CN116031935A - Grid-connected multi-microgrid system operation scheduling optimization method considering electric energy interaction and demand response - Google Patents
Grid-connected multi-microgrid system operation scheduling optimization method considering electric energy interaction and demand response Download PDFInfo
- Publication number
- CN116031935A CN116031935A CN202310048537.4A CN202310048537A CN116031935A CN 116031935 A CN116031935 A CN 116031935A CN 202310048537 A CN202310048537 A CN 202310048537A CN 116031935 A CN116031935 A CN 116031935A
- Authority
- CN
- China
- Prior art keywords
- grid
- micro
- power
- load
- demand
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides a micro-grid scheduling method considering electric energy interaction and demand response, which comprises the steps of firstly, constructing a grid-connected multi-micro-grid system structure model containing wind power and photovoltaic, and obtaining an optimized objective function with minimum total running cost of the system as a target according to equipment characteristics; establishing a grid-connected multi-micro-grid system dispatching optimization model according to the upper limit and the lower limit of the constraint condition; secondly, solving the optimization problem of the total operation cost through an improved marine predator algorithm; then, in the dispatching operation of the multi-micro-grid system, a system operation strategy considering the electric energy interaction and load demand response in the system is designed, and the purposes of peak clipping and valley filling are effectively realized; finally, through specific formula analysis, the rationality and the effectiveness of the provided scheduling optimization method are verified. The invention can effectively realize the running cost of the multi-micro-grid system and the corresponding perfection of the energy flow of each micro-grid, thereby realizing the economy and reliability of the operation of the micro-grid system.
Description
Technical Field
The technical scheme of the invention belongs to the technical field of power systems, and particularly relates to a micro-grid scheduling method considering electric energy interaction and demand response.
Background
The grid-connected multi-microgrid system operation scheduling method considering electric energy interaction and demand response is characterized in that in the multi-microgrid system, the total operation cost of the system is optimized under various constraint conditions, and an optimal scheduling scheme is finally obtained, so that stable operation of the power grid system is realized.
Along with the rapid development of the economy and the acceleration of the industrial process in China, the energy consumption becomes the primary problem of the industrial development in China. Therefore, new energy power generation modes such as wind power generation, photovoltaic power generation and the like are developed greatly in China, so that a large amount of renewable energy sources are connected into a power grid, and various output characteristics of a main power grid can be influenced by renewable energy sources such as wind power, photovoltaic power generation and the like. In order to improve the electric energy utilization rate of renewable energy sources such as wind power, photovoltaic and the like, a grid-connected multi-micro-grid system is proposed as a new grid mode.
In the multi-microgrid system, each independent microgrid system is kept in close contact, and can realize interconnected operation through a power interconnecting line while carrying out electric energy interaction with a main power grid, so that the full utilization of renewable energy sources in the system can be realized, the dependence of the multi-microgrid system on the main power grid can be reduced, and the economic benefit of the system is further improved. In addition, with the continuous reform of the electric power market, the load demand response reasonably digs the load potential of the user side by guiding the user to adjust the power demand in the peak period, so that the purpose of peak clipping and valley filling is achieved, and the method has become the popular direction of current research. Therefore, when a large number of micro-grid systems with complex structures are connected into the main power grid, the optimized scheduling of the grid-connected multi-micro-grid system operation is realized, the total operation cost of the system is reduced, and the method has important research value for power production and life electricity utilization. Aiming at the minimum problems of optimal scheduling and total running cost of a grid-connected multi-micro-grid system, the intelligent optimization algorithm can be effectively used for scheduling and optimizing the power system by virtue of the characteristics of fewer limiting conditions, higher solving efficiency and the like.
However, with the complexity of the multi-micro-grid system in terms of electric energy interaction and demand response, the solution results of some optimization algorithms have individual abnormal values, the solution efficiency is low, and the solution is in a locally optimal solution for a long time, so that the scheduling problem of the grid-connected multi-micro-grid system needs to be optimized by adopting an optimization algorithm with excellent optimizing capability.
Disclosure of Invention
The invention aims to provide a micro-grid scheduling method considering electric energy interaction and demand response when a micro-grid system with a large number of main power grids and a complex structure is accessed. In the grid-connected multi-microgrid system, the upper limit of the installed number and rated power of equipment of each microgrid is used as an optimization variable, the total running cost of the microgrid system is used as an objective function, and meanwhile, the balance constraint of electric loads in the system, the output constraint of equipment, the capacity constraint of a storage battery and the transaction constraint of the microgrid system and other microgrid systems or the microgrid and a main power grid system are also required to be considered. The wind power generation system and the photovoltaic power generation system in the micro-grid system have non-stable random characteristics in output, so that the optimization difficulty is increased; the invention solves the operation scheduling problem of the grid-connected multi-microgrid system by using an improved marine predator optimization algorithm, adopts disturbance type Tent chaotic mapping and dynamic inertia coefficient in a classical marine predator optimization algorithm to enhance local searching and optimizing capacity, so that the optimization model of the microgrid system can be optimized, the total operation cost of the system is reduced, the utilization rate of renewable energy sources in the system is improved, and the dependence of the multi-microgrid system on a main power grid is reduced.
The aim of the invention is achieved by the following technical scheme. A grid-connected multi-microgrid system operation scheduling optimization method considering electric energy interaction and demand response comprises the following steps:
As a further preferable technical solution, in the step 1, in the operation cost model of the multi-micro network system, the operation cost of the multi-micro network system includes investment operation cost of equipment, electric energy transaction cost with other micro network systems, electric energy transaction cost with a main power grid, fuel cost, pollutant treatment cost and penalty cost, and the modeling for each equipment cost is as follows:
(1) Mathematical models of investment, operation and maintenance costs of equipment are as follows (1):
wherein X is 1 Representing the total cost of investment, operation and maintenance of the equipment; n (N) E The number of kinds of distributed power supplies; p (P) N,e, Representing the installed capacity corresponding to each distributed power supply; f (u, v) is equipment depreciation cost in the micro-grid; u is the damage rate of the equipment; v is the equipment lifetime; x is X I,e ,X Ⅱ,e And X Ⅲ,e Investment, operation maintenance and equipment replacement costs of the micro-grid system are respectively; t (T) year Is an annual operating period;
(2) The mathematical model of the electric energy transaction cost of the micro-grid and the main power grid is as follows (2):
wherein X is 2 Representing the electric energy transaction cost of the micro-grid and the main power grid; p (P) BUY,1 (t) and P SELL,1 (t) representing the electric energy purchase and sale transaction amount between the micro-grid and the main grid at the time t respectively; d, d BUY,1 (t) and d SELL,1 (t) represents the price of the electric energy purchase and sale transaction between the micro-grid and the main grid at time t, respectively; wherein subscript 1 represents the power interaction between the microgrid system and the main grid;
(3) Mathematical models of the electric energy transaction costs of the micro-grid and other micro-grid systems are as follows (3):
wherein X is 3 Representing the electric energy transaction cost of the micro-grid and the main power grid; p (P) BUY,2 (t) and P SELL,2 (t) respectively representing the electric energy purchase and sale transaction amounts of the micro-grid and the main power grid at the moment t; d, d BUY,2 (t) and d SELL,2 (t) represents the electric energy purchase and sale transaction prices of the micro-grid and other micro-grids at the time t respectively; wherein subscript 2 represents power interactions inside the micro-grid system;
(4) Mathematical model of total fuel cost for a multi-microgrid system device is given by formula (4):
wherein X is 4 Representing the total fuel cost of the micro-grid system equipment; u (u) fuel Is the price of diesel; f (F) DG (t) is the diesel consumption of the diesel generator in operation at time t; n (N) DG The number of the diesel generators is the number of the diesel generators; p (P) N,DG The rated power of the Nth diesel generator is represented; p (P) DG Representing the actual power of the diesel generator at the time t; u (u) 1 And u 2 Is the fuel curve factor constant;
(5) Mathematical models of the cost of treatment of contaminants in a multi-microgrid system are given by formula (5):
wherein X is 5 Indicating the total cost of treatment of the contaminants; c μ Is the unit cost of different pollutants; lambda (lambda) DG,μ The emission coefficient of the diesel generator; lambda (lambda) GRID,μ The power purchase coefficient is the main power grid; subscript μ is the type of contaminant;
(6) The mathematical model of penalty cost due to shortage or waste of electric power is as follows (6):
wherein X is 6 Represents penalty cost due to shortage or waste of electric power; p (P) vacancy (t) represents a shortage amount of electric power; p (P) waste (t) is the amount of wasted power; w is the unit penalty cost.
On the other hand, constraint conditions in the multi-microgrid system comprise balance constraint of electric load, output constraint energy storage of equipment, capacity constraint of a storage battery and transaction constraint of the microgrid system and other microgrid systems or a main power grid;
(1) Establishing optimization variables of a multi-micro-grid system optimization model:
in a grid-connected multi-microgrid system, rated power of each microgrid diesel generator and the number of diesel generators are used; the power corresponding to each device in the micro-grid system; the output power of the wind power generation system and the number of wind power generation devices; the output power of the photovoltaic power generation system and the number of photovoltaic power generation devices; the charge and discharge power of the storage battery in the micro-grid system and the quantity of the storage battery are used as optimization variables;
wherein P is N,DG Represents the N < th DG Rated power of the diesel generator; n (N) DG The total number of the diesel generators; p (P) N,e, Representing the power corresponding to each device in the micro-grid system; n (N) E The number of kinds of distributed power supplies; p (P) WT The output power of the wind power generation system; p (P) PV The output power of the photovoltaic power generation system; p (P) c ，P f The charging and discharging power of the storage battery are respectively; e (E) BAT Is the energy storage capacity of the storage battery; n (N) BAT The number of the storage batteries; n (N) DG Is the number of gas turbines; n (N) PV The number of photovoltaic power generation devices; n (N) WT The number of the wind power generation devices;
(2) Establishing an objective function of a multi-micro-grid system optimization model:
in order to minimize the total operation cost of the grid-connected multi-microgrid system, an objective function is established in the aspect of the operation cost of the microgrid system equipment, the purpose of reducing the system operation cost is achieved, and the objective function is shown in formulas (7) and (8):
X MG,n ＝X 1 +X 2 +X 3 +X 4 +X 5 +X 6 (8)
wherein N is MG Is the number of micro-grid systems; x is X MG,n The total running cost of the micro-grid system;
(3) Establishing constraint conditions of a multi-microgrid system optimization model:
the mathematical model of the balance constraint of the electrical load is shown in the following formula (9):
wherein P is WT (t) is the output power of the wind power generation system at the moment t; p (P) PV (t) is the output power of the photovoltaic power generation system at the moment t; p (P) c (t)、P f (t) the charge and discharge power of the storage battery at the moment t respectively; l (L) Load (t) is the electric load demand of the micro-grid system in implementing load transfer at the moment t;
the mathematical model of the energy storage constraint of the storage battery is shown as the following formula (10):
In the method, in the process of the invention,
is the minimum energy storage capacity of the storage battery; />
The maximum energy storage capacity of the storage battery;E BAT (t) is the residual electric energy of the storage battery at the moment t; p (P) r (t) is the charging power of the storage battery at the time t; />
And->
Respectively the minimum and maximum charging power of the storage battery; p (P) f (t) is the discharge power of the storage battery at the time t; />
And->
Respectively the minimum and maximum discharge power of the storage battery;
the mathematical model of the force constraint of the device is shown in the following formula (11):
in the method, in the process of the invention,
and->
The minimum output of the wind power generation system, the photovoltaic power generation system and the diesel generator are respectively; />
And->
Maximum output of the diesel generators of the wind power generation system and the photovoltaic power generation system respectively; />
The mathematical model of the transaction constraints of the microgrid system with other microgrid systems is shown in the following equation (12):
in the method, in the process of the invention,
and->
The minimum and maximum electricity purchasing amounts of the micro-grid system to the nth micro-grid system are respectively;
and->
The minimum and maximum sales power of the micro-grid system to the nth micro-grid system are respectively;
the mathematical model of the trade constraint of the micro-grid system and the main grid is shown in the following formula (13):
in the method, in the process of the invention,
and->
The minimum and maximum electricity purchasing quantity of the micro-grid system to the main grid system is respectively;
and->
The minimum and maximum sales power of the micro-grid system to the main power grid system are respectively;
The number constraint of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery in the micro-grid system is shown as the following (14)
In the method, in the process of the invention,
and->
The minimum configuration quantity of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery is respectively; n (N) WT 、N PV 、N GT 、N BAT The configuration quantity of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery is respectively; />
The maximum configuration number of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery is respectively.
As a further preferable technical solution, the step 2 includes the following steps:
firstly, dividing a power load into peak-flat-valley periods by adopting a fuzzy clustering method, wherein the total interval value of each period is not less than 6 hours, and the peak period during the valley period is not more than 2 hours;
in addition, the present invention also contemplates efficient consumption of renewable energy sources and unnecessary electrical load responses. If the renewable energy production during peak and valley periods is able to meet the power demand, no load response will occur. If the power demand is not met, the remaining power load will take a demand response with the renewable energy fully absorbed.
As a further preferable technical solution, the demand response policy is: according to the importance of the power load and whether the power load has schedulability, the power load in the micro-grid system is divided into a non-response load, a transferable load and an interruptible load; for the interruptible load, excitation demand response is adopted, namely, when the load is in a peak period of electricity consumption or in an emergency, the power generation side actively stops supplying power to a user, and the power generation side needs to carry out economic compensation on the user; for transferable loads, price demand response is adopted, an electricity price elastic matrix is determined, and initial transfer of the power loads is completed;
Wherein, the electricity price elastic matrix is shown in formulas (14), (15) and (16):
the self-elasticity coefficient in the price elasticity matrix Q is represented by the following formula (15):
the cross elastic coefficient in the price elastic matrix Q is represented by the following formula (16):
wherein P is i Representing the amount of charge before a response is required during the i period; ΔP i Representing the load variation after the response is required in the i time period; e, e j Indicating the electricity price before the response is required in the j time period; Δe j Indicating the change amount of electricity price after the response is required in the j time period;
the price elastic matrix Q is represented by the following formula (17):
wherein Q is a price elastic matrix, and f, p and g respectively represent peak, flat and valley periods in a time-sharing electricity price mechanism;
and the electricity price change amount after demand response is shown in the following formula (18):
wherein P is f 、P p And P g Respectively considering the average load quantity of each period after price type demand response; p (P) f,0 、P p,0 And P g,0 The average load quantity of each period before price type demand response is respectively; Δe f 、Δe p And Δe g Respectively representing the electricity price change quantity of each period after the demand response; e, e f 、e p And e g The electricity price before time-sharing electricity price is adopted is represented;
the formula of the total load demand is shown in the following formula (19);
L Load (t)＝L Load,I (t)+L Load,trans (t)+L Load,break (t) (19)
wherein L is Load,I (t)、L Load,trans (t) and L Load,break (t) the demand of unresponsive load, transferable load and interruptible load inside the microgrid system at time t respectively;
Considering the characteristics of transferable load and interruptible load, the mathematical model of the load demand is shown in formula (20):
wherein L is after (t) and L before (t) electrical load demand before and after load transfer and load interruption, respectively; ΔL II,trans (t) is the load actually transferred; ΔL III,break (t) is the actual interrupt load; sigma (sigma) t',t =1 represents that the transferable load is transferred from time t' to time; sigma (sigma) t',t -1 represents a transferable load transfer from time t to time t'; lambda (lambda) t =1 indicates a load interruption, λ t =0 indicates no load interruption.
In the step 3, based on the classical marine predator algorithm, aiming at the problems of different speed ratios in the algorithm, the disturbance type Tent chaotic map, the dynamic inertia coefficient, the differential evolution combined with elite retention and the elite reverse learning combined with the cauchy variation are adopted to improve the same respectively to obtain an improved marine predator algorithm (HIMPA);
(1) Disturbance type Tent chaotic mapping
At present, the Tent chaotic mapping is regarded as an improvement strategy for enhancing population diversity by virtue of the characteristics of strong randomness, good uniformity and the like. However, some expert scholars find that the Tent chaotic map is easy to sink into a small cycle period or a fixed point, so that the invention designs a disturbance type Tent chaotic map, as shown in a formula (21), for improving the spatial distribution of a marine environment:
In which x is i Representing the ith chaos number; a is a chaotic variable, where a e (0, 1);
meanwhile, the problem of out-of-range search is solved by adopting a random out-of-range processing strategy, as shown in a formula (22), the phenomenon of single assignment is avoided, and the diversity of the population is ensured.
Wherein i and j represent search space ranges of individuals in the population; l (L) s Representing a prey matrix; l (L) s,0 Representing an initial prey matrix;
(2) Coefficient of dynamic inertia
In the high speed ratio stage of the classical marine predator algorithm, the prey does not utilize its own position when the position is updated, and in order to improve the optimizing capability of the high speed ratio stage in the later stage, a dynamic inertia coefficient is introduced, as shown in formula (23), so as to increase the capability of local searching. In the early global search, the convergence accuracy of the algorithm is further improved.
Wherein beta is 1 And beta 2 Is an adjustment coefficient, wherein beta 1 Has a value of 0.9, beta 2 The value of (2) is 0.4;
(3) Differential evolution combining elite retention
In the unit speed ratio, three different individuals in the population are randomly selected after the hunting population completes the Lewy movement or the Brownian movement, the variation of the population individuals is realized by adopting differential evolution and combining with a Cauchy mutation operator, as shown in a formula (24),
wherein q1, q2 and q3 are three random positions within the population size;
Meanwhile, in order to prevent individual variation from exceeding the boundary range, a random out-of-range processing strategy is adopted. After individual variation, an elite strategy is adopted to reserve individuals with better fitness values, so that population diversity is further improved, and local optimizing capability is enhanced.
(4) Elite reverse learning with fused cauchy variation
In the low speed ratio stage, the optimizing capability of the algorithm is influenced by the degradation of population diversity in the later stage, elite reverse learning is adopted for the hunting in the final stage, and the cauchy variation is fused, as shown in a formula (25), so that the algorithm is prevented from being sunk into a local optimal solution prematurely.
Wherein PR is PR s Population individuals generated after elite reverse learning; PC (personal computer) s Are individuals of the population generated after cauchy mutation; e is the reverse learned exchange coefficient.
As a further preferable technical solution, the step 4 includes the following steps:
(1) First, an initial prey matrix L with a population number S and a dimension D is generated 0 Obtaining a top predator matrix E according to the fitness value result, and finally finishing population initialization of the marine environment;
adopting disturbance type Tent chaotic mapping to improve the spatial distribution of the marine environment and obtain an initial prey matrix P 0 As shown in formula (26):
wherein s is population number; d is the population dimension; x is X ij Representing a prey individual;
for each hunting subject X ij Calculating the fitness value, and finding the most suitable individual with the fitness as Y ij Forming an initial top predator matrix E 0 As shown in formula (27);
(2) The optimizing process is divided into a high speed ratio (H < H/3), a unit speed ratio (H/3 < H < 2H/3) and a low speed ratio (2H/3 < H < H) according to the current iteration number and the maximum iteration number Hmax.
In high speed ratios (H < H/3), i.e. the top predator has a much lower speed than the prey itself in the early stages of the iteration; enhancing the early local search capability by introducing dynamic inertia coefficients, allows the prey to perform Brownian motion for a better global search while the predator is maintained at the current individual location, as shown in formula (28) below
Wherein Ls and E s A prey matrix and a top predator matrix, respectively; p is a constant, which is 0.5; step s Is the hunting movement step length; RB is Brownian operator; beta 1 And beta 2 Is an adjustment coefficient, wherein beta 1 Has a value of 0.9, beta 2 The value of (2) is 0.4; hmax is the maximum iteration number, h is the current iteration number;
(3) In the unit speed ratio (H/3 < H < 2H/3), the hunting population is equally divided into two parts, wherein one part of hunting is subjected to the Lewy motion, the other part of hunting is subjected to the Brownian motion, and development and exploration of fitness values in a search space are respectively completed, as shown in a formula (29) and a formula (30):
C in the formula F Is an adaptive coefficient; p is a constant, which is 0.5; as the number of iterations changes; RL is the Levy operator; step s Is the hunting movement step length;
after the hunter population completes the Lewy movement or Brownian movement, three different individuals in the population are randomly selected, and the variation of the population individuals is realized by adopting differential evolution and combining with a Cauchy mutation operator; meanwhile, individuals with good fitness values are reserved by adopting elite strategies, so that population diversity is further improved, and local optimizing capability is enhanced.
(4) In the low speed ratio (2H/3 < H < H), namely the self speed of the top predator is far higher than that of the prey in the later period of iteration, the position update of the prey is avoided from being caught according to the Lewy track of the top predator, as shown in a formula (31), and the local optimizing capability of an algorithm is improved.
And (3) performing elite reverse learning on the hunting matters in the final stage, and fusing the cauchy variation to avoid the algorithm from sinking into a local optimal solution prematurely.
(5) Finally, the ocean eddy and fish gathering phenomenon (FADs) are regarded as the local optimal solution of the search space, and the predators complete the position update with larger amplitude according to the update mechanism of the formula (32), so that the phenomenon of sinking into the local optimal solution is further avoided.
Wherein R is a binary number set randomly generated from a binary vector; r is a random number between 0 and 1; PF is a disturbance probability coefficient generated by FADs, and the value is 0.2; l (L) s1 And L s2 Is a random individual in the hunter population;
in addition, the ocean memory is updated for individuals of the top predator matrix E, and when the fitness value of the prey matrix L is calculated, if the optimal fitness value is changed, the corresponding individuals in the original top predator matrix are replaced.
(6) Judging whether the maximum iteration times are reached, if so, outputting an optimal result of the algorithm to obtain the minimum running cost of the multi-micro-grid system.
As a further preferable technical solution, in the step 6, the total amount of supply and demand in the multi-micro network system is shown in formula (33), and the supply and demand ratio and the demand supply are shown in formula (34):
wherein P is sell (t) and P buy (t) the total energy supply amount and the total demand amount of the multi-micro-grid system at the time t respectively;
wherein A (t) and B (t) are respectively the supply-demand ratio and the demand-supply ratio of the multi-micro-grid system at the moment t;
(1) When the supply and demand relationship does not exist in the micro-grid system, the internal electricity price is the same as the electricity price set by the main grid, and d is the same as the electricity price set by the main grid sell (t) and d buy (t) are all 0, and the internal electricity price is shown as a formula (35);
(2) When the supply and demand are balanced, the internal electricity price should be set at the intermediate price, and d sell (t)And d buy (t) equal internal electricity prices are as shown in formula (36):
(3) D when the total energy supply of the micro-grid system is smaller than the total demand sell (t) is less than d buy (t) the internal electricity price is represented by formula (37):
d when the total energy supply of the micro-grid system is greater than the total demand sell (t) is greater than d buy (t) the internal electricity price is represented by formula (38):
the grid-connected multi-micro-grid system operation scheduling optimization strategy taking the electric energy interaction and the demand response into consideration is that the demand response of users in the grid-connected multi-micro-grid system and the pricing strategy in the micro-grid are known in the prior art and are well known to those skilled in the art.
The grid-connected multi-microgrid system operation scheduling optimization strategy with the electric energy interaction and demand response, wherein the fuzzy clustering method, the marine predator algorithm and the Tent chaotic mapping are existing methods.
The grid-connected multi-microgrid system with the electric energy interaction and the demand response operates a scheduling optimization strategy, and the power, the load and the like of each device in the microgrid system are well known to those skilled in the art.
The invention has the advantages and positive effects that:
1. the invention is based on the dispatching optimization research of the operation of the grid-connected multi-micro-grid system, takes the energy flow and the operation cost of the multi-micro-grid system as main bodies, and carries out the example simulation analysis on the economy and the reliability of the operation of the grid-connected multi-micro-grid system. Meanwhile, the electric energy interaction and the load demand response inside the system are researched, and the time interval division of peak-to-valley is considered in the load demand response, so that the electric energy interaction and the load response can be effectively realized under the condition of gentle load fluctuation, and the running cost can be reduced;
2. The invention improves the classical Tent chaotic map and is applied to the initial stage of the marine predator algorithm, thereby improving the algorithm environment to a limited extent and ensuring the diversity of the population; differential evolution is adopted in the medium speed ratio stage of the algorithm, and a cauchy mutation operator is combined to realize the mutation of population individuals, so that the population diversity is further improved, and the local optimizing capability is enhanced; the method has the advantages that the method adopts elite reverse learning to fuse the cauchy variation in the prey matrix adopted in the low speed ratio stage, solves the problem that the algorithm is sunk into the optimal solution prematurely, and remarkably improves the convergence speed and precision of the algorithm;
3. in order to solve the problem of optimizing the total running cost of the system, the invention utilizes disturbance type Tent chaotic mapping, dynamic inertia coefficient, differential evolution combined with elite retention and elite reverse learning combined with cauchy variation to improve the traditional MPA, and provides the HIMPA, thereby remarkably improving the optimizing capability of the algorithm, effectively improving the economic benefit of the running of the system on the basis of ensuring the stable output and load balance of the equipment, and verifying the feasibility and stability of the HIMPA in solving the total running cost of the system.
Drawings
FIG. 1 is a schematic flow chart of a grid-connected multi-microgrid system operation scheduling strategy according to the present invention;
FIG. 2 is a schematic flow diagram of the improved marine predator algorithm of the present invention;
FIG. 3 is a diagram of a model structure of a grid-connected multi-microgrid system of the present invention;
FIG. 4 is a graph of the result of dividing peak-to-valley time periods in a micro-grid system by a fuzzy clustering method in the embodiment of the invention;
FIG. 5 is a schematic diagram of specific flow conditions of energy of each micro-grid system under different scenes in each micro-grid according to the embodiment of the present invention; wherein FIG. 5 (a) shows an energy flow diagram for each micro-grid system without regard to power interactions within the system and demand responses; 5 (b) represents that each micro-grid system considers the power interaction inside the system, but the MG2 and the MG3 do not consider the energy flow diagram under the requirement response; FIG. 5 (c) shows an energy flow diagram for MG2 and MG3 taking into account demand response, but each micro-grid system taking into account power interactions within the system; fig. 5 (d) shows an energy flow diagram under consideration of the power interaction inside the system by each micro-grid system, and the demand response by MG2 and MG 3.
FIG. 6 is a schematic diagram showing the specific flow of energy of each micro-grid system after each micro-grid system implements the scheduling scheme according to the embodiment of the present invention; wherein fig. 6 (a) shows an energy flow diagram of the micro-grid system MG1 after performing scheduling optimization; fig. 6 (b) shows an energy flow diagram of the micro-grid system MG2 after performing scheduling optimization; fig. 6 (c) shows an energy flow diagram of the micro-grid system MG3 after performing scheduling optimization; fig. 6 (d) shows the price of the electric energy transaction inside the micro-grid system.
Detailed Description
The invention will now be described in detail with reference to the accompanying drawings and examples:
examples: a grid-connected multi-micro-grid system operation scheduling optimization method considering electric energy interaction and demand response is disclosed as a program flow of a multi-micro-grid system operation scheduling strategy shown in fig. 1: starting, establishing a multi-microgrid system dispatching optimization model, setting upper and lower limits of constraint conditions, determining renewable energy sources and load demand sources, carrying out peak-flat-valley period division on the microgrid system, carrying out demand response analysis on internal electric loads, solving the multi-microgrid system by using HIMPA, judging the obtained optimal result, judging whether an abnormal value appears, executing a pricing strategy and completing internal transaction when the abnormal value does not appear, carrying out electric energy transaction with a main power grid or starting a diesel generator according to the internal optimization result, judging the final result of the dispatching optimization, if the operation cost is low, storing the related result, judging whether the termination condition of a daily operation period T is met, outputting the final result and ending.
The method comprises the following specific steps:
In the operation cost model of the multi-micro-grid system, the operation cost of the multi-micro-grid system comprises investment operation cost of equipment, electric energy transaction cost with other micro-grid systems, electric energy transaction cost with a main power grid, fuel cost, pollutant treatment cost and punishment cost, and the modeling of the cost of each equipment is as follows:
(1) Mathematical models of investment, operation and maintenance costs of equipment are as follows (1):
wherein X is 1 Representing the total cost of investment, operation and maintenance of the equipment; n (N) E The number of kinds of distributed power supplies; p (P) N,e, Representing the installed capacity corresponding to each distributed power supply; f (u, v) is equipment depreciation cost in the micro-grid; u is the damage rate of the equipment; v is the equipment lifetime; x is X I,e ,X Ⅱ,e And X Ⅲ,e Investment, operation maintenance and equipment replacement costs of the micro-grid system are respectively; t (T) year Is an annual operating period;
(2) The mathematical model of the electric energy transaction cost of the micro-grid and the main power grid is as follows (2):
wherein X is 2 Representing the electric energy transaction cost of the micro-grid and the main power grid; p (P) BUY,1 (t) and P SELL,1 (t) representing the electric energy purchase and sale transaction amount between the micro-grid and the main grid at the time t respectively; d, d BUY,1 (t) and d SELL,1 (t) represents a micro-net and respectivelyThe purchase and sale transaction price of the electric energy at the time t between the main power grids; wherein subscript 1 represents the power interaction between the microgrid system and the main grid;
(3) Mathematical models of the electric energy transaction costs of the micro-grid and other micro-grid systems are as follows (3):
wherein X is 3 Representing the electric energy transaction cost of the micro-grid and the main power grid; p (P) BUY,2 (t) and P SELL,2 (t) respectively representing the electric energy purchase and sale transaction amounts of the micro-grid and the main power grid at the moment t; d, d BUY,2 (t) and d SELL,2 (t) represents the electric energy purchase and sale transaction prices of the micro-grid and other micro-grids at the time t respectively; wherein subscript 2 represents power interactions inside the micro-grid system;
(4) Mathematical model of total fuel cost for a multi-microgrid system device is given by formula (4):
wherein X is 4 Representing the total fuel cost of the micro-grid system equipment; u (u) fuel Is the price of diesel; f (F) DG (t) is the diesel consumption of the diesel generator in operation at time t; n (N) DG The number of the diesel generators is the number of the diesel generators; p (P) N,DG The rated power of the Nth diesel generator is represented; p (P) DG Representing the actual power of the diesel generator at the time t; u (u) 1 And u 2 Is the fuel curve factor constant;
(5) Mathematical models of the cost of treatment of contaminants in a multi-microgrid system are given by formula (5):
wherein X is 5 Indicating the total cost of treatment of the contaminants; c μ Is the unit cost of different pollutants; lambda (lambda) DG,μ The emission coefficient of the diesel generator; lambda (lambda) GRID,μ The power purchase coefficient is the main power grid; subscript μ is the type of contaminant;
(6) The mathematical model of penalty cost due to shortage or waste of electric power is as follows (6):
wherein X is 6 Represents penalty cost due to shortage or waste of electric power; p (P) vacancy (t) represents a shortage amount of electric power; p (P) waste (t) is the amount of wasted power; w is the unit penalty cost.
On the other hand, constraint conditions in the multi-microgrid system comprise balance constraint of electric load, output constraint energy storage of equipment, capacity constraint of a storage battery and transaction constraint of the microgrid system and other microgrid systems or a main power grid;
(1) Establishing optimization variables of a multi-micro-grid system optimization model:
in a grid-connected multi-microgrid system, rated power of each microgrid diesel generator and the number of diesel generators are used; the power corresponding to each device in the micro-grid system; the output power of the wind power generation system and the number of wind power generation devices; the output power of the photovoltaic power generation system and the number of photovoltaic power generation devices; the charge and discharge power of the storage battery in the micro-grid system and the quantity of the storage battery are used as optimization variables;
wherein P is N,DG Represents the N < th DG Rated power of the diesel generator; n (N) DG The total number of the diesel generators; p (P) N,e, Representing the power corresponding to each device in the micro-grid system; n (N) E The number of kinds of distributed power supplies; p (P) WT The output power of the wind power generation system; p (P) PV The output power of the photovoltaic power generation system; p (P) c ，P f Respectively charge and discharge of the accumulatorA power; e (E) BAT Is the energy storage capacity of the storage battery; n (N) BAT The number of the storage batteries; n (N) DG Is the number of gas turbines; n (N) PV The number of photovoltaic power generation devices; n (N) WT The number of the wind power generation devices;
(2) Establishing an objective function of a multi-micro-grid system optimization model:
in order to minimize the total operation cost of the grid-connected multi-microgrid system, an objective function is established in the aspect of the operation cost of the microgrid system equipment, the purpose of reducing the system operation cost is achieved, and the objective function is shown in formulas (7) and (8):
X MG,n ＝X 1 +X 2 +X 3 +X 4 +X 5 +X 6 (8)
wherein N is MG Is the number of micro-grid systems; x is X MG,n The total running cost of the micro-grid system;
(3) Establishing constraint conditions of a multi-microgrid system optimization model:
the mathematical model of the balance constraint of the electrical load is shown in the following formula (9):
wherein P is WT (t) is the output power of the wind power generation system at the moment t; p (P) PV (t) is the output power of the photovoltaic power generation system at the moment t; p (P) c (t)、P f (t) the charge and discharge power of the storage battery at the moment t respectively; l (L) Load (t) is the electric load demand of the micro-grid system in implementing load transfer at the moment t;
the mathematical model of the energy storage constraint of the storage battery is shown as the following formula (10):
In the method, in the process of the invention,
is the minimum energy storage capacity of the storage battery; />
The maximum energy storage capacity of the storage battery; e (E) BAT (t) is the residual electric energy of the storage battery at the moment t; p (P) r (t) is the charging power of the storage battery at the time t; />
And->
Respectively the minimum and maximum charging power of the storage battery; p (P) f (t) is the discharge power of the storage battery at the time t; />
And->
Respectively the minimum and maximum discharge power of the storage battery;
the mathematical model of the force constraint of the device is shown in the following formula (11):
in the method, in the process of the invention,
and->
The minimum output of the wind power generation system, the photovoltaic power generation system and the diesel generator are respectively; />
And->
Maximum output of the diesel generators of the wind power generation system and the photovoltaic power generation system respectively;
the mathematical model of the transaction constraints of the microgrid system with other microgrid systems is shown in the following equation (12):
in the method, in the process of the invention,
and->
The minimum and maximum electricity purchasing amounts of the micro-grid system to the nth micro-grid system are respectively;
and->
The minimum and maximum sales power of the micro-grid system to the nth micro-grid system are respectively;
the mathematical model of the trade constraint of the micro-grid system and the main grid is shown in the following formula (13):
in the method, in the process of the invention,
and->
The minimum and maximum electricity purchasing quantity of the micro-grid system to the main grid system is respectively;
and->
The minimum and maximum sales power of the micro-grid system to the main power grid system are respectively;
The number constraint of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery in the multi-microgrid system is shown as the following (14)
In the method, in the process of the invention,
and->
The minimum configuration quantity of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery is respectively; n (N) WT 、N PV 、N GT 、N BAT The configuration quantity of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery is respectively; />
The maximum configuration number of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery is respectively.
Firstly, constructing a mathematical model of investment operation cost of micro-grid system equipment according to the minimum total operation cost of the system as an optimization objective function, constructing a mathematical model of electric energy transaction cost of the micro-grid and a main power grid or other micro-grid systems by using formulas (2) and (3), constructing a mathematical model of fuel total cost of multi-micro-grid system equipment by using formula (4), and constructing a mathematical model of treatment cost of pollutants and punishment cost of electric loss in the multi-micro-grid system by using formulas (5) and (6); then adding the formulas (1) to (6) to establish the total running cost of the system as shown in the formula (8); finally, the minimum total running cost of the system is used as an optimization objective function, as shown in a formula (7);
then establishing constraint conditions: as shown in fig. 3, the present invention is a model structure of a grid-connected multi-microgrid system, where each microgrid has a trade of electric energy with a main grid. The system comprises three micro-grid systems, and each micro-grid system is connected by a power interconnecting line. Meanwhile, the wind power generation system (WT), the photovoltaic power generation system (PV), the storage battery energy storage system (BAT) and the Diesel Generator (DG) of each micro-grid system all adopt equivalent output models, and the internal electricity load is self-demand, so that earning of benefits is not considered. MG1 is a wind power micro-grid system, MG2 is a photovoltaic micro-grid system, and MG3 simultaneously comprises a wind power generation system (WT) and a photovoltaic power generation system (PV). Furthermore, the power loss of the entire multi-microgrid system is not considered.
In the grid-connected multi-microgrid system constructed by the invention, the balance constraint of electric load, the output constraint of equipment, the capacity constraint of a storage battery and the transaction constraint of the microgrid system and other microgrid systems or a main power grid are respectively established through formulas (9) - (13); further, the minimum number of devices to be arranged of the multi-microgrid system is set to 1, and the upper limit of the installed number and the constitution type thereof are shown in table 1 below.
TABLE 1 installation upper limit and composition type number of each micro-grid system
In the formulas (1) - (13) of the present embodiment, the load data, the environmental data and the relevant parameter references of the devices of the multi-micro network system are "hierarchical energy optimization management of active distribution network including multi-micro network system" (2022), and specific data are shown in the following table 2; meanwhile, for the treatment cost of pollutants, the treatment price and the emission quantity of each pollutant refer to grid-connected micro-grid optimal configuration model (2017) considering the comprehensive performance of the micro-grid, and specific data are shown in the following table 3; the specific prices of the peak-flat-valley period and the transaction electricity price reference price-guided multi-microgrid system coordinated autonomous optimization operation strategy (2019) formulated by the main power grid are shown in the following table 4; in the daily operation scheduling of the system, the maximum ratio of the interruptible load in unit time is 0.1, and the compensation cost is 0.3 yuan/kWh; the electricity price elastic matrix uses an industrial consumer electricity price elastic matrix.
Table 2 parameters related to devices in a multi-microgrid system
TABLE 3 emission coefficient of polluted gas and treatment cost
TABLE 4 time-of-use electricity price
firstly, dividing a power load into peak-flat-valley periods by adopting a fuzzy clustering method, wherein the total interval value of each period is not less than 6 hours, and the peak period during the valley period is not more than 2 hours;
in addition, the present invention also contemplates efficient consumption of renewable energy sources and unnecessary electrical load responses. If the renewable energy production during peak and valley periods is able to meet the power demand, no load response will occur. If the power demand is not met, the remaining power load will take a demand response with the renewable energy fully absorbed.
As a further preferable technical solution, the demand response policy is: according to the importance of the power load and whether the power load has schedulability, the power load in the micro-grid system is divided into a non-response load, a transferable load and an interruptible load; for the interruptible load, excitation demand response is adopted, namely, when the load is in a peak period of electricity consumption or in an emergency, the power generation side actively stops supplying power to a user, and the power generation side needs to carry out economic compensation on the user; for transferable loads, price demand response is adopted, an electricity price elastic matrix is determined, and initial transfer of the power loads is completed;
Wherein, the electricity price elastic matrix is shown in formulas (14), (15) and (16):
the self-elasticity coefficient in the price elasticity matrix Q is represented by the following formula (15):
the cross elastic coefficient in the price elastic matrix Q is represented by the following formula (16):
wherein P is i Representing the amount of charge before a response is required during the i period; ΔP i Representing the load variation after the response is required in the i time period; e, e j Indicating the electricity price before the response is required in the j time period; Δe j Indicating the change amount of electricity price after the response is required in the j time period;
the price elastic matrix Q is represented by the following formula (17):
wherein Q is a price elastic matrix, and f, p and g respectively represent peak, flat and valley periods in a time-sharing electricity price mechanism;
and the electricity price change amount after demand response is shown in the following formula (18):
wherein P is f 、P p And P g Respectively isConsidering the average load of each period after price type demand response; p (P) f,0 、P p,0 And P g,0 The average load quantity of each period before price type demand response is respectively; Δe f 、Δe p And Δe g Respectively representing the electricity price change quantity of each period after the demand response; e, e f 、e p And e g The electricity price before time-sharing electricity price is adopted is represented;
the formula of the total load demand is shown in the following formula (19);
L Load (t)＝L Load,I (t)+L Load,trans (t)+L Load,break (t) (19)
wherein L is Load,I (t)、L Load,trans (t) and L Load,break (t) the demand of unresponsive load, transferable load and interruptible load inside the microgrid system at time t respectively;
Considering the characteristics of transferable load and interruptible load, the mathematical model of the load demand is shown in formula (20):
wherein L is after (t) and L before (t) electrical load demand before and after load transfer and load interruption, respectively; ΔL II,trans (t) is the load actually transferred; ΔL III,break (t) is the actual interrupt load; sigma (sigma) t',t =1 represents that the transferable load is transferred from time t' to time; sigma (sigma) t',t -1 represents a transferable load transfer from time t to time t'; lambda (lambda) t =1 indicates a load interruption, λ t =0 indicates no load interruption.
In a multi-microgrid system, MG2 and MG3 are selected as research objects, peak-valley time period division is carried out on internal loads of the multi-microgrid system by using a fuzzy clustering method, and then demand response analysis is carried out; wherein MG1 is a comparison object, and no demand response analysis is performed.
The division result of the initial peak-to-valley period according to the fuzzy clustering method is shown in fig. 4, wherein a red curve represents the peak period, a black curve represents the flat period, and a blue curve represents the valley period. Based on the initial division result and the revised principle, the peak-to-valley final division results of MG2 and MG3 are shown in table 5.
Table 5 peak-to-valley period division results for MG2 and MG3
In order to intuitively analyze the influence of electric energy interaction and demand response on the energy flow of the system, according to the electric energy transaction between the multi-micro-grid system and a main power grid and the electric energy interaction inside the multi-micro-grid system, the following four scenes are selected:
(1) Scene 1: each micro-grid system does not consider the electric energy interaction and the demand response inside the system;
(2) Scene 2: each micro-grid system considers the electric energy interaction inside the system, but MG2 and MG3 do not consider the demand response;
(3) Scene 3: MG2 and MG3 consider the demand response, but each micro-grid system does not consider the power interaction inside the system;
(4) Scene 4: each micro-grid system considers power interactions inside the system, and MG2 and MG3 consider demand responses.
A specific flow of system energy is shown in fig. 5. In both fig. 5 (a) and fig. 5 (b), the power interaction within the system occurs mainly within the time periods 0:00-4:00, 10:00-12:00, and 14:00-18:00, without considering the demand response. In the range of 0:00-4:00, the electric power sold by MG1 to MG3 is 570.41kW, and the electric power sold by the multi-micro-grid system to the main power grid at the stage is correspondingly reduced compared with that of the multi-micro-grid system shown in the (a) of fig. 5. In 10:00-12:00, the power purchased by MG2 from MG1 is 703.34kW. At 12:00-18:00, there is a simultaneous electric energy transaction of the multi-microgrid system with the main electric network and electric energy interaction inside the multi-microgrid system at this stage. Within 14:00-18:00, the power purchased by MG2 from MG1 was 562.43kW, the power purchased by MG3 was 779.50kW, and the power purchased by MG3 from MG1 was 255.81kW. Compared with fig. 5 (a), the power purchasing power of the multi-micro-grid system from the main power grid is correspondingly reduced by 2301.08kW, and the power supply pressure of the main power grid is relieved.
Under the condition that the electric energy interaction in the system is not considered, the MG2 and the MG3 transfer the electric loads of the peak period to the flat and valley period preferentially according to the period division result, and the electric loads of the ordinary period can be transferred to the valley period correspondingly. As shown in fig. 5 (a) and 5 (c), in the valley period 0:00-4:00, the electricity purchase amount of the multi-microgrid system to the main power grid is obviously increased, and the electricity purchase power is increased by 1978.96kW. In the peak, the normal period 7:00-12:00 and the normal period 16:00-21:00, the electricity purchasing quantity of the multi-micro-grid system to the main power grid is obviously reduced, and the electricity purchasing power is respectively reduced by 1500.53kW and 2655.31kW. On the premise of ensuring basic power consumption requirements, the running cost of the multi-micro-grid system can be further reduced. As shown in fig. 5 (d), when the multi-micro-grid system has internal power interaction and demand response, the energy flow is complex, and a reasonable and effective scheduling optimization scheme is needed to be adopted, so that the smooth operation of the system is realized.
According to the energy flow result shown in fig. 5 (d), the scheduling optimization scheme of each micro-grid system, the electricity price change of the internal electric energy interaction and the load change before and after the demand response are shown in fig. 6.
In MG1, as shown in FIG. 6 (a), it is obvious that the WT has a larger output in daily operation, and can meet most of load requirements, and only needs to purchase electricity to the main power grid in the time periods of 4:00-8:00 and 20:00-24:00. Insufficient electrical energy may be provided by the battery, and the diesel generator is not started. In addition, the surplus energy can be stored by a storage battery or sold to a main power grid or other micro-grid systems.
In MG2, as shown in FIG. 6 (b), the PV has a certain output only in the period of 6:00-17:00, and especially in the period of 10:00-16:00, the load requirement can be met, and the surplus electric energy is used for storing energy or selling electricity. Because of the smaller output of the PV during other periods, electricity needs to be purchased from the main grid or other microgrid system. In addition, the battery provides limited power, and thus a diesel generator is required as a backup power source.
In MG3, as shown in fig. 6 (c), WT and PV provide renewable energy sources simultaneously, and can effectively meet internal load requirements in a 12:00-14:00 time period, and store energy and sell electricity. In other time periods, electricity is purchased from a main power grid or other micro-grid systems, but the running time of the diesel generator is smaller than that of the MG2.
In MG2 and MG3, the red portion of the load curve indicates that no demand response has occurred. In the time period, the output of the renewable energy source can completely meet the load demand, and a user does not need to carry out corresponding load interruption and load transfer, so that the internal consumption of the renewable energy source is increased, and the electricity utilization feeling of the user is improved. The peak-to-valley difference change after the demand response was performed is shown in table 6. As is clear from table 6, the load peak-to-valley difference of MG2 was reduced from 1044.66kW to 853.06kW, and the reduction range was 18.34%. Meanwhile, the load peak-valley difference of the MG3 is reduced from 779.90kW to 755.92kW, and the reduction amplitude is 3.07%. Therefore, the demand response strategy can reduce the load demand in the electricity consumption peak period so as to achieve the purpose of peak clipping and valley filling.
Table 6 load variation kW after demand response
In the change of the electricity prices of the system internal power interactions, as shown in fig. 6 (d), the internal interaction electricity prices thereof are always kept between the trading electricity prices established by the main power grid. The internal electricity prices change due to the energy flow inside the multi-microgrid system during the time periods 0:00-4:00, 10:00-12:00 and 15:00-18:00.
And 3, respectively adopting disturbance type Tent chaotic mapping, dynamic inertia coefficient, differential evolution combined with elite preservation and elite reverse learning combined with cauchy variation to improve the disturbance type Tent chaotic mapping, dynamic inertia coefficient and elite reverse learning aiming at the problem of different speed ratios in the algorithm on the basis of a classical marine predator algorithm, so as to obtain an improved marine predator algorithm (HIMPA). The invention has the innovation points that the classical Tent chaotic map is improved and applied to the initial stage of the marine predator algorithm, the differential evolution is adopted in the medium speed ratio stage of the algorithm, the variation of population individuals is realized by combining with the Cauchy mutation operator, the elite reverse learning fusion Cauchy variation is adopted in the hunting matrix adopted in the low speed ratio stage, and the improvement of the marine predator algorithm is finally realized.
(1) Disturbance type Tent chaotic mapping
At present, the Tent chaotic mapping is regarded as an improvement strategy for enhancing population diversity by virtue of the characteristics of strong randomness, good uniformity and the like. However, some expert scholars find that the Tent chaotic map is easy to sink into a small cycle period or a fixed point, so that the invention designs a disturbance type Tent chaotic map, as shown in a formula (21), for improving the spatial distribution of a marine environment:
In which x is i Representing the ith chaos number; a is a chaotic variable, where a e (0, 1);
meanwhile, the problem of out-of-range search is solved by adopting a random out-of-range processing strategy, as shown in a formula (22), the phenomenon of single assignment is avoided, and the diversity of the population is ensured.
Wherein i and j represent search space ranges of individuals in the population; l (L) s Representing a prey matrix; l (L) s,0 Representing an initial prey matrix;
(2) Coefficient of dynamic inertia
In the high speed ratio stage of the classical marine predator algorithm, the prey does not utilize its own position when the position is updated, and in order to improve the optimizing capability of the high speed ratio stage in the later stage, a dynamic inertia coefficient is introduced, as shown in formula (23), so as to increase the capability of local searching. In the early global search, the convergence accuracy of the algorithm is further improved.
Wherein beta is 1 And beta 2 Is an adjustment coefficient, wherein beta 1 Has a value of 0.9, beta 2 The value of (2) is 0.4;
(3) Differential evolution combining elite retention
In the unit speed ratio, three different individuals in the population are randomly selected after the hunting population completes the Lewy movement or the Brownian movement, the variation of the population individuals is realized by adopting differential evolution and combining with a Cauchy mutation operator, as shown in a formula (24),
wherein q1, q2 and q3 are three random positions within the population size;
Meanwhile, in order to prevent individual variation from exceeding the boundary range, a random out-of-range processing strategy is adopted. After individual variation, an elite strategy is adopted to reserve individuals with better fitness values, so that population diversity is further improved, and local optimizing capability is enhanced.
(4) Elite reverse learning with fused cauchy variation
In the low speed ratio stage, the optimizing capability of the algorithm is influenced by the degradation of population diversity in the later stage, elite reverse learning is adopted for the hunting in the final stage, and the cauchy variation is fused, as shown in a formula (25), so that the algorithm is prevented from being sunk into a local optimal solution prematurely.
Wherein PR is PR s Population individuals generated after elite reverse learning; PC (personal computer) s Are individuals of the population generated after cauchy mutation; e is the reverse learned exchange coefficient.
(1) First, an initial prey matrix L with a population number S and a dimension D is generated 0 Obtaining a top predator matrix E according to the fitness value result, and finally finishing population initialization of the marine environment;
adopting disturbance type Tent chaotic mapping to improve the spatial distribution of the marine environment and obtain an initial prey matrix P 0 As shown in formula (26):
wherein s is population number; d is the population dimension; x is X ij Representing a prey individual;
for each hunting subject X ij Calculating the fitness value, and finding the most suitable individual with the fitness as Y ij Forming an initial top predator matrix E 0 As shown in formula (27);
(2) The optimizing process is divided into a high speed ratio (H < H/3), a unit speed ratio (H/3 < H < 2H/3) and a low speed ratio (2H/3 < H < H) according to the current iteration number and the maximum iteration number Hmax.
In high speed ratios (H < H/3), i.e. the top predator has a much lower speed than the prey itself in the early stages of the iteration; enhancing the early local search capability by introducing dynamic inertia coefficients, allows the prey to perform Brownian motion for a better global search while the predator is maintained at the current individual location, as shown in formula (28) below
Wherein Ls and E s A prey matrix and a top predator matrix, respectively; p is a constant, which is 0.5; step s Is the hunting movement step length; RB is Brownian operator; beta 1 And beta 2 Is an adjustment coefficient, wherein beta 1 Has a value of 0.9, beta 2 The value of (2) is 0.4; hmax is the maximum iteration number, h is the current iteration number;
(3) In the unit speed ratio (H/3 < H < 2H/3), the hunting population is equally divided into two parts, wherein one part of hunting is subjected to the Lewy motion, the other part of hunting is subjected to the Brownian motion, and development and exploration of fitness values in a search space are respectively completed, as shown in a formula (29) and a formula (30):
c in the formula F Is an adaptive coefficient; p is a constant, which is 0.5; as the number of iterations changes; RL is the Levy operator; step s Is the hunting movement step length;
after the hunter population completes the Lewy movement or Brownian movement, three different individuals in the population are randomly selected, and the variation of the population individuals is realized by adopting differential evolution and combining with a Cauchy mutation operator; meanwhile, individuals with good fitness values are reserved by adopting elite strategies, so that population diversity is further improved, and local optimizing capability is enhanced.
(4) In the low speed ratio (2H/3 < H < H), namely the self speed of the top predator is far higher than that of the prey in the later period of iteration, the position update of the prey is avoided from being caught according to the Lewy track of the top predator, as shown in a formula (31), and the local optimizing capability of an algorithm is improved.
And (3) performing elite reverse learning on the hunting matters in the final stage, and fusing the cauchy variation to avoid the algorithm from sinking into a local optimal solution prematurely.
(5) Finally, the ocean eddy and fish gathering phenomenon (FADs) are regarded as the local optimal solution of the search space, and the predators complete the position update with larger amplitude according to the update mechanism of the formula (32), so that the phenomenon of sinking into the local optimal solution is further avoided.
Wherein R is a binary number set randomly generated from a binary vector; r is a random number between 0 and 1; PF is a disturbance probability coefficient generated by FADs, and the value is 0.2; l (L) s1 And L s2 Is a random individual in the hunter population;
in addition, the ocean memory is updated for individuals of the top predator matrix E, and when the fitness value of the prey matrix L is calculated, if the optimal fitness value is changed, the corresponding individuals in the original top predator matrix are replaced.
(6) Judging whether the maximum iteration times are reached, if so, outputting an optimal result of the algorithm to obtain the minimum running cost of the multi-micro-grid system.
Based on four scenes, the optimal running cost CMG of each micro-grid system is obtained by utilizing HIMPA solution, and the scheduling optimization results of the grid-connected multi-micro-grid system under different scenes are subjected to economic analysis, wherein the results are shown in Table 7. In addition, CMMG in scenario 3 and scenario 4 takes into account the cost of compensation for load outages and analyzes the power trade of the multi-microgrid system with the main grid.
As can be seen from table 7, in the scheduling optimization scheme adopted in the present invention and shown in scenario 4, compared with scenario 1, the running cost of each micro-grid system is the lowest, the total running cost is 84488.28RMB, and the reduction range is 11.19%. Meanwhile, the total power purchase power is 34738.45kW, and the reduction amplitude is 13.22%. The total electricity selling power is 4244.52kW, and the reduction amplitude is 32.96%. Furthermore, compared to scenario 1, scenario 2 and scenario 3 have reduced the overall running cost of the system by 1.45% and 10.18%, respectively.
Table 7 results in different scenarios
the total amount of supply and demand in the multi-microgrid system is shown in formula (33), and the supply and demand ratio and demand supply are shown in formula (34):
wherein P is sell (t) and P buy (t) the total energy supply amount and the total demand amount of the multi-micro-grid system at the time t respectively;
wherein A (t) and B (t) are respectively the supply-demand ratio and the demand-supply ratio of the multi-micro-grid system at the moment t;
(1) When the supply and demand relationship does not exist in the micro-grid system, the internal electricity price is the same as the electricity price set by the main grid, and d is the same as the electricity price set by the main grid sell (t) and d buy (t) are all 0, and the internal electricity price is as shown in formula (35)Showing;
(2) When the supply and demand are balanced, the internal electricity price should be set at the intermediate price, and d sell (t) and d buy (t) equal internal electricity prices are as shown in formula (36):
(3) D when the total energy supply of the micro-grid system is smaller than the total demand sell (t) is less than d buy (t) the internal electricity price is represented by formula (37):
d when the total energy supply of the micro-grid system is greater than the total demand sell (t) is greater than d buy (t) the internal electricity price is represented by formula (38):
In conclusion, the internal electric energy interaction and the demand response can effectively reduce the running cost of the multi-micro-grid system, achieve the purpose of peak clipping and valley filling, and verify the effectiveness of the system scheduling optimization scheme designed by the invention. Meanwhile, the power supply pressure of the main power grid can be relieved by reducing the electricity purchasing power, and the impact of the multi-micro-grid system on the main power grid can be reduced by reducing the electricity selling power, so that the coordinated operation of the multi-micro-grid system and the main power grid is ensured.
The foregoing is illustrative of the present invention, but the scope of the invention is not limited to this example, and every person skilled in the art is within the scope of the present invention, and needs to be equivalently replaced or changed according to the technical scheme and the inventive concept of the present invention, and should fall within the scope of the present invention.
Claims (8)
1. A grid-connected multi-microgrid system operation scheduling optimization method considering electric energy interaction and demand response is characterized by comprising the following steps of: the method comprises the following steps:
step 1, setting upper and lower limits of equipment parameters and constraint conditions of a multi-micro-grid system, determining renewable energy output and load demand sources, and establishing an operation cost model of the multi-micro-grid system; setting up a grid-connected multi-micro-grid system dispatching optimization model according to the upper limit and the lower limit of constraint conditions by taking the installed quantity of the equipment and the upper limit value of rated power in each micro-grid system as an optimization variable of an optimization objective function and taking the running cost of the minimized multi-micro-grid system as the objective function;
step 2, dividing peak-flat-valley time periods of internal loads of each grid-connected multi-microgrid system based on a fuzzy clustering method, classifying power loads in a user load layer, and obtaining mathematical models of different power load demand responses according to different power loads through a demand response strategy;
step 3, improving the marine predator algorithm MPA to obtain an improved marine predator algorithm HIMPA;
step 4, solving the minimum running cost of the multi-micro-grid system by using an improved marine predator algorithm to obtain an optimal result;
Step 5, analyzing the optimal result obtained in the step 4, namely the minimum running cost of the multi-micro-grid system, judging whether the value meets various constraints, if not, defining the value as an abnormal value, readjusting the upper limit and the lower limit of the constraint condition, returning to the step 4, and calculating again by using the improved marine predator algorithm until the abnormal value does not appear any more;
step 6, when the obtained optimal result has no abnormal value, executing a pricing strategy and completing the electric energy transaction in each micro-grid system;
step 7, after internal electric energy dispatching is carried out according to each micro-grid system, an internal optimization result is obtained, and whether each micro-grid system needs to carry out electric energy transaction with a main power grid or not for meeting the self electric power demand or whether a diesel generator is started in the micro-grid system to carry out electric power supplement is determined according to the result;
step 8, judging the final result of the optimized scheduling, if the operation cost can be effectively reduced, storing the related result, if the expected target is not reached, readjusting the upper limit and the lower limit of the constraint condition, and returning to the step 4;
step 9, judging whether the dispatching operation strategy of the multi-micro-grid system meets the iterative termination condition, namely whether the operation time T reaches the daily operation period T, and if so, outputting a final result; if the stopping condition is not met, updating the time period, re-entering the step 4, repeatedly iterating the operation solution until the operation time T reaches the daily operation period T, and finally ending the operation.
2. The grid-connected multi-microgrid system operation scheduling optimization method considering power interaction and demand response according to claim 1, wherein the method comprises the following steps of: in the step 1, in the operation cost model of the multi-microgrid system, the operation cost of the multi-microgrid system includes investment operation cost of equipment, electric energy transaction cost with other microgrid systems, electric energy transaction cost with a main power grid, fuel cost, pollutant treatment cost and punishment cost, and the modeling of each equipment cost is as follows:
(1) Mathematical models of investment, operation and maintenance costs of equipment are as follows (1):
wherein X is 1 Representing the total cost of investment, operation and maintenance of the equipment; n (N) E The number of kinds of distributed power supplies; p (P) N,e, Representing the installed capacity corresponding to each distributed power supply; f (u, v) is equipment depreciation cost in the micro-grid; u is the damage rate of the equipment; v is the equipment lifetime; x is X I,e ,X Ⅱ,e And X Ⅲ,e Investment, operation maintenance and equipment replacement costs of the micro-grid system are respectively; t (T) year Is an annual operating period;
(2) The mathematical model of the electric energy transaction cost of the micro-grid and the main power grid is as follows (2):
wherein X is 2 Representing the electric energy transaction cost of the micro-grid and the main power grid; p (P) BUY,1 (t) and P SELL,1 (t) representing the electric energy purchase and sale transaction amount between the micro-grid and the main grid at the time t respectively; d, d BUY,1 (t) and d SELL,1 (t) represents the price of the electric energy purchase and sale transaction between the micro-grid and the main grid at time t, respectively; wherein subscript 1 represents the power interaction between the microgrid system and the main grid;
(3) Mathematical models of the electric energy transaction costs of the micro-grid and other micro-grid systems are as follows (3):
wherein X is 3 Representing the electric energy transaction cost of the micro-grid and the main power grid; p (P) BUY,2 (t) and P SELL,2 (t) respectively representing the electric energy purchase and sale transaction amounts of the micro-grid and the main power grid at the moment t; d, d BUY,2 (t) and d SELL,2 (t) represents the electric energy purchase and sale transaction prices of the micro-grid and other micro-grids at the time t respectively; wherein subscript 2 represents power interactions inside the micro-grid system;
(4) Mathematical model of total fuel cost for a multi-microgrid system device is given by formula (4):
wherein X is 4 Representing the total fuel cost of the micro-grid system equipment; u (u) fuel Is the price of diesel; f (F) DG (t) is the diesel consumption of the diesel generator in operation at time t; n (N) DG The number of the diesel generators is the number of the diesel generators; p (P) N,DG The rated power of the Nth diesel generator is represented; p (P) DG Representing the actual power of the diesel generator at the time t; u (u) 1 And u 2 Is the fuel curve factor constant;
(5) Mathematical models of the cost of treatment of contaminants in a multi-microgrid system are given by formula (5):
wherein X is 5 Indicating the total cost of treatment of the contaminants; c μ Is the unit cost of different pollutants; lambda (lambda) DG,μ The emission coefficient of the diesel generator; lambda (lambda) GRID,μ The power purchase coefficient is the main power grid; subscript μ is the type of contaminant;
(6) The mathematical model of penalty cost due to shortage or waste of electric power is as follows (6):
wherein X is 6 Represents penalty cost due to shortage or waste of electric power; p (P) vacancy (t) represents a shortage amount of electric power; p (P) waste (t) is the amount of wasted power; w is the unit penalty cost.
3. The grid-connected multi-microgrid system operation scheduling optimization method considering power interaction and demand response according to claim 2, wherein the method comprises the following steps of: in the step 1, constraint conditions in the multi-microgrid system comprise balance constraint of electric load, output constraint energy storage of equipment, capacity constraint of a storage battery and transaction constraint of the microgrid system and other microgrid systems or a main power grid;
(1) Establishing optimization variables of a multi-micro-grid system optimization model:
in a grid-connected multi-microgrid system, rated power of each microgrid diesel generator and the number of diesel generators are used; the power corresponding to each device in the micro-grid system; the output power of the wind power generation system and the number of wind power generation devices; the output power of the photovoltaic power generation system and the number of photovoltaic power generation devices; the charge and discharge power of the storage battery in the micro-grid system and the quantity of the storage battery are used as optimization variables;
Wherein P is N,DG Represents the N < th DG Rated power of the diesel generator; n (N) DG The total number of the diesel generators; p (P) N,e, Representing the power corresponding to each device in the micro-grid system; n (N) E The number of kinds of distributed power supplies; p (P) WT The output power of the wind power generation system; p (P) PV The output power of the photovoltaic power generation system; p (P) c ，P f The charging and discharging power of the storage battery are respectively; e (E) BAT Is the energy storage capacity of the storage battery; n (N) BAT The number of the storage batteries; n (N) DG Is the number of gas turbines; n (N) PV The number of photovoltaic power generation devices; n (N) WT Is wind powerThe number of power generation devices;
(2) Establishing an objective function of a multi-micro-grid system optimization model:
establishing an objective function in the aspect of the running cost of the micro-grid system equipment, wherein the objective function is shown as the following formulas (7) and (8):
X MG,n ＝X 1 +X 2 +X 3 +X 4 +X 5 +X 6 (8)
wherein N is MG Is the number of micro-grid systems; x is X MG,n The total running cost of the micro-grid system;
(3) Establishing constraint conditions of a multi-microgrid system optimization model:
the mathematical model of the balance constraint of the electrical load is shown in the following formula (9):
wherein P is WT (t) is the output power of the wind power generation system at the moment t; p (P) PV (t) is the output power of the photovoltaic power generation system at the moment t; p (P) c (t)、P f (t) the charge and discharge power of the storage battery at the moment t respectively; l (L) Load (t) is the electric load demand of the micro-grid system in implementing load transfer at the moment t;
The mathematical model of the energy storage constraint of the storage battery is shown as the following formula (10):
in the method, in the process of the invention,
is the minimum energy storage capacity of the storage battery; />
The maximum energy storage capacity of the storage battery; e (E) BAT (t) is the residual electric energy of the storage battery at the moment t; p (P) r (t) is the charging power of the storage battery at the time t; />
And->
Respectively the minimum and maximum charging power of the storage battery; p (P) f (t) is the discharge power of the storage battery at the time t; />
And->
Respectively the minimum and maximum discharge power of the storage battery;
the mathematical model of the force constraint of the device is shown in the following formula (11):
in the method, in the process of the invention,
and->
The minimum output of the wind power generation system, the photovoltaic power generation system and the diesel generator are respectively;
and->
Respectively wind power generation system and lightMaximum output of a diesel generator of the photovoltaic power generation system;
the mathematical model of the transaction constraints of the microgrid system with other microgrid systems is shown in the following equation (12):
in the method, in the process of the invention,
and->
The minimum and maximum electricity purchasing amounts of the micro-grid system to the nth micro-grid system are respectively; />
And->
The minimum and maximum sales power of the micro-grid system to the nth micro-grid system are respectively;
the mathematical model of the trade constraint of the micro-grid system and the main grid is shown in the following formula (13):
in the method, in the process of the invention,
and->
The minimum and maximum electricity purchasing quantity of the micro-grid system to the main grid system is respectively; / >
And
the minimum and maximum sales power of the micro-grid system to the main power grid system are respectively;
the number constraint of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery in the multi-microgrid system is shown as the following (14)
In the method, in the process of the invention,
and->
The minimum configuration quantity of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery is respectively; n (N) WT 、N PV 、N GT 、N BAT The configuration quantity of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery is respectively; />
The maximum configuration number of the wind power generation device, the photovoltaic power generation device, the gas turbine and the storage battery is respectively.
4. The grid-connected multi-microgrid system operation scheduling optimization method considering power interaction and demand response according to claim 3, wherein the method comprises the following steps of: the step 2 includes the following steps:
(1) Dividing the power load into peak-flat-valley periods by adopting a fuzzy clustering method, wherein the total interval value of each period is not less than 6 hours, and the peak period in the valley period is not more than 2 hours;
(2) If the renewable energy production during peak and valley periods is able to meet the power demand, no load response will occur; if the power demand is not met, the remaining power load will take a demand response with the renewable energy fully absorbed.
5. The grid-connected multi-microgrid system operation scheduling optimization method considering power interaction and demand response according to claim 4, wherein the method comprises the following steps of: the strategies of the demand response are as follows: according to the importance of the power load and whether the power load has schedulability, the power load in the micro-grid system is divided into a non-response load, a transferable load and an interruptible load; for the interruptible load, excitation demand response is adopted, namely, when the load is in a peak period of electricity consumption or in an emergency, the power generation side actively stops supplying power to a user, and the power generation side needs to carry out economic compensation on the user; for transferable loads, price demand response is adopted, an electricity price elastic matrix is determined, and initial transfer of the power loads is completed;
wherein, the electricity price elastic matrix is shown in formulas (14), (15) and (16):
the self-elasticity coefficient in the price elasticity matrix Q is represented by the following formula (15):
the cross elastic coefficient in the price elastic matrix Q is represented by the following formula (16):
wherein P is i Representing the amount of charge before a response is required during the i period; ΔP i Representing the load variation after the response is required in the i time period; e, e j Indicating the electricity price before the response is required in the j time period; Δe j Indicating the change amount of electricity price after the response is required in the j time period;
The price elastic matrix Q is represented by the following formula (17):
wherein Q is a price elastic matrix, and f, p and g respectively represent peak, flat and valley periods in a time-sharing electricity price mechanism;
and the electricity price change amount after demand response is shown in the following formula (18):
wherein P is f 、P p And P g Respectively considering the average load quantity of each period after price type demand response; p (P) f,0 、P p,0 And P g,0 The average load quantity of each period before price type demand response is respectively; Δe f 、Δe p And Δe g Respectively representing the electricity price change quantity of each period after the demand response; e, e f 、e p And e g The electricity price before time-sharing electricity price is adopted is represented;
the formula of the total load demand is shown in the following formula (19);
L Load (t)＝L Load,I (t)+L Load,trans (t)+L Load,break (t) (19)
wherein L is Load,I (t)、L Load,trans (t) and L Load,break (t) the demand of unresponsive load, transferable load and interruptible load inside the microgrid system at time t respectively;
the mathematical model of the load demand is shown in formula (20):
wherein L is after (t) and L before (t) electrical load demand before and after load transfer and load interruption, respectively; ΔL II,trans (t) is the load actually transferred; ΔL III,break (t) is the actual interrupt load; sigma (sigma) t',t =1 represents that the transferable load is transferred from time t' to time; sigma (sigma) t',t -1 represents a transferable load transfer from time t to time t'; lambda (lambda) t =1 indicates a load interruption, λ t =0 indicates no load interruption.
6. The grid-connected multi-microgrid system operation scheduling optimization method considering power interaction and demand response according to claim 5, wherein the method comprises the following steps of: in the step 3, based on the marine predator algorithm, aiming at the problem of different speed ratios in the algorithm, disturbance type Tent chaotic mapping, dynamic inertia coefficient, differential evolution combined with elite reservation and elite reverse learning combined with cauchy variation are adopted to improve the algorithm, so that an improved marine predator algorithm HIMPA is obtained;
(1) Disturbance type Tent chaotic map
The disturbance type Tent chaotic map is shown as a formula (21) and is used for improving the spatial distribution of the marine environment;
in which x is i Representing the ith chaos number; a is a chaotic variable, where a e (0, 1);
meanwhile, a random out-of-range processing strategy is adopted to solve the problem of out-of-range search, as shown in a formula (22);
wherein i and j represent search space ranges of individuals in the population; l (L) s Representing a prey matrix; l (L) s,0 Representing an initial prey matrix;
(2) Coefficient of dynamic inertia
Dynamic inertia coefficients, as shown in equation (23), for increasing the ability of local searches;
wherein beta is 1 And beta 2 Is an adjustment coefficient;
(3) Differential evolution combining elite retention
In the unit speed ratio, three different individuals in the population are randomly selected after the hunting population completes the Lewy movement or the Brownian movement, the variation of the population individuals is realized by adopting differential evolution and combining with a Cauchy mutation operator, as shown in a formula (24),
Wherein q1, q2 and q3 are three random positions within the population size;
meanwhile, in order to prevent individuals from exceeding the boundary range after mutation, a random out-of-range processing strategy is adopted; after individual variation, retaining individuals with better fitness values by adopting elite strategies;
(4) Elite reverse learning with fused cauchy variation
Reverse learning is carried out on the hunting matters in the final stage, and the cauchy variation is fused, as shown in a formula (25);
wherein PR is PR s Population individuals generated after elite reverse learning; PC (personal computer) s Are individuals of the population generated after cauchy mutation; e is the reverse learned exchange coefficient.
7. The grid-connected multi-microgrid system operation scheduling optimization method considering power interaction and demand response according to claim 6, wherein the method comprises the following steps of: the step 4 comprises the following steps:
(1) Generating an initial prey matrix L with population number S and dimension D 0 Obtaining a top predator matrix E according to the fitness value result, and finally finishing population initialization of the marine environment;
adopting disturbance type Tent chaotic mapping to improve the spatial distribution of the marine environment and obtain an initial prey matrix P 0 As shown in formula (26):
wherein s is population number; d is the population dimension; x is X ij Representing a prey individual;
For each hunting subject X ij Calculating the fitness value, and finding the most suitable individual with the fitness as Y ij Forming an initial top predator matrix E 0 As shown in formula (27);
(2) Dividing the optimizing process into a high speed ratio H < H/3, a unit speed ratio H/3< H <2H/3 and a low speed ratio 2H/3< H < H according to the current iteration times and the maximum iteration times Hmax;
in a high speed ratio H < H/3, i.e. the top predator has a much lower speed than the prey itself in the initial stages of the iteration; enhancing the early local search capability by introducing dynamic inertia coefficients, allows the prey to perform Brownian motion for a better global search while the predator is maintained at the current individual location, as shown in formula (28) below
Wherein Ls and E s A prey matrix and a top predator matrix, respectively; p is a constant; step s Is the hunting movement step length; RB is Brownian operator; beta 1 And beta 2 The adjustment coefficient is Hmax is the maximum iteration number, and h is the current iteration number;
(3) In the unit speed ratio H/3< H <2H/3, the hunting population is divided into two parts, wherein one part of hunting species performs the Lewy motion, and the other part of hunting species performs the Brownian motion, so that the development and the exploration of fitness values in a search space are respectively completed, as shown in the formula (29) and the formula (30):
C in the formula F Is an adaptive coefficient; p is a constant; as the number of iterations changes; RL is the Levy operator; step s Is the hunting movement step length;
after the hunter population finishes the Lewy movement or the Brownian movement, three different individuals in the population are randomly selected, the variation of the individuals in the population is realized by adopting differential evolution and combining a Cauchy mutation operator, and simultaneously, individuals with better fitness values are reserved by adopting elite strategies;
(4) In the low speed ratio of 2H/3< H < H, namely the self speed of the top predator is far higher than that of the prey in the later period of iteration, the position update of the prey is avoided being caught according to the Lewy track of the top predator, as shown in a formula (31);
reverse learning is carried out on the hunting matters in the final stage, and the cauchy variation is fused;
(5) Regarding ocean vortexes and fish gathering phenomenon FADs as local optimal solutions of the search space, and finishing position updating with larger amplitude by predators according to an updating mechanism of a formula (32);
wherein R is a binary number set randomly generated from a binary vector; r is a random number between 0 and 1; PF is the disturbance probability coefficient generated by FADs; l (L) s1 And L s2 Is a random individual in the hunter population;
in addition, the ocean memory is updated for individuals of the top predator matrix E, and when the fitness value of the prey matrix L is calculated, if the optimal fitness value is changed, the corresponding individuals in the original top predator matrix are replaced;
(6) Judging whether the maximum iteration times are reached, and outputting an optimal result of the algorithm if the maximum iteration times are met, so as to obtain the minimum running cost of the multi-microgrid system.
8. The grid-connected multi-microgrid system operation scheduling optimization method considering power interaction and demand response according to claim 7, wherein the method comprises the following steps of: in the step 6, the total amount of supply and demand in the multi-micro network system is shown in formula (33), and the supply-demand ratio and the demand supply are shown in formula (34):
wherein P is sell (t) and P buy (t) the total energy supply amount and the total demand amount of the multi-micro-grid system at the time t respectively;
wherein A (t) and B (t) are respectively the supply-demand ratio and the demand-supply ratio of the multi-micro-grid system at the moment t;
(1) When the supply and demand relationship does not exist in the micro-grid system, the internal electricity price is the same as the electricity price set by the main grid, and d is the same as the electricity price set by the main grid sell (t) and d buy (t) are all 0, and the internal electricity price is shown as a formula (35);
(2) When the supply and demand are balanced, the internal electricity price should be set at the intermediate price, and d sell (t) and d buy (t) equal internal electricity prices are as shown in formula (36):
(3) D when the total energy supply of the micro-grid system is smaller than the total demand sell (t) is less than d buy (t) the internal electricity price is represented by formula (37):
d when the total energy supply of the micro-grid system is greater than the total demand sell (t) is greater than d buy (t) the internal electricity price is represented by formula (38):
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310048537.4A CN116031935A (en) | 2023-01-31 | 2023-01-31 | Grid-connected multi-microgrid system operation scheduling optimization method considering electric energy interaction and demand response |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310048537.4A CN116031935A (en) | 2023-01-31 | 2023-01-31 | Grid-connected multi-microgrid system operation scheduling optimization method considering electric energy interaction and demand response |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116031935A true CN116031935A (en) | 2023-04-28 |
Family
ID=86080944
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310048537.4A Pending CN116031935A (en) | 2023-01-31 | 2023-01-31 | Grid-connected multi-microgrid system operation scheduling optimization method considering electric energy interaction and demand response |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116031935A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117498353A (en) * | 2024-01-03 | 2024-02-02 | 国网浙江省电力有限公司金华供电公司 | Voltage support adjustment method and system for new energy station grid-connected system |
-
2023
- 2023-01-31 CN CN202310048537.4A patent/CN116031935A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117498353A (en) * | 2024-01-03 | 2024-02-02 | 国网浙江省电力有限公司金华供电公司 | Voltage support adjustment method and system for new energy station grid-connected system |
| CN117498353B (en) * | 2024-01-03 | 2024-03-05 | 国网浙江省电力有限公司金华供电公司 | Voltage support adjustment method and system for new energy station grid-connected system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105703369B (en) | Optimal energy flow modeling and solving method for multi-energy coupling transmission and distribution network | |
| CN112529256B (en) | Multi-uncertainty-considered distributed power supply cluster day-ahead scheduling method and system | |
| CN108734350A (en) | A kind of independent method for solving with combined dispatching of the power distribution network containing micro-capacitance sensor | |
| CN110659830A (en) | Multi-energy micro-grid planning method for comprehensive energy system | |
| Chen et al. | A robust optimization framework for energy management of CCHP users with integrated demand response in electricity market | |
| CN108347062A (en) | Microgrid energy based on gesture game manages distributed multiple target Cooperative Optimization Algorithm | |
| CN107464010A (en) | A kind of virtual plant capacity configuration optimizing method | |
| CN109884888B (en) | Multi-building micro-grid model prediction regulation and control method based on non-cooperative game | |
| Iqbal et al. | Optimal scheduling of grid transactive home demand responsive appliances using polar bear optimization algorithm | |
| CN114069678A (en) | Light storage direct current micro-grid energy scheduling method considering energy storage degradation cost | |
| CN112068436A (en) | Layered and distributed control method and system for comprehensive energy system of industrial park | |
| CN109167347A (en) | Based on the adaptive population multiple target electric car charge and discharge Optimization Scheduling of cloud | |
| CN111353128A (en) | Multi-energy hub optimization operation method based on non-cooperative game | |
| CN111224432B (en) | Micro-grid optimal scheduling method and device | |
| CN112734116A (en) | Optimal scheduling method for active power distribution network containing distributed energy storage | |
| CN116031935A (en) | Grid-connected multi-microgrid system operation scheduling optimization method considering electric energy interaction and demand response | |
| Yuanyuan et al. | Optimization scheduling method of power grid energy-saving based on fuzzy decision | |
| CN114970191A (en) | Power grid traffic system day-ahead distribution robust scheduling method based on potential game | |
| TWI639962B (en) | Particle Swarm Optimization Fuzzy Logic Control Charging Method Applied to Smart Grid | |
| CN112541612B (en) | Multi-network coupling-based virtual power plant economic optimization scheduling method | |
| CN110504684B (en) | Day-ahead optimal scheduling method for regional multi-microgrid system | |
| Zhang et al. | Research on economic optimal dispatching of microgrid cluster based on improved butterfly optimization algorithm | |
| CN110766285A (en) | Day-ahead energy scheduling method based on virtual power plant | |
| Taniguchi et al. | Decentralized trading and demand side response in inter-intelligent renewable energy network | |
| CN115796533A (en) | Virtual power plant double-layer optimization scheduling method and device considering clean energy consumption |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |