Micro-grid scheduling method based on hybrid genetic particle swarm optimization algorithm

Abstract

The invention provides a micro-grid dispatching method based on a hybrid genetic particle swarm optimization algorithm. The method comprises the following steps: constructing a micro-grid system comprising a photovoltaic cell PV, a fan WT, a diesel generator DE and a micro gas turbine MT; based on the micro-grid system, a multi-objective optimized micro-grid optimization scheduling model is constructed, an objective function of the optimization scheduling model is set to be cost for minimizing the micro-grid in a grid-connected mode, and the cost comprises operation cost and environment cost; and solving the optimal scheduling model of the micro-grid by adopting a hybrid genetic particle swarm optimization algorithm MOPSO, and outputting an optimization result meeting a scheduling strategy. According to the method, the operation cost is optimized, the microgrid dispatching strategy with the optimal economical efficiency can be formulated by accurately predicting the future load and the new energy output and fully utilizing adjustable and controllable means such as energy storage and the hybrid genetic particle swarm optimization algorithm, and the operation cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrid dispatching, and in particular to a microgrid dispatching method based on a hybrid genetic particle swarm optimization algorithm. Background Art

[0002] Microgrid is a new type of network structure. It is a system unit composed of micro power sources, loads, energy storage systems and control devices. It is an autonomous system that can realize self-protection control and energy management. It can be connected to the external power grid or operate in isolation. It can meet the power supply needs of users as well as the needs of trigeneration of cooling, heating and electricity.

[0003] A microgrid system structure in the prior art is as follows: Figure 1 As shown in Figure 1, microgrid systems usually include devices such as wind power generation, photovoltaic power generation, fuel cells, micro gas turbines, small internal combustion engines and batteries. These devices are coordinated through optimized scheduling strategies to meet the overall optimization goals of the system. Microgrid optimization scheduling refers to the reasonable arrangement of different distributed energy outputs and transmission power between microgrids and the main grid under various constraints of the system, so as to achieve different goals such as low operating costs, low emissions, high reliability and high power generation efficiency. For the demand side, the optimized scheduling of microgrids can effectively reduce the electricity costs of users. For the power supply side, the optimized scheduling of microgrids can improve the stability of the power grid, reduce energy loss in power production and environmental pollution. Therefore, it is of great practical significance to optimize the scheduling of microgrids.

[0004] At present, there have been a lot of studies on the operation and dispatching of microgrids in the prior art. There are existing solutions that use the economic benefits of operation and the peak-shaving and valley-filling effect as optimization targets to build a multi-objective model of a gas-electricity interconnected virtual power plant. There are also solutions that use scheduling targets such as operating costs and power fluctuations to establish a multi-objective model that considers the charging and discharging strategy of electric vehicles, but none of the above solutions consider environmental pollution issues. At the same time, the economic dispatch of microgrids is mostly a multi-constraint, nonlinear optimization problem. The particle swarm algorithm has the advantages of easy implementation, few adjustable parameters, fast convergence speed, and strong optimization ability, so it is more widely used in the field of optimization. There are also solutions that use traditional particle swarm algorithms to solve multi-objective models based on economic costs, environmental costs, and power fluctuations.

[0005] The PSO (Particle Swarm optimization) algorithm in the prior art is prone to fall into the local optimal solution during the search process, which limits its application in complex systems such as microgrid optimization scheduling. In order to overcome this limitation, some researchers have conducted in-depth explorations and proposed innovative solutions in related schemes. Specifically, a scheme in the prior art proposes a multi-objective optimization scheduling model for microgrids in the grid-connected mode that comprehensively considers the operating cost and environmental protection cost of the microgrid system. In order to solve complex multi-objective optimization problems, the model adopts an improved particle swarm algorithm. In the improved particle swarm algorithm, attention is paid to the adjustment of the inertia weight factor and the learning factor. By carefully designing these parameters, the algorithm can maintain a good global exploration ability during the search process, thereby avoiding the dilemma of falling into the local optimal solution. This improvement not only improves the search efficiency of the algorithm, but also makes the solution result closer to the global optimal solution, providing strong support for the optimal scheduling of microgrids.

[0006] The disadvantages of the microgrid dispatching method in the above prior art include:

[0007] (1) The improved inertia weight factor in the prior art is linear attenuation, which has the following main disadvantages.

[0008] ● Lack of dynamic adaptability: The linear inertia weight factor decreases monotonically during the execution of the algorithm. This fixed decreasing method lacks dynamic adaptability to the search requirements at different stages. In some problems, it may be necessary to maintain a larger inertia weight at a certain stage of the search for a wider global search, while at another stage it may be necessary to reduce the inertia weight for a more refined local search. The linear inertia weight factor cannot meet this requirement.

[0009] ● Parameter sensitivity: The performance of the linear inertia weight factor is highly dependent on the selection of its initial value and final value. If the initial value is set too large, the algorithm may diverge too much in the early stage of the search and miss some potential excellent solutions; if the initial value is set too small, the algorithm may fall into a local optimum in the early stage of the search.

[0010] Similarly, the choice of termination value will also affect the performance of the algorithm in the later stages of the search.

[0011] ●Search efficiency problem: Since the linear inertia weight factor is monotonically decreasing, it may become too small in the later stages of the search, causing the particle velocity update to be too dependent on the historical optimal position and the group optimal position, and lacking sufficient exploration. This may cause the algorithm to stagnate in the later stages of the search and fail to find a better solution.

[0012] ● Not suitable for all problems: The linear inertia weight factor is based on a simplifying assumption that the inertia weight should be gradually reduced to strengthen the local search as the search proceeds. However, not all problems meet this assumption. For some problems with multi-peak or complex search spaces, it may be necessary to keep a large inertia weight for a wider search. In this case, the linear inertia weight factor may not be the best choice.

[0013] (2) The individual learning factor c1 and the social learning factor c2 determine the impact of the particle's own experience information and the experience information of other particles on the particle's trajectory, which reflects the information exchange between the particle group. A larger c1 value will cause the particle to search too much in its own local range, and a larger c2 value will cause the particle to converge to the local optimal value too early. The improvement of c1 and c2 in the prior art does not have a good balance between the local and global relationship, and its dynamic change has no significant effect on the improvement of the algorithm. Summary of the invention

[0014] The embodiment of the present invention provides a microgrid scheduling method based on a hybrid genetic particle swarm optimization algorithm to effectively improve the stability of the power grid.

[0015] In order to achieve the above object, the present invention adopts the following technical scheme.

[0016] A microgrid dispatching method based on a hybrid genetic particle swarm optimization algorithm, comprising:

[0017] Construct a microgrid system including photovoltaic cells PV, wind turbines WT, diesel generators DE, and micro gas turbines MT;

[0018] Constructing an optimal dispatching model of a multi-objective optimized microgrid based on the microgrid system, and setting the objective function of the optimal dispatching model to minimize the cost of the microgrid in the grid-connected mode, the cost including operating cost and environmental cost;

[0019] The hybrid genetic particle swarm optimization algorithm MOPSO is used to solve the optimal scheduling model of the microgrid and output the optimization result that meets the scheduling strategy.

[0020] Preferably, the microgrid system comprising photovoltaic cells PV, wind turbines WT, diesel generators DE, and micro gas turbines MT comprises:

[0021] A microgrid system including PV, WT, DE and MT is constructed, and the output power model of the PV is expressed as:

[0022]

[0023] Where: P p 'v is the output active power of the photovoltaic cell, R' pv is the photovoltaic output power under standard test conditions, q' pv is the derating factor of photovoltaic, I' T is the actual solar radiation intensity; I' STC is the solar radiation intensity under standard test conditions; α' p is the temperature coefficient of the PV panel; T′ c is the PV cell temperature at the current time step; T′ stc is the PV cell temperature under standard test;

[0024] The wind speed power characteristic curve expression of the WT is:

[0025]

[0026] Where: P′ WT is the output power of WT; P′ r is the rated power of WT; v′ ci , v′ r and v′ co represents the cut-in wind speed, rated wind speed and cut-out wind speed of WT; a′, b′, c′, d′ are wind speed parameters;

[0027] The expression of the DE model is:

[0028]

[0029] Where: C DEOM (t), C DEF (t), C DEEN (t) are the operation and maintenance cost, fuel cost and pollutant treatment cost of the diesel engine at time t. DE (t) is the power generation of the diesel engine at time t; K DEOM is the operation and maintenance cost coefficient of the diesel engine; γ de,k is the emission of type k pollutants generated by the operation of the diesel engine; C k is the cost coefficient for treating type k pollutants; α, β, γ are the coefficients for diesel engines;

[0030] The expression of MT is:

[0031]

[0032] Where: PMT(t) is the active output power of the micro gas turbine; η MT(t) is the operating efficiency of the micro gas turbine. The gas turbine generates electricity by consuming fuel. During operation, fuel costs, operation and maintenance costs, and pollutant treatment costs will be incurred. The expression is as follows:

[0033]

[0034] Where: C MTOM (t), C MTF (t), C MTEN (t) are the operation and maintenance cost, fuel cost, and pollutant treatment cost of the diesel engine at time t; P MT (t) is the power generation of the diesel engine at time t; K OM is the operation and maintenance cost of the diesel engine; is a constant term, which represents the ratio of total energy capacity to unit energy lower heating value, γ mt,k is the emission of type k pollutants generated by the operation of the diesel engine; C k is the cost coefficient for treating type k pollutants;

[0035] The expression of the battery model is set as:

[0036]

[0037] Where: SOC(t) is the remaining capacity of the battery at time t; P bess (t) is the charging and discharging power of the battery at time t, P bess (t) is positive, indicating charging, P bess (t) is negative, indicating discharge; η + , η ― are the charge and discharge efficiency respectively.

[0038] Preferably, the optimization dispatching model of the microgrid based on the microgrid system for multi-objective optimization is constructed, and the objective function of the optimization dispatching model is set to minimize the cost of the microgrid in the grid-connected mode, and the cost includes operating cost and environmental cost, including:

[0039] Constructing a multi-objective optimized scheduling model of the microgrid based on the microgrid system, and setting the objective function of the optimization scheduling model to minimize the cost of the microgrid system in the grid-connected mode, the cost including the operating cost and the environmental cost;

[0040] (1) The operating cost of the microgrid system is expressed as:

[0041]

[0042] Where: C grid (t), C MT (t), C DE (t) are the total cost of interaction between the microgrid and the main grid, the total operating cost of the micro gas turbine and the total operating cost of the diesel generator during period t; P sell (t), Pbuy (t) are the power sold and purchased by the microgrid and the large grid at time t; C buy (t), c sell (t) are the electricity purchase and sales prices of the microgrid and the large grid at time t;

[0043] (2) The environmental cost of the microgrid system is expressed as:

[0044]

[0045] Where: C GRID,EN (t) is the pollutant treatment cost of the large power grid; γ grid,k is the emission of type k pollutants generated by the operation of the large power grid; C k is the cost coefficient for treating type k pollutants;

[0046] The objective function of the optimal dispatch model of the microgrid system is defined as follows:

[0047] Z=f1+f2

[0048] Where: Z is the total cost of the microgrid;

[0049] The constraints of the optimal dispatch model of the microgrid are set as follows:

[0050] (1) Power balance constraints

[0051] P′ PV (t)+P′ WT (t)+P grid (t)+P DE (t)+P MT (t)+P bess (t) = P L (t)

[0052] P′ PV (t) is the photovoltaic power at time t, P′ WT (t) is the wind speed power at time t, P grid (t) is the power of interaction between the microgrid and the main grid at time t, P DE (t) is the power generated by the diesel engine at time t, P MT (t) is the active output power of the micro gas turbine at time t, P bess (t) is the charging and discharging power of the battery at time t, P L (t) is the load demand power at time t;

[0053] (2) Diesel engine output constraints

[0054]

[0055] (3) Micro gas turbine output constraints

[0056]

[0057] (4) Transmission power constraints of interconnection lines

[0058]

[0059] (5) Energy storage device constraints

[0060]

[0061] Where: are the upper and lower limits of diesel engine output and micro gas turbine output respectively; r DE 、r MT They are the upper limits of the climbing power of diesel engines and micro gas turbines respectively; The upper and lower limits of the transmission power of the tie line; It is the upper and lower limits of the energy storage device output. A positive value indicates power input, and a negative value indicates power output. SOC max (t), SOC min (t) is the upper and lower limits of the energy storage capacity at time t.

[0062] Preferably, the hybrid genetic particle swarm optimization algorithm MOPSO is used to solve the optimal scheduling model of the microgrid, and the optimization result satisfying the scheduling strategy is output, including:

[0063] (1) Specify system parameters and MOPSO algorithm parameters, including parameters of the power generation unit, energy storage device, and load in the microgrid, as well as parameters such as the population size, maximum number of iterations, and acceleration factor of the MOPSO algorithm, and initialize the particle population. Each particle individual corresponds to a microgrid scheduling plan, which includes the operating power of the power generation unit, the charging and discharging status of the energy storage device, and the power exchange strategy between the microgrid and the main grid;

[0064] (2) Each particle individual is input into the simulation model as an independent microgrid optimization scheduling strategy for calculation. The fitness value of each particle individual is determined by the objective function value Z and constraints generated by the microgrid scheduling scheme. If all constraints are not met, the objective function Z is assigned a maximum value.

[0065] (3) Taking individual fitness as the input of the optimization model, the strategy of inertia weight normal distribution attenuation is introduced, the movement strategy of particle position is updated in real time, and the next generation population is obtained by controlling the convergence speed and search direction of particles using the compression factor Φ. The state of each particle is updated by using the particle speed and position update formula, a new microgrid scheduling plan is generated, and the fitness value of each individual particle is continuously improved;

[0066] (4) According to the historical optimal solution pbest of the individual particle and the historical optimal solution gbest of the group, determine whether the solution of the current individual particle is better than its historical optimal solution. If the current solution is better than the historical optimal solution, then the current solution is used as the new historical optimal solution; otherwise, the historical optimal solution is maintained unchanged. For each particle, compare the fitness value of its individual historical optimal solution with the historical optimal solution of other particles, count the number of times it is dominated by other particles, and select the particle with the less number of domination as the new individual historical optimal solution;

[0067] (5) Sort all individual particles in the population according to non-dominated order, calculate the crowding distance of each individual particle, and after comprehensively considering the crowding distance and non-dominated level, store the non-dominated Pareto solution set into the external archive set. When the size of the external archive set reaches the upper limit of the capacity, sort all solutions according to the crowding distance between particles, and select the first Nmax particles with better fitness values ​​and larger crowding distances and add them to the external archive set;

[0068] (6) Select the global optimal solution gbest from the external archive set. Calculate the comprehensive fitness value of each particle according to the non-dominated level and crowding distance of each particle, sort them according to the fitness value, and randomly select one from the first few particles as the global optimal solution;

[0069] (7) Return to step (3) and decide whether to continue the selection process based on the comparison between the current number of iterations and the maximum number of iterations. If the number of iterations reaches the maximum number of iterations, output the final Pareto non-inferior solution. Otherwise, update the speed and position of each particle to find a new individual solution and add mutation operations so that particles have the probability of mutation in each iteration process, increasing the diversity of particles. Then enter the next round of iterations. For each particle, calculate its new speed and position, and update its individual extreme value and global extreme value to the current optimal solution.

[0070] (8) For each solution in the final Pareto non-inferior solution set, further constraint processing is performed to make it meet the constraint conditions of the independent microgrid energy optimization scheduling strategy, and finally the optimization result that meets the scheduling strategy is output. The optimization result includes the power consumption of wind power, photovoltaic power, energy storage device, diesel, main grid, and gas turbine device at different time points under the condition of lowest total cost.

[0071] It can be seen from the technical solutions provided by the above embodiments of the present invention that the method of the present invention optimizes the operating cost. By accurately predicting future loads and new energy output, and making full use of adjustable means such as energy storage, the hybrid genetic particle swarm optimization algorithm can formulate the most economical microgrid scheduling strategy and reduce operating costs. Balance supply and demand: The hybrid genetic particle swarm optimization algorithm can optimize the scheduling of distributed energy within the microgrid according to user needs and grid characteristics, ensure the balance of supply and demand of the grid, and improve the stability of the grid.

[0072] Additional aspects and advantages of the present invention will be given in part in the following description, which will become obvious from the following description, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0074] Figure 1 A microgrid system structure diagram in the prior art;

[0075] Figure 2 A processing flow chart of the threshold MOPSO algorithm provided in an embodiment of the present invention;

[0076] Figure 3 A schematic diagram comparing the effects of a traditional PSO algorithm and a MOPSO algorithm provided in an embodiment of the present invention;

[0077] Figure 4 A schematic diagram of the distribution of Pareto non-inferior solutions of a traditional PSO algorithm and a MOPSO algorithm after 100 iterations provided in an embodiment of the present invention;

[0078] Figure 5-Figure 8 A schematic diagram of wind power generation prediction and actual consumption situation is provided for an embodiment of the present invention, which considers only the lowest operating cost, only the lowest environmental protection cost, a compromise between operating cost and environmental protection cost, and only the total cost.

[0079] Figure 9-12 A schematic diagram of photovoltaic power generation prediction and actual consumption situation is provided for an embodiment of the present invention, which considers only the lowest operating cost, only the lowest environmental protection cost, a compromise between operating cost and environmental protection cost, and only the total cost.

[0080] Fig.13 A schematic diagram of various microgrid distribution situations provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0081] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0082] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or coupling. The term "and / or" used herein includes any unit and all combinations of one or more associated listed items.

[0083] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0084] To facilitate understanding of the embodiments of the present invention, several specific embodiments will be further explained below with reference to the accompanying drawings, and each embodiment does not constitute a limitation on the embodiments of the present invention.

[0085] The microgrid system in the embodiment of the present invention includes clean uncontrollable power generation units such as photovoltaic cells (PV) and wind turbines (WT), clean controllable power generation units such as diesel generators (DE) and micro turbines (MT), and energy storage elements (ES).

[0086] The output power of solar photovoltaic cells PV is related to the light intensity. Photovoltaic cells generally work in the maximum power point tracking (MPPT) mode. In the microgrid optimization scheduling, the power of photovoltaic cells must be predicted. The output power model of photovoltaic cells can be expressed as:

[0087]

[0088] Where: P p ' v is the output active power of the photovoltaic cell, R' pv is the photovoltaic output power under standard test conditions, q' pv is the derating factor of photovoltaic, usually 0.8; I' T is the actual solar radiation intensity; I' STC is the solar radiation intensity under standard test conditions; α' p is the temperature coefficient of the PV panel; T′ c is the PV cell temperature at the current time step; T′ stc is the PV cell temperature under standard test. The above expression of the output power of the PV cell is reflected in the formula of the microgrid power balance constraint.

[0089] The wind speed power characteristic curve expression of the wind turbine WT is:

[0090]

[0091] Where: P′ WT is the output power of WT; P′ r is the rated power of WT; v′ ci , v′ r and v′ co represents the cut-in wind speed, rated wind speed and cut-out wind speed of WT; a′, b′, c′, d′ are wind speed parameters. The above WT wind speed power characteristic curve expression is reflected in the formula of microgrid power balance constraint.

[0092] Diesel engine is a commonly used fuel generator, which will generate fuel costs, operation and maintenance costs, and pollutant treatment costs during operation. The expression of the diesel generator model is:

[0093]

[0094] Where: C DEOM (t), C DEF (t), C DEEN (t) are the operation and maintenance cost, fuel cost and pollutant treatment cost of the diesel engine at time t. DE (t) is the power generation of the diesel engine at time t; K DEOMis the diesel engine operation and maintenance cost coefficient; γ de,k is the emission of type k pollutants generated by the operation of the diesel engine; C k is the cost coefficient for treating type k pollutants; α, β, γ are the coefficients of the diesel engine. The expression of the diesel generator model above is reflected in the formula of the microgrid operation cost and its constraint problem.

[0095] Gas turbines generate electricity by consuming fuel. The output power of micro gas turbines can be freely adjusted and has a fast response speed. The expression of gas turbine model MT is:

[0096]

[0097] Where: P MT (t) is the active output power of the micro gas turbine; η MT (t) is the operating efficiency of the micro gas turbine. The gas turbine consumes fuel to generate electricity, and in the process of operation, it will generate fuel costs, operation and maintenance costs, and pollutant treatment costs, which are expressed as follows:

[0098]

[0099] Where: C MTOM (t), C MTF (t), C MTEN (t) are the operation and maintenance cost, fuel cost, and pollutant treatment cost of the diesel engine at time t; P TM (t) is the power generation of the diesel engine at time t; K OM is the operation and maintenance cost of the diesel engine; Is a constant term that represents the ratio of total energy capacity to the lower heating value of a unit of energy. It gives a benchmark for comparing the efficiency of different energy sources or systems. mt,k is the emission of type k pollutants generated by the operation of the diesel engine; C k is the cost coefficient for treating type k pollutants.

[0100] As a kind of energy storage device, the battery will release the stored energy when renewable energy cannot meet the load, provide users with stable power supply and ensure the safe and stable operation of the system. The expression of the battery model is:

[0101]

[0102] Where: SOC(t) is the remaining capacity of the battery at time t; P bess (t) is the charging and discharging power of the battery at time t, P bess (t) is positive, indicating charging, P bess (t) is negative, indicating discharge; η + , η ―are the charging and discharging efficiencies, respectively. The above expressions of the battery model are used in the expressions of the energy storage device constraints.

[0103] Based on the PV model, WT model, DE model and MT model of the above microgrid system, a multi-objective optimization microgrid optimization scheduling model is constructed. The structural characteristics of the microgrid optimization scheduling model are:

[0104] 1. Modular design: The modular design concept is adopted to model distributed energy systems such as PV, WT, DE and MT as independent modules. This design enables the model to flexibly adapt to microgrid systems of different sizes and types, and also facilitates subsequent maintenance and expansion.

[0105] 2. Multi-objective optimization: The optimization goal of the model is not limited to economic efficiency, but also takes environmental protection into consideration. By setting two objective functions, operating cost and environmental protection cost, the model can be optimized in multiple dimensions to maximize the comprehensive benefits.

[0106] 3. Comprehensive constraints: The model takes into full consideration various constraints when it is constructed, including power balance constraints, equipment operating status constraints, etc. The setting of these constraints ensures that the model can maintain the stable operation of the microgrid system during the optimization process and meet various requirements in actual operation.

[0107] 4. Intelligent scheduling algorithm: The model adopts an intelligent scheduling algorithm, which is a scheduling strategy based on the hybrid genetic particle swarm optimization algorithm.

[0108] The objective function of the optimization scheduling model is set to minimize the cost of the microgrid in the grid-connected mode, which includes operating cost and environmental cost.

[0109] (1) The operating cost of the microgrid is expressed as:

[0110]

[0111]

[0112] Where: C grid (t), C MT (t), C DE (t) are the total cost of interaction between the microgrid and the main grid, the total operating cost of the micro gas turbine and the total operating cost of the diesel generator during period t; P sell (t), P buy (t) are the power sold and purchased by the microgrid and the large grid at time t; c buy (t), c sell (t) are the electricity purchase and sales prices of the microgrid and the large grid at time t.

[0113] (2) The expression of the environmental cost of the microgrid is:

[0114]

[0115] Where: C GRID,EN (t) is the pollutant treatment cost of the large power grid; γ grid,k is the emission of type k pollutants generated by the operation of the large power grid; C k is the cost coefficient for treating type k pollutants.

[0116] The objective function of the microgrid optimization dispatch model is defined as follows:

[0117] Z=f1+f2

[0118] Where: Z is the total cost of the microgrid, which is composed of the sum of the microgrid operation cost and environmental protection cost.

[0119] The constraints of the optimal dispatch model of the microgrid are set as follows:

[0120] (1) Power balance constraints

[0121] P′ PV (t)+P′ WT (t)+P grid (t)+P DE (t)+P MT (t)+P bess (t) = P L (t)

[0122] P′ PV (t) is the photovoltaic power at time t, P′ WT (t) is the wind speed power at time t, P grid (t) is the power of interaction between the microgrid and the main grid at time t, P DE (t) is the power generated by the diesel engine at time t, P MT (t) is the active output power of the micro gas turbine at time t, P bess (t) is the charging and discharging power of the battery at time t, P L (t) is the load demand power at time t.

[0123] (2) Diesel engine output constraints

[0124]

[0125] (3) Micro gas turbine output constraints

[0126]

[0127] (4) Transmission power constraints of interconnection lines

[0128]

[0129] (5) Energy storage device constraints

[0130]

[0131] Where: are the upper and lower limits of diesel engine output and micro gas turbine output respectively; r DE 、r MT They are the upper limits of the climbing power of diesel engines and micro gas turbines respectively; The upper and lower limits of the transmission power of the tie line; It is the upper and lower limits of the energy storage device output. A positive value indicates power input, and a negative value indicates power output. SOC max (t), SOC min (t) is the upper and lower limits of the energy storage capacity at time t.

[0132] The optimal dispatch of microgrid is a typical power system optimization problem, which has the characteristics of high dimension, nonlinearity, and multiple constraints. Compared with algorithms such as genetic algorithms, the PSO algorithm has stronger optimization ability. At the same time, it is easier to apply when solving optimization problems. The present invention adopts MOPSO (Multi-Objective Particle Swarm Optimization, hybrid genetic particle swarm optimization algorithm) to solve the optimal dispatch model of microgrid.

[0133] The MOPSO algorithm is an algorithm developed based on the PSO algorithm for multi-objective optimization problems. Unlike the traditional PSO algorithm, the MOPSO algorithm uses multiple particle swarms, each of which is used to optimize an objective function. These particle swarms converge to multiple Pareto optimal solutions through information sharing. In this way, the MOPSO algorithm can optimize multiple objective functions simultaneously without knowing multiple Pareto frontiers in advance.

[0134] 1) Dynamic inertia factor

[0135] The performance of the PSO algorithm is affected by the choice of parameters. The inertia weight factor of the traditional PSO algorithm is fixed and easy to fall into the local optimal value. To address this shortcoming, the MOPSO algorithm introduces a strategy of normal distribution attenuation of the inertia weight, and a moving strategy that updates the particle position in real time:

[0136]

[0137] Where: w max and w minare the maximum and minimum values ​​of the inertia weight respectively; t is the current iteration number; T is the total iteration number. The δ value of the optimal dynamic attenuation of the inertia weight is 0.443. The improved algorithm w has a faster update speed in the early stage and a slower update speed in the later stage, which makes the particles transition globally quickly in the early stage and converge locally in the later stage, with higher accuracy.

[0138] 2) Compression factor

[0139] In the prior art, it is difficult to balance the global search capability and local search capability of the algorithm by dynamically changing the value of the learning factor. Therefore, the present invention introduces a compression factor Φ to ensure that the algorithm can smoothly converge to the optimal solution by controlling the convergence speed and search direction of the particles.

[0140]

[0141] Among them, C = c1 + c2. So the speed formula is updated to

[0142]

[0143] 3) Mutation operation

[0144] In order to strengthen the local and global search of the algorithm, a mutation operation is added and a mutation link is introduced, so that particles have the probability of mutation in each iteration process, which increases the diversity of particles and improves the accuracy of the algorithm. The specific method is:

[0145]

[0146] in is the optimal position of the population in the kth iteration; ε is a random variable that obeys the standard normal distribution.

[0147] The processing flow of the MOPSO algorithm provided by the embodiment of the present invention is as follows: Figure 2 As shown, the processing steps include the following:

[0148] (1) Initialization: First, the system parameters and MOPSO algorithm parameters need to be specified, including the parameters of the power generation unit, energy storage device, and load in the microgrid, as well as the population size, maximum number of iterations, acceleration factor, and other parameters of the MOPSO algorithm. Next, the particle population needs to be initialized. Each individual particle corresponds to a microgrid scheduling plan, including the operating power of the power generation unit, the charging and discharging status of the energy storage device, and the power exchange strategy between the microgrid and the main grid.

[0149] (2) Calculate the fitness value through the simulation model: Input each particle individual as the optimization scheduling strategy of an independent microgrid into the simulation model for calculation. For each particle individual, its fitness value is determined by the objective function value Z and the constraints generated by the microgrid scheduling scheme. If all constraints are not met, the objective function Z is assigned a maximum value. The main task of this stage is to use the simulation model to calculate and optimize various microgrid systems and assign a fitness value to each particle individual for use in subsequent steps.

[0150] (3) Update the population: Take individual fitness as the input of the optimization model, introduce the strategy of inertia weight normal distribution attenuation, update the movement strategy of particle position in real time, and use the compression factor Φ to control the convergence speed and search direction of the particles to obtain the next generation of population. At this stage, we need to update the state of each particle by using the particle speed and position update formula to generate a new microgrid scheduling plan to continuously improve the fitness value of each individual particle. Specifically, the update process includes adjusting the speed and position of each particle and generating a new population according to the new microgrid scheduling plan.

[0151] (4) Calculate the local extreme value pbest: According to the individual historical optimal solution pbest and the group historical optimal solution gbest, determine whether the solution of the current particle is better than its historical optimal solution. If the current solution is better than the historical optimal solution, the current solution will be used as the new historical optimal solution; otherwise, the historical optimal solution will remain unchanged. At the same time, for each particle, compare the fitness value of its individual historical optimal solution and the historical optimal solution of other particles, count the number of times it is dominated by other particles, and select the one with the least number of domination as the new individual historical optimal solution. The main task of this stage is to calculate the individual extreme value of each particle, that is, the local extreme value. We need to compare the individual extreme value of the current particle with the individual extreme value of the original particle, and determine the optimal solution of each particle based on the domination relationship.

[0152] (5) Storing non-dominated solutions: Sort all individuals in the population by non-dominated order, and then calculate the crowding distance of each individual. The crowding distance reflects the density of the distribution around an individual. The larger the value, the sparser the distribution around the individual. After considering the crowding distance and the non-dominated level, the non-dominated Pareto solution set is stored in the external archive set. At the same time, when the size of the external archive set reaches the upper limit of the capacity, we need to sort all solutions according to the crowding distance between particles, and select the first Nmax particles with better fitness values ​​and larger crowding distances and add them to the external archive set. The purpose of this step is to save the optimal solution in the non-dominated solution set and ensure that the size of the external archive set does not exceed the specified capacity so that it can be used in the next step.

[0153] (6) Select the global optimal solution gbest. Select the global optimal solution gbest from the external archive set. Calculate the comprehensive fitness value of each particle according to the non-dominated level and crowding distance of each particle, sort them according to the fitness value, and then randomly select one from the first few particles as the global optimal solution.

[0154] (7) Return to step (3) and decide whether to continue the selection based on the comparison between the current number of iterations and the maximum number of iterations. If the number of iterations reaches the maximum number of iterations, the final Pareto non-inferior solution is output; otherwise, the speed and position of each particle are updated to find a new individual solution, and mutation operations are added so that particles have the probability of mutation in each iteration process, increasing the diversity of particles, and then entering the next round of iteration. When updating the particle speed and position, the speed and position need to be adjusted according to the current individual extreme value and global extreme value. Specifically, for each particle, its new speed and position are calculated, and its individual extreme value and global extreme value are updated to the current optimal solution.

[0155] (8) At this stage, the energy optimization scheduling strategy of the independent microgrid is a comprehensive process, which involves the prediction and management of distributed power sources, energy storage devices and loads. It aims to achieve reasonable allocation and efficient utilization of resources through the coordinated control of distributed power output and energy storage device charging and discharging, as well as the management of load demand side, so as to reduce costs. We need to apply the independent microgrid energy optimization scheduling strategy, that is, the optimization problem proposed previously, again to ensure that the solution of each optimal individual meets the constraints of the energy scheduling strategy. In this way, we can obtain the optimization result that satisfies the scheduling strategy. For each solution in the final Pareto non-inferior solution set, further constraint processing is required to make it meet the constraints of the independent microgrid energy optimization scheduling strategy, that is, to ensure the feasibility and effectiveness of the scheduling scheme. The final output is the optimization result that meets the scheduling strategy, which includes the power consumption of wind, photovoltaic, energy storage, diesel, main grid, gas turbine and other devices at different time points under the condition of the lowest total cost.

[0156] The microgrid system of the present invention includes various distributed power sources, including photovoltaic, wind, diesel, gas turbine and energy storage. The operating parameters and costs of each distributed power source in the microgrid are shown in Table 1, the pollutant emission coefficient and cost of each distributed power source are shown in Table 2, and the energy storage parameters are shown in Table 3.

[0157] Table 1 Unit parameters

[0158]

[0159] Table 2 Pollutant emission coefficients

[0160]

[0161] Table 3 Energy storage parameters

[0162]

[0163] A schematic diagram comparing the effects of a traditional PSO algorithm and a MOPSO algorithm provided by an embodiment of the present invention is shown in FIG. Figure 3 As shown, from Figure 3 It can be seen that the convergence speed of the traditional PSO algorithm and the MOPSO algorithm is faster than that of the hybrid genetic-particle swarm algorithm, which means that the existing particle swarm algorithm is easy to fall into the local optimal value, while the hybrid genetic-particle swarm algorithm has better global exploration ability. Therefore, the MOPSO algorithm proposed in the present invention is superior to the existing PSO algorithm.

[0164] A schematic diagram of the distribution of Pareto non-inferior solutions of a traditional PSO algorithm and a MOPSO algorithm after 100 iterations provided by an embodiment of the present invention is shown in FIG. Figure 4 As shown. Figure 4 It can be seen that the MOPSO algorithm has simple operation steps and stronger optimization ability when solving multi-objective optimization problems. In addition, the MOPSO algorithm can find a Pareto frontier with a relatively uniform distribution in a relatively short time, which further shows its superiority. The MOPSO algorithm proposed in the present invention has a better distribution of Pareto non-inferior solutions after 100 iterations and is more concentrated.

[0165] Figure 5-Figure 8 A schematic diagram of wind power generation prediction and actual consumption situation provided by an embodiment of the present invention, which considers only the lowest operating cost, only the lowest environmental protection cost, a compromise between operating cost and environmental protection cost, and only the total cost. Figure 5-Figure 8 It can be seen that the demand for wind power generation when only considering the lowest environmental protection cost is significantly less than the two methods of compromising operating costs and environmental protection costs and only considering the total cost. Although reducing the use of wind power generation has reduced costs and reduced environmental pollution to a certain extent, it will lead to increased power supply on other power supply sides and cannot effectively achieve the optimal scheduling of microgrids.

[0166] Figure 9-12 A schematic diagram of photovoltaic power generation prediction and actual consumption situation provided by an embodiment of the present invention, which considers only the lowest operating cost, only the lowest environmental protection cost, a compromise between operating cost and environmental protection cost, and only the total cost. Figure 9-12It can be seen that the demand for photovoltaic power generation is significantly greater than the total cost when only considering the lowest operating cost and the lowest environmental protection cost. Although increasing the use of photovoltaic power generation has reduced costs and reduced environmental pollution to a certain extent, it will lead to increased power supply on other power supply sides and cannot effectively achieve the optimal scheduling of microgrids. In addition, photovoltaic power generation has many restrictions and cannot guarantee stable power supply.

[0167] Fig.13 A schematic diagram of various microgrid distribution situations provided by an embodiment of the present invention, from Fig.13 It can be seen that energy storage plays an auxiliary regulatory role, charging when electricity prices are low and discharging when electricity prices are high, which reduces economic and environmental costs. At the same time, since the total power generation of photovoltaic and wind power cannot meet the load demand of the system, the system can only purchase electricity from the main grid and output electricity from distributed power sources to meet the load demand. When selecting the total cost as the objective function, after comprehensively considering the operating cost and environmental protection cost, MT and DE basically fully meet the load demand. This is because the comprehensive benefits of MT and DE are greater than the comprehensive benefits of purchasing electricity from the main grid.

[0168] In summary, the MOPSO algorithm provided by the present invention can better balance global search and local search, can make full use of the global search capability of the genetic algorithm, and widely search for potential optimal solutions in the solution space; at the same time, through the local search capability of the particle swarm optimization algorithm, the potential optimal solution searched is finely adjusted, thereby improving the accuracy and efficiency of the scheduling strategy and increasing the probability of finding the global optimal solution. The method of the embodiment of the present invention has better global exploration capabilities and improves search accuracy. It can effectively increase the diversity of the population, prevent the algorithm from converging to the local optimal solution too early, and increase the possibility of the algorithm finding the global optimal solution.

[0169] The method of the present invention optimizes the operating cost: by accurately predicting future loads and new energy output, and making full use of adjustable means such as energy storage, the hybrid genetic particle swarm optimization algorithm can formulate the most economical microgrid scheduling strategy and reduce operating costs.

[0170] Balancing supply and demand: The hybrid genetic particle swarm optimization algorithm can optimize the scheduling of distributed energy within the microgrid according to user needs and grid characteristics, ensure the balance of supply and demand of the grid, and improve the stability of the grid.

[0171] Adaptable to various scenarios: The hybrid genetic particle swarm optimization algorithm can be applied to different types of microgrids and different scheduling requirements, and has strong adaptability.

[0172] Easy to expand: With the continuous development and improvement of microgrid technology, the algorithm can interact with other smart devices, such as smart grids and new energy vehicles, to build a smart energy supply system.

[0173] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0174] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention or certain parts of the embodiments.

[0175] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0176] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A microgrid scheduling method based on hybrid genetic particle swarm optimization algorithm, characterized in that: include: Construct a microgrid system including photovoltaic cells PV, wind turbines WT, diesel generators DE, and micro gas turbines MT; Constructing a multi-objective optimized scheduling model of the microgrid based on the microgrid system, and setting the objective function of the optimized scheduling model to minimize the cost of the microgrid in the grid-connected mode, which includes operating cost and environmental cost; The hybrid genetic particle swarm optimization algorithm MOPSO is used to solve the optimal scheduling model of the microgrid and output the optimization result that meets the scheduling strategy.

2. The method according to claim 1, characterized in that The microgrid system comprising photovoltaic cells PV, wind turbines WT, diesel generators DE, and micro gas turbines MT includes: A microgrid system including PV, WT, DE and MT is constructed, and the output power model of the PV is expressed as: Where: P p ' v is the output active power of the photovoltaic cell, R' pv is the photovoltaic output power under standard test conditions, q' pv is the derating factor of photovoltaic, I' T is the actual solar radiation intensity; I' STC is the solar radiation intensity under standard test conditions; α' p is the temperature coefficient of the PV panel; T c ' is the PV cell temperature at the current time step; T s ' tc is the PV cell temperature under standard test; The wind speed power characteristic curve expression of the WT is: Where: P′ WT is the output power of WT; P′ r is the rated power of WT; v′ ci , v′ r and v′ co represents the cut-in wind speed, rated wind speed and cut-out wind speed of WT; a′, b′, c′, d′ are wind speed parameters; The expression of the DE model is: Where: C DEOM (t), C DEF (t), C DEEN (t) are the operation and maintenance cost, fuel cost and pollutant treatment cost of the diesel engine at time t. DE (t) is the power generation of the diesel engine at time t; K DEOM is the diesel engine operation and maintenance cost coefficient; γ de,k is the emission of type k pollutants generated by the operation of the diesel engine; C k is the cost coefficient for treating type k pollutants; α, β, γ are the coefficients for diesel engines; The expression of MT is: Where: PMT(t) is the active output power of the micro gas turbine; η MT(t) is the operating efficiency of the micro gas turbine. The gas turbine generates electricity by consuming fuel. During operation, fuel costs, operation and maintenance costs, and pollutant treatment costs will be incurred. The expression is as follows: Where: C MTOM (t), C MTF (t), C MTEN (t) are the operation and maintenance cost, fuel cost, and pollutant treatment cost of the diesel engine at time t; P MT (t) is the power generation of the diesel engine at time t; K OM is the operation and maintenance cost of the diesel engine; is a constant term, which represents the ratio of total energy capacity to unit energy lower heating value, γ mt,k is the emission of type k pollutants generated by the operation of the diesel engine; C k is the cost coefficient for treating type k pollutants; The expression of the battery model is set as: Where: SOC(t) is the remaining capacity of the battery at time t; P bess (t) is the charging and discharging power of the battery at time t, P bess (t) is positive, indicating charging, P bess (t) is negative, indicating discharge; η + , η ― are the charge and discharge efficiency respectively.

3. The method according to claim 2, characterized in that The optimization dispatching model of the microgrid based on the microgrid system for multi-objective optimization is constructed, and the objective function of the optimization dispatching model is set to minimize the cost of the microgrid in the grid-connected mode, which includes operating cost and environmental cost, including: Constructing an optimal dispatching model of a multi-objective optimized microgrid based on the microgrid system, and setting the objective function of the optimal dispatching model to minimize the cost of the microgrid system in a grid-connected mode, the cost including operating cost and environmental cost; (1) The operating cost of the microgrid system is expressed as: Where: C grid (t), C MT (t), C DE (t) are the total cost of interaction between the microgrid and the main grid, the total operating cost of the micro gas turbine and the total operating cost of the diesel generator during period t; P sell (t), P buy (t) are the power sold and purchased by the microgrid and the large grid at time t; c buy (t), c sell (t) are the electricity purchase and sales prices of the microgrid and the large grid at time t; (2) The environmental cost of the microgrid system is expressed as: Where: C RGID,EN (t) is the pollutant treatment cost of the large power grid; γ grid,k is the emission of type k pollutants generated by the operation of the large power grid; C k is the cost coefficient for treating type k pollutants; The objective function of the optimal dispatch model of the microgrid system is defined as follows: Z=f1+f2 Where: Z is the total cost of the microgrid; The constraints of the optimal dispatch model of the microgrid are set as follows: (1) Power balance constraints P′ PV (t)+P′ WT (t)+P grid (t)+P DE (t)+P MT (t)+P bess (t)=P L (t) P′ PV (t) is the photovoltaic power at time t, P′ WT (t) is the wind speed power at time t, P grid (t) is the power of interaction between the microgrid and the main grid at time t, P DE (t) is the power generated by the diesel engine at time t, P MT (t) is the active output power of the micro gas turbine at time t, P bess (t) is the charging and discharging power of the battery at time t, P L (t) is the load demand power at time t; (2) Diesel engine output constraints (3) Micro gas turbine output constraints (4) Transmission power constraints of interconnection lines (5) Energy storage device constraints Where: are the upper and lower limits of diesel engine output and micro gas turbine output respectively; r DE 、r MT They are the upper limits of the climbing power of diesel engines and micro gas turbines respectively; The upper and lower limits of the transmission power of the tie line; It is the upper and lower limits of the energy storage device output. A positive value indicates power input, and a negative value indicates power output. SOC max (t), SOC min (t) is the upper and lower limits of the energy storage capacity at time t.

4. The method according to claim 3, characterized in that The hybrid genetic particle swarm optimization algorithm MOPSO is used to solve the optimal scheduling model of the microgrid and output the optimization result that meets the scheduling strategy, including: (1) Specify system parameters and MOPSO algorithm parameters, including parameters of the power generation unit, energy storage device, and load in the microgrid, as well as parameters such as the population size, maximum number of iterations, and acceleration factor of the MOPSO algorithm, and initialize the particle population. Each particle individual corresponds to a microgrid scheduling plan, which includes the operating power of the power generation unit, the charging and discharging status of the energy storage device, and the power exchange strategy between the microgrid and the main grid; (2) Each particle individual is input into the simulation model as an independent microgrid optimization scheduling strategy for calculation. The fitness value of each particle individual is determined by the objective function value Z and constraints generated by the microgrid scheduling scheme. If all constraints are not met, the objective function Z is assigned a maximum value. (3) Taking individual fitness as the input of the optimization model, the strategy of inertia weight normal distribution attenuation is introduced, the movement strategy of particle position is updated in real time, and the next generation population is obtained by controlling the convergence speed and search direction of particles using the compression factor Φ. The state of each particle is updated by using the particle speed and position update formula, a new microgrid scheduling plan is generated, and the fitness value of each individual particle is continuously improved; (4) According to the historical optimal solution pbest of the individual particle and the historical optimal solution gbest of the group, determine whether the solution of the current individual particle is better than its historical optimal solution. If the current solution is better than the historical optimal solution, then the current solution is used as the new historical optimal solution; otherwise, the historical optimal solution is maintained unchanged. For each particle, compare the fitness value of its individual historical optimal solution with the historical optimal solution of other particles, count the number of times it is dominated by other particles, and select the particle with the less number of domination as the new individual historical optimal solution; (5) Sort all individual particles in the population according to non-dominated order, calculate the crowding distance of each individual particle, and after comprehensively considering the crowding distance and non-dominated level, store the non-dominated Pareto solution set into the external archive set. When the size of the external archive set reaches the upper limit of the capacity, sort all solutions according to the crowding distance between particles, and select the first Nmax particles with better fitness values ​​and larger crowding distances and add them to the external archive set; (6) Select the global optimal solution gbest from the external archive set. Calculate the comprehensive fitness value of each particle according to the non-dominated level and crowding distance of each particle, sort them according to the fitness value, and randomly select one from the first few particles as the global optimal solution; (7) Return to step (3) and decide whether to continue the selection process based on the comparison between the current number of iterations and the maximum number of iterations. If the number of iterations reaches the maximum number of iterations, output the final Pareto non-inferior solution. Otherwise, update the speed and position of each particle to find a new individual solution and add mutation operations so that particles have the probability of mutation in each iteration process, increasing the diversity of particles. Then enter the next round of iterations. For each particle, calculate its new speed and position, and update its individual extreme value and global extreme value to the current optimal solution. (8) For each solution in the final Pareto non-inferior solution set, further constraint processing is performed to make it meet the constraint conditions of the independent microgrid energy optimization scheduling strategy, and finally the optimization result that meets the scheduling strategy is output. The optimization result includes the power consumption of wind power, photovoltaic power, energy storage device, diesel, main grid, and gas turbine device at different time points under the condition of lowest total cost.

Images