Wind-solar-fuel-storage micro-grid optimization scheduling method and system based on improved particle swarm optimization algorithm

Abstract

The invention relates to the technical field of micro-grid operation control and optimal scheduling, in particular to a wind-light-fuel-storage micro-grid optimal scheduling method and system based on an improved particle swarm optimization algorithm, and the method comprises the steps: constructing a multi-objective optimal scheduling model; solving the multi-objective optimization scheduling model by adopting an improved particle swarm optimization algorithm to obtain an optimal scheduling scheme; on the basis of the optimal scheduling scheme obtained through solving, starting, stopping and output of a gas turbine and a diesel generator in the micro-grid, charging and discharging of an energy storage system and power interaction with a main power grid are controlled, and different optimization scheduling directions are selected for different optimization targets according to the requirements of decision makers. According to the method, the uncertainty of wind and light output and load can be effectively handled, the multi-target, multi-constraint and nonlinear micro-grid optimization scheduling problem is efficiently solved, and economic, environment-friendly and stable operation of the micro-grid is realized.

Description

A method and system for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm. Technical Field

[0001] This invention belongs to the field of microgrid operation control and optimization scheduling technology, specifically relating to an optimization scheduling method and system for wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm. Background Technology

[0002] Microgrids, as an effective form of integrating distributed renewable energy sources (such as wind power and solar power), significantly improve energy utilization efficiency and power supply reliability. Wind-solar-gas-storage microgrids combine the cleanliness of renewable energy, the rapid response and dispatchability of gas turbines, and the energy time-shifting and power support capabilities of energy storage, making them promising for broad applications in the energy supply sector. However, the significant randomness and volatility of wind and solar power output and load demand pose substantial challenges to microgrid dispatching.

[0003] Microgrid dispatching needs to simultaneously consider multiple objectives: operational economy, environmental protection, renewable energy integration rate, etc., and these objectives often conflict with each other. Furthermore, the dispatching process must satisfy a series of complex constraints, including system power balance, unit ramp-up rate limits, energy storage charging and discharging status and capacity constraints, minimum start-up and shutdown time and output limits of gas turbines, and grid interaction power limits. This results in dispatching models typically exhibiting nonlinear, mixed-integer programming problems with high variable dimensionality (involving multiple time periods and multiple unit states).

[0004] Currently, traditional mathematical programming methods (such as mixed-integer linear programming, MILP) have high requirements for model linearization, limited ability to handle strong nonlinearity and uncertainty, and their computational complexity increases sharply with the size of the system. Basic optimization algorithms (such as the standard particle swarm optimization algorithm) are prone to getting trapped in local optima, have slow convergence speeds, and exhibit premature convergence when solving such complex optimization problems. They also struggle to effectively handle a large number of constraints and cannot meet the practical needs of efficient and optimized scheduling in microgrids.

[0005] Therefore, there is an urgent need for a microgrid optimization scheduling method that can take into account multiple objectives, effectively handle complex constraints, and has high solution efficiency. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a wind-solar-gas-storage microgrid optimization scheduling method and system based on an improved particle swarm optimization algorithm. This method effectively addresses the uncertainties in wind and solar power output and load, efficiently solves multi-objective, multi-constraint, and nonlinear microgrid optimization scheduling problems, and achieves economical, environmentally friendly, and stable operation of the microgrid.

[0007] The technical solution of this invention is as follows: A wind-solar-gas-storage microgrid optimization scheduling method based on an improved particle swarm optimization algorithm, comprising the following steps: Step 1: Constructing a multi-objective optimization scheduling model for a wind-solar-gas-storage microgrid including wind power generation, photovoltaic power generation, gas turbines, diesel generators, energy storage systems, and loads, wherein the objectives include minimizing total operating costs and minimizing pollutant emissions; Step 2: Solving the multi-objective optimization scheduling model using an improved particle swarm optimization algorithm to obtain the optimal scheduling scheme, wherein the improvements include an adaptive location update mechanism, a novel elimination mechanism, and a location distance decision strategy, wherein in the novel elimination mechanism, the vacant positions of eliminated particles are randomly selected from the winning particles; Step 3: Based on the optimal scheduling scheme obtained by the solution, controlling the start-up, shutdown, and output of gas turbines and diesel generators in the microgrid, the charging and discharging of the energy storage system, and the power interaction with the main grid, and selecting different optimization scheduling directions for different optimization objectives according to the needs of the decision-maker.

[0008] Furthermore, the total operating cost includes gas turbine fuel cost and start-up and shutdown cost, energy storage operation and maintenance cost, electricity purchase cost or electricity sales revenue from grid interaction, and cost of penalties for wind and solar curtailment; the pollutant emissions include CO2 and NO produced by the gas turbine. X SO X .

[0009] Furthermore, step 2 includes the following steps: Step 21: With the power grid operating in an islanded mode, under the condition of satisfying system constraints, and comprehensively considering the economy, reliability, and environmental protection of the microgrid, a multi-objective optimization scheduling model for the microgrid system with the minimum operating cost and the minimum pollutant emission cost is established, and the parameters are initialized; Step 22: The fitness values ​​of the initial N populations are calculated, i.e., power generation cost and environmental cost, and the local optimum and global optimum are found; Step 23: The velocity update formula and position update formula of the original particle swarm optimization algorithm are improved; Step 24: Based on the new elimination mechanism and position distance decision strategy, the individuals are sorted in descending order according to their fitness values, the worst group of individuals is selected, the elimination mechanism is used to process the individuals, and the position distance decision strategy is used to make the selected particles generate new positions and velocities. After iterating until the termination condition is met, the optimal scheduling scheme is output.

[0010] Furthermore, the constraints of the multi-objective optimization scheduling model include: photovoltaic cell constraints: ;in, For the actual output of photovoltaic cells; Intensity of sunlight; Efficiency in maximum power point tracking mode; The area of ​​the solar panel; For the efficiency of photovoltaic cells; For the incident angle of the photovoltaic image; constraints of the wind turbine generator: ;in, , The actual power and rated power of the wind turbine generator; , , These are the wind turbine's inlet velocity, outlet velocity, and rated velocity; constraints for the micro gas turbine: ; ;in, Fuel costs for micro gas turbines; This represents the price of natural gas; LHV represents the low value of natural gas. This refers to the output power of a micro gas turbine. The efficiency of a micro gas turbine is related to its value. Forming a trigonometric function relationship; diesel generator constraints: ;in, Fuel cost for diesel generators; This refers to the output power of the diesel generator; , , Fuel cost coefficient for diesel generators; energy storage constraints: ;in, , These represent the battery capacities at time t and time t-1, respectively. The total energy storage output at time t; Let t be the total load of the system at time t; These refer to the inverter's operating efficiency and the energy storage's charging and discharging efficiency, respectively.

[0011] Furthermore, the improved original particle swarm optimization algorithm includes velocity update formulas and position update formulas, wherein the velocity update formula is: The position update formula is: ;

[0012] Where G represents the current iteration number, and i represents the current particle. Indicates the reference optimal position. This represents the position of the i-th particle in generation G. This represents the velocity of the i-th particle in generation G. Take 2, Take 1.49445, and And rand represents a uniform distribution following (0,1). The position representing the individual optimal value of a particle. This indicates the position of the most central particle in the population. express The fitness value at this location; This represents the weighting factor.

[0013] Furthermore, the aforementioned The value can be: ;in, This means randomly selecting the positions of 10 particles from the population, choosing the position of the particle with the lowest fitness value. express The fitness value for this location.

[0014] Furthermore, the location update strategy employs the following location update methods, the specific location update method depending on... The expression is as follows: Where ps represents the total number of individual particles in the population. This represents the optimal value for the current individual.

[0015] Furthermore, weighting factor weights It can be expressed by the following formula: ;in, Initialize to a random number; Take 0.4, Take 0.9, This represents the maximum number of iterations; exp represents the weighted average algorithm.

[0016] Furthermore, in the novel elimination mechanism, the formula for generating new positions is: Location-distance decision-making strategy paradigm:

[0017] in, , and represents the maximum and minimum particle values ​​generated under the Cauchy distribution, respectively, which are the global optimal values ​​at the current iteration number.

[0018] An optimized scheduling system for a wind-solar-gas-storage microgrid based on an improved particle swarm optimization algorithm includes a model building module, an optimization solution module, and a scheduling control module. The model building module constructs a multi-objective optimized scheduling model for the wind-solar-gas-storage microgrid, encompassing wind power generation, photovoltaic power generation, gas turbines, energy storage systems, and loads. The objectives include at least minimizing total operating costs and minimizing pollutant emissions. The optimization solution module solves the optimized scheduling model using an improved particle swarm optimization algorithm. The improvements include an adaptive location update mechanism, a novel elimination mechanism, and a location distance decision strategy. In the novel elimination mechanism, the vacant positions of eliminated particles are randomly selected from the winning particles, and the optimal scheduling scheme is obtained through the solution. The scheduling control module, based on the optimal scheduling scheme obtained by the optimization solution module, controls the start-up, shutdown, and output of the gas turbines, the charging and discharging of the energy storage system, and the power interaction with the main grid within the microgrid. Different optimized scheduling directions are selected according to the decision-maker's needs and for different optimization objectives.

[0019] Compared with the prior art, the present invention has the following beneficial effects: (1) Excellent solution performance: By improving the particle swarm optimization algorithm, the shortcomings of the traditional particle swarm optimization algorithm, such as premature convergence and difficulty in handling constraints and discrete variables, are effectively overcome, and the efficiency and accuracy of solving complex nonlinear, high-dimensional, multi-constraint optimization scheduling problems of wind, solar, gas and energy storage microgrids are significantly improved; (2) Multi-objective balance optimization: The constructed mathematical model comprehensively considers operating costs and pollutant emissions. Through the optimization algorithm, different scheduling schemes with different focuses can be selected according to actual needs, providing decision-makers with diversified choices; (3) Strong robustness: By integrating relevant characteristics of scenario analysis, the scheduling scheme has stronger adaptability and robustness to wind and solar power output and load prediction errors, and improves the stability of actual operation of microgrids; (4) Improved energy utilization efficiency: By optimizing energy storage charging and discharging and gas turbine regulation, wind and solar fluctuations are effectively smoothed, wind and solar curtailment is reduced, clean energy is maximized, and the flexibility and economy of microgrid operation are improved. Attached Figure Description

[0020] Figure 1 is a schematic diagram of the overall process of the present invention; Figure 2 is a curve of power allocation of each energy source in the microgrid under the condition of minimum total cost; Figure 3 is a comparison of convergence curves of different algorithms on the f20 test function 30D problem; Figure 4 is a comparison of convergence curves of different algorithms on the f25 test function 30D problem. Detailed Implementation

[0021] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0022] Referring to Figure 1, an optimized scheduling method for a wind-solar-gas-storage microgrid based on an improved particle swarm optimization algorithm includes the following steps: Step 1: Construct a multi-objective optimized scheduling model for a wind-solar-gas-storage microgrid comprising wind power generation, photovoltaic power generation, gas turbines, diesel generators, energy storage systems, and loads. The objectives include minimizing total operating costs and minimizing pollutant emissions. Step 2: Solve the multi-objective optimized scheduling model using an improved particle swarm optimization algorithm to obtain the optimal scheduling scheme. The improvements include an adaptive location update mechanism, a novel elimination mechanism, and a location distance decision strategy. In the novel elimination mechanism, the vacant positions of eliminated particles are randomly selected from the winning particles. Step 3: Based on the optimal scheduling scheme obtained from the solution, control the start-up, shutdown, and output of gas turbines and diesel generators in the microgrid, the charging and discharging of the energy storage system, and the power interaction with the main grid. Different optimized scheduling directions are selected for different optimization objectives according to the needs of the decision-maker.

[0023] The following is a further explanation based on a specific embodiment of the present invention: 1. Construction of microgrid scheduling model for wind, solar and gas storage: Specifically, the present invention proposes a microgrid scheduling optimization method and system based on an improved particle swarm optimization algorithm, and establishes a multi-objective optimization model that comprehensively considers minimizing total operating costs (including: gas turbine fuel cost and start-up and shutdown cost, energy storage operation and maintenance cost, electricity purchase cost or electricity sales revenue from interaction with the main grid, and wind and solar curtailment penalty cost) and minimizing pollutant emissions (such as CO2, NOx, SOx, mainly from gas turbines). The weighted summation method or Pareto optimization method can be used to handle the multi-objective modeling of the following key constraints: Microgrid system structure: The system includes photovoltaic cells (PV), wind turbines and other clean uncontrollable power generation units, diesel generators (DE), micro gas turbines (MT) and other clean controllable power generation units, as well as energy storage units (ES). In the present invention, the microgrid is set to operate in an islanded manner, and the micro power source supplies power to the load inside the microgrid; (1) Photovoltaic cells (PV) and constraints: The output power of photovoltaic cells is related to the light intensity. The maximum power point tracking (MPPT) mode power of photovoltaic cells can be expressed as: ;in, For the actual output of photovoltaic cells; Intensity of sunlight; Efficiency in maximum power point tracking mode; The area of ​​the solar panel; For the efficiency of photovoltaic cells; (2) Wind turbine (WT) and constraints: The output power of the wind turbine is related to the wind speed, and its power output model can be expressed as: ;in, , The actual power and rated power of the wind turbine generator; , , The cut-in wind speed, cut-out wind speed, and rated wind speed of the wind turbine are respectively taken as 3 m / s, 25 m / s, and 14 m / s in this embodiment; (3) Micro gas turbine (MT) and constraints: The fuel cost of the micro gas turbine is related to its working efficiency. The expression for the fuel cost of MT is: ; ;in, Fuel costs for micro gas turbines; For the price of natural gas, this embodiment takes... LHV represents the low value of natural gas; in this embodiment, it is taken as... ; This refers to the output power of a micro gas turbine. The efficiency of a micro gas turbine is related to its value. The relationship is a trigonometric function; (4) Diesel generator (DE) and constraints: The fuel cost of the diesel generator is its consumption characteristic function. The expression for the fuel cost of the diesel generator is: ;in, Fuel cost for diesel generators; This refers to the output power of the diesel generator; , , This patent uses the fuel cost coefficient of a diesel generator. , , (5) Energy Storage (ES) and Constraints: Energy storage can track the changes in wind and solar power output and perform charging and discharging. In the power grid, it plays a role in buffering the uncertain output of wind and solar power, improving the power supply reliability and continuity of the power grid. When the total output power of wind and photovoltaic is greater than the total load, the energy storage charges; otherwise, the energy storage discharges. The charging and discharging state of the energy storage can be expressed as: ;in, , These represent the battery capacities at time t and time t-1, respectively. The total energy storage output at time t; Let t be the total load of the system at time t; These refer to the inverter's operating efficiency and the energy storage's charging and discharging efficiency, respectively.

[0024] 2. Minimizing operating costs as the optimization objective: The generation cost of a microgrid mainly considers the operating cost and the compensation cost for interruptible loads. Therefore, the generation cost should be minimized while satisfying the system equality and inequality constraints. Where: T is the number of time periods in the microgrid's dispatch cycle; N is the number of micro-source types. Let be the power generation cost of the micro-electric park at time t; Let be the power output of the i-th micro-power source at time t; Let be the interruptibility cost of the micro-power source at time t.

[0025] (1) Power generation cost: The operating cost of a microgrid mainly considers the fuel cost, depreciation cost, and maintenance cost of the generator units. Since PV photovoltaic cells and WT wind turbines are clean energy sources and do not consume fossil fuels during operation, the fuel cost of PV photovoltaic cells and WT wind turbines is not considered. (2) Interruptible cost: When a microgrid is in islanded operation, there may be insufficient power supply, requiring the interruption of some non-critical loads to ensure the normal power supply of critical loads. In this embodiment, the strategy for interrupting some interruptible loads is as follows: only when the controllable power source, uncontrollable power source, and energy storage device are all in full power generation state, the load with insufficient power supply is interrupted, and corresponding economic compensation is required for the interrupted load.

[0026] 3. Propose an improved particle swarm optimization algorithm: Improve the velocity and position update formulas of the original particle swarm optimization algorithm to update the individual optimal value and the overall optimal value, as follows: ; Where G represents the current iteration number, and i represents the current particle. This represents the position of the i-th particle in generation G. This represents the velocity of the i-th particle in generation G. Take 2, Take 1.49445, and And rand represents a uniform distribution following (0,1):

[0027] Where, represents the position of the particle with the smallest fitness value among 10 randomly selected particles from the population, represents the position of the particle's individual optimal value, represents the position of the central particle in the population, represents the fitness value of that position; represents the weight factor, whose value is expressed by the following formula: where is initialized to a random number; takes values ​​of 0.4 and 0.9, representing the maximum number of iterations; exp represents the weighted average algorithm.

[0028] The position update strategy employs two position update methods, and the specific method used depends on the population size, where ps represents the total number of individual particles in the population.

[0029] Step 24: To avoid the population easily getting trapped in local optima, an elimination mechanism and position-distance decision strategy are proposed. Individuals are sorted in descending order based on their fitness values, and the worst group of individuals is selected. These individuals are then eliminated using the elimination mechanism, and the position-distance decision strategy is used to make the selected particles generate new positions and velocities. The elimination mechanism and position-distance decision strategy are as follows: Location-distance decision-making strategy paradigm: ;in, , , and These represent the maximum and minimum particle values ​​generated under the Cauchy distribution, respectively. This is the globally optimal value for the current iteration number.

[0030] 4. Optimize the scheduling process by inputting the predicted wind power output curve, photovoltaic power output curve, load demand curve, time-of-use electricity price, fuel price, pollutant emission coefficient, equipment parameters, and constraints for the next day (future 24 hours); set the parameters of the improved particle swarm optimization algorithm, randomly initialize the particle swarm according to the constraints, and output the optimal scheduling scheme according to the improved particle swarm optimization algorithm process, including the following for each future time period: gas turbine start-stop status and output, energy storage charging and discharging status and power, power interaction with the main grid, actual utilization power of wind power and photovoltaic power generation, total system cost, and total emissions.

[0031] Data from the experiments of this invention: MATLAB 7.6 was used for programming. To verify that the above model is better than existing algorithms, the following are the comparative experimental results: Algorithm CCPSO LPSO LPSO proposed algorithm function Mean / StdMean / StdMean / StdMean / StdF 12.09E-13 / 7.66E-14 2.27E-13 / 0 1.38E-13 / 1.12E-13 0 / 0f 51.94E-13 / 1.28E-13 1.37E-11 / 5.63E-12 1.14E-13 / 0 0 / 5.43E-14f 82.09E+01 / 5.91E-02 2.09E+01 / 4.92E-02 2.09E+01 / 5.2E-02 2.09E+01 / 6.06E-02f 12 1.15E+02 / 3.90E+011.27E+02 / 1.54E+011.59E+02 / 9.11E+003.59E+01 / 1.44E+01f 20 1.26E+01 / 9.25E-011.36E+01 / 4.56E-011.31E+01 / 1.56E+001.09E+01 / 5.28E-01f 253.04E+02 / 9.59E+002.94E+02 / 5.22E+002.84E+02 / 1.15E+012.83E+02 / 8.81E+00 The table compares the different algorithms on multiple test functions (f1, f5, f8, f) in the CEC2013 test set. 12 ,f 20 ,f 25 The new algorithm performs best on most functions: f1 and f5. The mean and standard deviation (Mean / Std) of the new algorithm are 0 or close to 0, and the standard deviation is 0 or a minimum value, indicating that it can stably find the global optimum.

[0032] f 12 The mean of the new algorithm (3.59E+01) is significantly lower than that of other algorithms (1.15E+02~1.59E+02), and the standard deviation is smaller, indicating that it is more robust to high-dimensional complex problems.

[0033] f 20 and f 25 The new algorithm achieves optimal mean and standard deviation, further validating its superiority in multimodal problems.

[0034] Referring to Figures 3 and 4, the experimental data shows that the newly proposed particle swarm optimization algorithm can converge to a better value with the same number of iterations. The experimental results are shown in the figure: The figure shows the convergence curves of different algorithms on the 30-dimensional problem. The horizontal axis is the number of function evaluations (nfe / 100), and the vertical axis is the fitness error.

[0035] As can be seen from the curve trend, the proposed new algorithm significantly outperforms other algorithms (PSO, iwPSO, ccpSO, CLPSO, SLPSO) in terms of convergence speed and accuracy. Specifically, it achieves faster convergence by requiring fewer function evaluations to reach a lower fitness error.

[0036] Better solution: The fitness error value at final convergence is lower, indicating that the new algorithm has a greater advantage in the quality of the solution.

[0037] This improvement may be attributed to the elimination mechanism of the location distance decision strategy, which effectively avoids premature convergence while enhancing global search capabilities.

[0038] This invention analyzes a practical microgrid system, using a 24-hour calculation cycle. Typical daily 24-hour load data from a factory area is selected, with a peak load of 90kW. To improve the utilization rate of renewable energy, both PV and WT operate in maximum power point tracking (MPPT) mode. The energy storage device (ES) is a 20kWh battery pack with a maximum charging / discharging power of 5kW, and interruptible loads account for 15% of the total load demand.

[0039] Figure 4 illustrates the power allocation of various energy sources (energy storage, photovoltaic, wind power, grid power, and gas turbines) in the microgrid under the condition of minimum total cost. The scheduling results of the new algorithm have the following characteristics: Economy: The total cost is the lowest, indicating that the algorithm effectively balances economic costs in the objective function. Environmental friendliness: The utilization rate of renewable energy (photovoltaic and wind power) is high, reducing dependence on gas turbines and grid power, which meets environmental emission targets. Stability: The power allocation curve is smooth without drastic fluctuations, indicating that the algorithm can meet power balance and unit constraints. The proposed new algorithm significantly outperforms the traditional particle swarm optimization algorithm and its improved versions in terms of convergence speed, solution quality, robustness, and practical application performance, effectively solving the problem of multi-objective optimization scheduling in microgrids and providing reliable support for the efficient utilization of renewable energy.

[0040] An optimized scheduling system for a wind-solar-gas-storage microgrid based on an improved particle swarm optimization algorithm includes a model building module, an optimization solution module, and a scheduling control module. The model building module constructs a multi-objective optimized scheduling model for the wind-solar-gas-storage microgrid, encompassing wind power generation, photovoltaic power generation, gas turbines, energy storage systems, and loads. The objectives include at least minimizing total operating costs and minimizing pollutant emissions. The optimization solution module solves the optimized scheduling model using an improved particle swarm optimization algorithm. The improvements include an adaptive location update mechanism, a novel elimination mechanism, and a location distance decision strategy. In the novel elimination mechanism, the vacant positions of eliminated particles are randomly selected from the winning particles, and the optimal scheduling scheme is obtained through the solution. The scheduling control module, based on the optimal scheduling scheme obtained by the optimization solution module, controls the start-up, shutdown, and output of the gas turbines, the charging and discharging of the energy storage system, and the power interaction with the main grid within the microgrid. Different optimized scheduling directions are selected according to the decision-maker's needs and for different optimization objectives.

[0041] By constructing a multi-objective optimization scheduling model and improving the particle swarm optimization algorithm, the shortcomings of the particle swarm optimization algorithm, such as premature convergence and difficulty in handling constraints and discrete variables, are effectively overcome. This significantly improves the efficiency and accuracy of solving complex nonlinear, high-dimensional, and multi-constraint optimization scheduling problems in wind-solar-gas-storage microgrids. The constructed mathematical model comprehensively considers operating costs and pollutant emissions. The optimization algorithm can select scheduling schemes with different requirements, providing decision-makers with multiple options. Integrating scenario analysis or robust optimization methods makes the scheduling scheme more adaptable and robust to wind and solar power output and load forecasting errors, improving the stability of actual microgrid operation. By optimizing energy storage charging and discharging and gas turbine regulation, wind and solar fluctuations are effectively mitigated, wind and solar curtailment is reduced, and the utilization of clean energy is maximized. The optimization results can effectively guide the start-up, shutdown, and output of gas turbines, the charging and discharging strategies of energy storage, and energy interaction with the main grid, improving the flexibility and economy of microgrid operation.

[0042] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm, characterized in that, Includes the following steps: Step 1: Construct a multi-objective optimization scheduling model for a wind-solar-gas-storage microgrid, including wind power generation, photovoltaic power generation, gas turbines, diesel generators, energy storage systems, and loads. The objectives include minimizing total operating costs and minimizing pollutant emissions. Step 2: Solve the multi-objective optimization scheduling model using an improved particle swarm optimization algorithm to obtain the optimal scheduling scheme. The improvements include an adaptive location update mechanism, a novel elimination mechanism, and a location distance decision strategy. In the novel elimination mechanism, the vacant positions of eliminated particles are randomly selected from the winning particles. Step 3: Based on the obtained optimal scheduling scheme, control the start-up, shutdown, and output of gas turbines and diesel generators in the microgrid, the charging and discharging of the energy storage system, and the power interaction with the main grid. Different optimization scheduling directions are selected according to the decision-maker's needs and for different optimization objectives.

2. The method for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm according to claim 1, characterized in that, The total operating cost includes gas turbine fuel cost and start-up and shutdown cost, energy storage operation and maintenance cost, electricity purchase cost or electricity sales revenue from the main grid, and cost of penalties for wind and solar curtailment; the pollutant emissions include CO2 and NO produced by the gas turbine. X SO X .

3. The method for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm according to claim 1, characterized in that, Step 2 includes the following steps: Step 21: With the power grid operating in an islanded mode, under the condition of satisfying system constraints, and comprehensively considering the economy, reliability, and environmental protection of the microgrid, a multi-objective optimization scheduling model for the microgrid system with the minimum operating cost and the minimum pollutant emission cost is established, and the parameters are initialized; Step 22: The fitness values ​​of the initial N populations are calculated, i.e., the power generation cost and the environmental cost, and the local optimum and the global optimum are found; Step 23: The velocity update formula and the position update formula of the original particle swarm optimization algorithm are improved; Step 24: Based on the new elimination mechanism and the position distance decision strategy, the individuals are sorted in descending order according to their fitness values, the worst group of individuals is selected, the elimination mechanism is used to process the individuals, and the position distance decision strategy is used to make the selected particles generate new positions and velocities. After iterating until the termination condition is met, the optimal scheduling scheme is output.

4. The method for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm according to claim 3, characterized in that, The constraints of the multi-objective optimization scheduling model include: photovoltaic cell constraints: ;in, For the actual output of photovoltaic cells; Intensity of sunlight; Efficiency in maximum power point tracking mode; The area of ​​the solar panel; For the efficiency of photovoltaic cells; For the incident angle of the photovoltaic image; constraints of the wind turbine generator: ;in, 、 The actual power and rated power of the wind turbine generator; 、 、 These are the wind turbine's inlet velocity, outlet velocity, and rated velocity; constraints for the micro gas turbine: ； ;in, Fuel costs for micro gas turbines; This represents the price of natural gas; LHV represents the low value of natural gas. This refers to the output power of a micro gas turbine. The efficiency of a micro gas turbine is related to its value. Forming a trigonometric function relationship; diesel generator constraints: ;in, Fuel cost for diesel generators; This refers to the output power of the diesel generator; 、 、 Fuel cost coefficient for diesel generators; energy storage constraints: ;in, 、 These represent the battery capacities at time t and time t-1, respectively. The total energy storage output at time t; Let t be the total load of the system at time t; These refer to the inverter's operating efficiency and the energy storage's charging and discharging efficiency, respectively.

5. The method for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm according to claim 3, characterized in that, The improved original particle swarm optimization algorithm includes velocity update formulas and position update formulas. The velocity update formula is as follows: The position update formula is: Where G represents the current iteration number, and i represents the current particle. Indicates the reference optimal position. This represents the position of the i-th particle in generation G. This represents the velocity of the i-th particle in generation G. Take 2, Take 1.49445, and And rand represents a uniform distribution following (0,1). The position representing the individual optimal value of a particle. This indicates the position of the most central particle in the population. express The fitness value at this location; This represents the weighting factor.

6. The method for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm according to claim 5, characterized in that, The The value can be: ;in, This means randomly selecting the positions of 10 particles from the population, choosing the position of the particle with the lowest fitness value. express The fitness value for this location.

7. The method for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm according to claim 3, characterized in that, The location update strategy employs the following location update methods, the specific of which depends on... The expression is as follows: Where ps represents the total number of individual particles in the population. This represents the optimal value for the current individual.

8. The method for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm according to claim 5, characterized in that, The weighting factor is expressed by the following formula: ;in, Initialize to a random number; Take 0.4, Take 0.9, This represents the maximum number of iterations; exp represents the weighted average algorithm.

9. A method for optimal scheduling of wind-solar-gas-storage microgrids based on an improved particle swarm optimization algorithm as described in claim 3, characterized in that, In the novel elimination mechanism, the formula for generating new positions is: Location-distance decision-making strategy paradigm: ； ;in, ， ， and These represent the maximum and minimum particle values ​​generated under the Cauchy distribution, respectively. This is the globally optimal value for the current iteration number.

10. A wind-solar-gas-storage microgrid optimization scheduling system based on an improved particle swarm optimization algorithm, characterized in that, The system includes a model building module, an optimization solution module, and a scheduling control module. The model building module is used to construct a multi-objective optimization scheduling model for a wind-solar-gas-storage microgrid that includes wind power generation, photovoltaic power generation, gas turbines, energy storage systems, and loads. The objectives include at least minimizing total operating costs and minimizing pollutant emissions. The optimization solution module uses an improved particle swarm optimization algorithm to solve the optimization scheduling model. The improvements include an adaptive position update mechanism, a novel elimination mechanism, and a position distance decision strategy. In the novel elimination mechanism, the vacant positions of eliminated particles are randomly selected from the winning particles, and the optimal scheduling scheme is obtained through the solution. The scheduling control module is used to control the start-up, shutdown, and power output of the gas turbine in the microgrid, the charging and discharging of the energy storage system, and the power interaction with the main grid, based on the optimal scheduling scheme obtained by the optimization solution module. According to the decision-maker's needs, different optimization scheduling directions are selected for different optimization objectives.

Images