present invention provides an economic dispatch optimization method for a power system based on an improved dragonfly algorithm, comprising: determining the objective function and constraints of the economic dispatch model of the power system; taking the dragonfly algorithm as a benchmark algorithm, and using the improved dragonfly algorithm to optimize the economic dispatch model of the power system; initializing relevant parameters in the dragonfly algorithm, the relevant parameters including: population size, maximum number of iterations, individual dimensions, upper and lower limits of the individual motion position range, dragonfly population position vector, and step vector; performing iterations, calculating the fitness values ​​of individuals in the dragonfly population, and updating the positions of food and natural enemies in the population, that is, the two individuals with the best and worst fitness; updating the weight parameters of the five behaviors in the dragonfly population; number, and calculate five behaviors in the dragonfly population, update the search radius; determine whether there are other dragonflies of the same kind around the current dragonfly. If there are other dragonflies of the same kind around the current dragonfly, by changing the grouping behavior weight and inertia weight, obtain the new step vector formula and position vector update formula of the dragonfly population to update the step vector and position vector; if there are no other dragonflies of the same kind around the current dragonfly, use the calculation formula introduced by the dynamic learning factor into the Levy flight to update the position vector; then verify the updated position vector based on the constraint conditions; determine whether the maximum number of iterations has been reached, if so, stop the iteration and output the optimal solution; if the maximum number of iterations has not been reached, return to continue the iterative operation. The present invention not only takes into account the power generation cost of thermal power units, but also includes the power generation cost of wind power units and carbon trading costs. In order to ensure the safe and stable operation of the power system, the constraints are fully set, including system power balance constraints, thermal power unit output power constraints, wind power unit output power constraints, thermal power unit ramp constraints, thermal power unit start and stop constraints, reserve constraints, thermal power unit carbon emission quota constraints and purchase carbon emission quota constraints; based on the stability and low carbon nature of the power system, a reasonable power system economic dispatch model under wind power grid connection is formulated. By changing the weight of the teaming behavior, the problem of slow convergence speed and easy to fall into local optimality when dealing with complex problems is overcome by the dragonfly algorithm; the fixed value inertia weight is changed to a nonlinear decreasing function that decreases with the number of iterations, so as to facilitate the global optimization of the dragonfly algorithm. The learning factor is introduced into the Levy flight to solve the problem of too large step size and ignoring local optimization during Levy flight; the improved dragonfly algorithm is used to solve the established dispatch model, and the feasibility of the method is verified by examples; by comparing with the algorithm before improvement, the experimental results show that the improved dragonfly algorithm of the present invention has better optimization performance and improves the economy of the dispatch scheme to a certain extent.

[0204] The step division of the above methods is only for the purpose of clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this patent; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the scope of protection of this patent.

[0205] Second implementation method:

[0206] like Figure 4 As shown, the second embodiment of the present invention provides an economic dispatch optimization system for an electric power system based on an improved dragonfly algorithm, the system comprising: a preprocessing module 201, an improved modeling module 202, an initialization module 203, an iterative calculation module 204, a behavior calculation module 205, a position update module 206, and a judgment output module 207.

[0207] Specifically, the preprocessing module 201 is used to determine the objective function and constraints of the power system economic dispatch model; the improved modeling module 202 is used to use the dragonfly algorithm as the benchmark algorithm and adopt the improved dragonfly algorithm to optimize the power system economic dispatch model; the initialization module 203 is used to initialize the relevant parameters in the dragonfly algorithm, and the relevant parameters include: population size, maximum number of iterations, individual dimension, upper and lower limits of individual motion position range, dragonfly population position vector and step vector; the iterative calculation module 204 is used to iteratively calculate the fitness value of the dragonfly population individuals, update the position of food and natural enemies in the population, that is, the two individuals with the best and worst fitness; the behavior calculation module 205 is used to update the weight parameters of the five behaviors in the dragonfly population , and calculate the five behaviors in the dragonfly population, and update the search radius; the position update module 206 is used to determine whether there are other dragonflies of the same kind around the current dragonfly. If there are other dragonflies of the same kind around the current dragonfly, the step vector formula and position vector update formula of the dragonfly population are obtained by changing the grouping behavior weight and the inertia weight to update the step vector and the position vector; if there are no other dragonflies of the same kind around the current dragonfly, the calculation formula introduced into the Levy flight by the dynamic learning factor is used to update the position vector; then the updated position vector is verified based on the constraint conditions; the judgment output module 207 is used to judge whether the maximum number of iterations has been reached. If so, the iteration is stopped and the optimal solution is output; if the maximum number of iterations has not been reached, it returns to continue the iterative operation.

[0208] It is not difficult to find that this embodiment is a system embodiment corresponding to the first embodiment, and this embodiment can be implemented in conjunction with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the first embodiment.

[0209] It is worth mentioning that all modules involved in this embodiment are logic modules. In practical applications, a logic unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, this embodiment does not introduce units that are not closely related to solving the technical problem proposed by the present invention, but this does not mean that there are no other units in this embodiment.

[0210] A third embodiment of the present invention relates to a network side server, such as Figure 5 As shown, it includes at least one processor 302; and a memory 301 that is communicatively connected to the at least one processor 302; wherein the memory 301 stores instructions that can be executed by the at least one processor 302, and the instructions are executed by the at least one processor 302 so that the at least one processor 302 can execute the above-mentioned data processing method.

[0211] The memory 301 and the processor 302 are connected in a bus manner, and the bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors 302 and the memory 301 together. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. The data processed by the processor 302 is transmitted on a wireless medium via an antenna, and further, the antenna also receives data and transmits the data to the processor 302.

[0212] The processor 302 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management and other control functions. The memory 301 can be used to store data used by the processor 302 when performing operations.

[0213] The fourth embodiment of the present invention relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the power system economic dispatch optimization method based on the improved dragonfly algorithm in the first embodiment is implemented.

[0214] That is, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0215] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can know all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

[0216] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing economic dispatch of power system based on improved dragonfly algorithm, characterized in that: include: S1, determine the objective function and constraints of the power system economic dispatch model; S2, taking the dragonfly algorithm as the benchmark algorithm, the improved dragonfly algorithm is used to optimize the economic dispatch model of the power system; S3, initializing the relevant parameters in the dragonfly algorithm, including: population size, maximum number of iterations, individual dimension, upper and lower limits of individual motion position range, dragonfly population position vector and step vector; S4, iteratively calculate the fitness values ​​of the dragonfly population individuals and update the positions of food and natural enemies in the population, that is, the two individuals with the best and worst fitness; S5, update the weight parameters of the five behaviors in the dragonfly population, calculate the five behaviors in the dragonfly population, and update the search radius; S6, determining whether there are other dragonflies of the same species around the current dragonfly. If there are other dragonflies of the same species around the current dragonfly, the step vector and position vector are updated by changing the grouping behavior weight and inertia weight to obtain a new step vector formula and position vector update formula for the dragonfly population; if there are no other dragonflies of the same species around the current dragonfly, the position vector is updated by using the calculation formula introduced into the Levy flight by the dynamic learning factor; and then the updated position vector is verified based on the constraint conditions; S7, determine whether the maximum number of iterations has been reached, if so, stop the iteration and output the optimal solution; if not, return to step S4 to continue the iterative operation.

2. The power system economic dispatch optimization method based on the improved dragonfly algorithm according to claim 1 is characterized in that: The objective function of determining the power system economic dispatch model is the sum of the power generation cost of the thermal power unit, the power generation cost of the wind power unit and the carbon trading cost.

3. The power system economic dispatch optimization method based on the improved dragonfly algorithm according to claim 2 is characterized in that: The calculation formula for the power generation cost of the thermal power unit is: Among them, F G is the power generation cost of the thermal power unit, t is the dispatch time, T is a dispatch cycle, i is the group number of the thermal power unit, N is the total number of thermal power units in the system, P i,t is the output of the i-th thermal power unit at time t, parameter a i 、b i 、c i is the coal consumption coefficient of the i-th thermal power unit, e i and h i is the valve point effect coefficient of the i-th thermal power unit, g i E is the pollutant gas emitted per unit of power generation of thermal power units, i is the cost coefficient for treating polluted gases.

4. The power system economic dispatch optimization method based on the improved dragonfly algorithm according to claim 1 is characterized in that: The constraints of the power system economic dispatch model include: system power balance constraints, thermal power unit output power constraints, wind power unit output power constraints, thermal power unit ramp rate constraints, reserve capacity constraints, thermal power unit start and stop constraints, thermal power unit carbon emission quota constraints and carbon emission right purchase quota constraints.

5. The power system economic dispatch optimization method based on the improved dragonfly algorithm according to claim 4 is characterized in that: The system power balance constraint is: Among them, t is the scheduling time, T is a scheduling cycle, i is the group number of the thermal power unit, N is the total number of thermal power units in the system, P i,t is the output of the i-th thermal power unit at time t, M is the number of wind turbines connected to the grid, j is the group number of the wind turbine, P j,t is the output power of the jth wind turbine at the dispatching time t, P load,t is the total load of the system at time t.

6. The power system economic dispatch optimization method based on the improved dragonfly algorithm according to claim 1 is characterized in that: The improved dragonfly algorithm is used to optimize the economic dispatch model of the power system, including: Step S21, changing the weight a of the grouping behavior in the dragonfly algorithm; Step S22, changing the inertia weight ω in the dragonfly algorithm; Step S23, introducing dynamic learning factors into Levy flight.

7. The power system economic dispatch optimization method based on the improved dragonfly algorithm according to claim 1 is characterized in that: In S6, by changing the grouping behavior weight and inertia weight, the new step vector formula of the dragonfly population is obtained as follows: ΔX k+1 =(sS i +a′A i +cC i +fF i +eE i )+ω′ΔX k Where ΔX k+1 is the modified k+1-time step vector, s is the weight coefficient of the separation behavior, S i is the separation behavior position vector of the i-th dragonfly individual, a' is the changed grouping behavior weight, A i is the grouping behavior vector of the i-th dragonfly individual, c is the weight coefficient of the grouping behavior, C i is the aggregation behavior position vector of the i-th dragonfly individual, f is the weight coefficient of foraging behavior, F i is the foraging behavior position vector of the i-th dragonfly individual, e is the weight coefficient of the avoidance behavior, E i is the position vector of the enemy avoidance behavior of the i-th dragonfly individual, ω' is the changed inertia weight, ΔX k is the step vector of k times, where k is the current number of iterations.

8. The power system economic dispatch optimization method based on the improved dragonfly algorithm according to claim 1 is characterized in that: In S6, the new position vector update formula of the dragonfly population is obtained by changing the grouping behavior weight and the inertia weight: X k+1 =X k +ΔX k+1 Among them, X k+1 is the position of the individual at k+1 times, X k is the position of the individual at k times, ΔX k+1 is the step vector for k+1 times.

9. The power system economic dispatch optimization method based on the improved dragonfly algorithm according to claim 1 is characterized in that: In S6, the calculation formula introduced into the Levy flight using the dynamic learning factor is: n=2 λ-1 Among them, η is the adaptive learning factor, λ is the relevant control parameter, k max is the maximum number of iterations, k is the current number of iterations; The formula for updating the individual position of the dragonfly at this time is: X k+1 =X k +ηLe′vy(d)×X k Where η is the adaptive learning factor, Le′vy(x) is the Levy flight, and X k+1 is the position of the individual at k+1 times, X k is the position of the individual at k times.

10. The power system economic dispatch optimization system based on the improved dragonfly algorithm is characterized by: Applied to the power system economic dispatch optimization method based on the improved dragonfly algorithm described in claim 1, the system comprises: A preprocessing module is used to determine the objective function and constraints of the power system economic dispatch model; Improved modeling module, used to optimize the economic dispatch model of power system by using the improved dragonfly algorithm with the dragonfly algorithm as the benchmark algorithm; Initialization module, used to initialize the relevant parameters in the dragonfly algorithm, including: population size, maximum number of iterations, individual dimension, upper and lower limits of individual motion position range, dragonfly population position vector and step vector; Iterative calculation module, used to iteratively calculate the fitness value of dragonfly population individuals and update the positions of food and natural enemies in the population, that is, the two individuals with the best and worst fitness; The behavior calculation module is used to update the weight parameters of the five behaviors in the dragonfly population, calculate the five behaviors in the dragonfly population, and update the search radius; The position update module is used to determine whether there are other dragonflies of the same species around the current dragonfly. If there are other dragonflies of the same species around the current dragonfly, the step vector and position vector update formula of the dragonfly population are obtained by changing the grouping behavior weight and inertia weight to update the step vector and position vector; if there are no other dragonflies of the same species around the current dragonfly, the calculation formula introduced into the Levy flight by the dynamic learning factor is used to update the position vector; then the updated position vector is verified based on the constraint conditions; The output judgment module is used to judge whether the maximum number of iterations has been reached. If so, the iteration is stopped and the optimal solution is output; if not, the iteration is returned to continue the iterative operation.

Images