An optimal dispatching method for new energy power system based on artificial intelligence

Abstract

The present invention relates to the field of artificial intelligence technology, and in particular, to an optimization and dispatching method for a new energy power system based on artificial intelligence. It comprises the following steps: collecting multi-source data in the power system and preprocessing the collected data; obtaining the power generation of renewable energy according to the collected data and in combination with the needs of the power system; defining multiple objective functions and constraints based on wind energy and solar energy, combining genetic algorithm and particle swarm optimization to form a hybrid optimization algorithm, thereby using the hybrid optimization algorithm to solve the solution of the objective function and generate a real-time adjustment and dispatching strategy; applying the optimal solution set obtained by the hybrid optimization algorithm to the actual power system to generate a dispatching instruction. The method can better utilize solar energy and wind energy resources, reduce waste, and improve the overall efficiency of the system. At the same time, it helps to conduct local refined search after finding potential high-quality solutions in the optimization dispatching, thereby improving the accuracy of the solution.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based new energy power system optimization scheduling method. Background Art

[0002] With the increasing global awareness of environmental protection and the deepening pursuit of sustainable development, the demand for renewable energy is rising. Among them, wind energy and solar energy are favored because of their clean, pollution-free and renewable characteristics. However, these two energy sources face considerable challenges in practical applications, especially their inherent intermittent and volatile nature. For example, wind power generation depends on changes in wind speed, while solar power generation is affected by the duration and intensity of sunshine. These factors lead to the instability of renewable energy generation. This instability not only puts pressure on the stable operation of the power grid, but also increases the difficulty of power system dispatching. In order to ensure the safety and reliability of power supply, effective technical means and management measures must be taken to optimize dispatching, such as smoothing output through energy storage technology, improving prediction accuracy through intelligent prediction algorithms, and establishing a more flexible power market mechanism, so as to achieve an effective combination of new energy and traditional energy and promote the efficient and stable operation of the entire power system.

[0003] However, the scheduling methods of existing technologies often find it difficult to cope with the intermittency and uncertainty of wind and solar energy, resulting in problems such as low energy utilization efficiency, severe equipment wear, high maintenance costs, and poor system stability. In addition, traditional optimization methods find it difficult to strike a balance between global search and local search, resulting in low search efficiency and insufficient solution accuracy. Therefore, an optimization scheduling method for a new energy power system based on artificial intelligence is designed. Summary of the invention

[0004] The purpose of the present invention is to provide an optimization and scheduling method for a new energy power system based on artificial intelligence to solve the problem raised in the above background technology that it is difficult to cope with the intermittent and uncertain nature of wind energy and solar energy and that it is difficult to balance the optimization method between global search and local search.

[0005] To achieve the above object, the present invention aims to provide a method for optimizing and dispatching a new energy power system based on artificial intelligence, comprising the following steps:

[0006] S1. Collect multi-source data in the power system and pre-process the collected data;

[0007] S2. Obtain the power generation of renewable energy based on the collected data and in combination with the needs of the power system; renewable energy includes wind energy and solar energy; in the process of obtaining the power generation of wind energy, the influence factor of turbulence is introduced for optimization; in the process of obtaining the power generation of solar energy, the inverter efficiency and the efficiency and capacity of battery energy storage are introduced for optimization;

[0008] S3. Based on wind energy and solar energy, multiple objective functions and constraints are defined, and the genetic algorithm and particle swarm optimization are combined to form a hybrid optimization algorithm, so as to use the hybrid optimization algorithm to solve the solution of the objective function and generate a real-time adjustment scheduling strategy;

[0009] S4. Apply the optimal solution set obtained by the hybrid optimization algorithm to the actual power system to generate dispatch instructions; wherein the dispatch instructions include the output plan of each wind turbine and solar photovoltaic panel, the charging and discharging strategy of the energy storage system, and the load distribution plan of the power grid.

[0010] As a further improvement of the technical solution, the power generation amount of the renewable energy obtained in S2 is specifically as follows:

[0011] Wind power generation:

[0012] ;

[0013] in, For wind turbines in time Output power; is the air density; is the area swept by the wind turbine blades; For in time wind speed; is the power factor of the wind turbine; is the efficiency of the generator;

[0014] Solar Power:

[0015] ;

[0016] in, For solar panels in time Output power; For in time The intensity of solar radiation incident on the solar panel at 1000 Hz; is the effective area of ​​the solar panel; is the photoelectric conversion efficiency of the solar panel; is the temperature coefficient; For in time The operating temperature of the solar panel; is the reference temperature.

[0017] As a further improvement of the technical solution, in S2, the influencing factors of turbulence are introduced to optimize the process of obtaining wind energy power generation, as follows:

[0018] ；

[0019] in, The optimized wind turbine generator is Output power; is atmospheric pressure; is the molar mass of air; is the gas constant; is the air temperature; is the power coefficient under non-turbulence conditions; is the correction factor for turbulence intensity; is the correction factor of the pitch angle; For the temperature Generator efficiency under ; For aerodynamic efficiency.

[0020] As a further improvement of the technical solution, in S2, the efficiency of the inverter and the efficiency and capacity of the battery energy storage are introduced to optimize the process of obtaining the power generation of solar energy, and the optimization is as follows:

[0021] ;

[0022] in, For optimized solar panels in time Output power; The efficiency of the inverter in converting DC power to AC power; The efficiency of the battery during charging and discharging; is the maximum capacity of the battery; For battery in time remaining power when

[0023] As a further improvement of the technical solution, the objective function in S3 is specifically:

[0024] Maximize the utilization of new energy:

[0025] ;

[0026] in, The objective function is to maximize the utilization rate of new energy; The time interval for optimization;

[0027] Minimize energy storage system costs:

[0028] ;

[0029] in, The objective function for minimizing the cost of the energy storage system; ; For in time Charging cost; For in time The charging power of the energy storage system; For in time The discharge cost; For in time The discharge power of the energy storage system.

[0030] As a further improvement of the technical solution, the constraint conditions in S3 include power balance constraint, energy storage system capacity constraint and energy storage system charge and discharge power limit, specifically:

[0031] Power balance constraints:

[0032] ;

[0033] in, For in time electricity demand; is the base load; Load changes caused by demand response;

[0034] ;

[0035] in, is the number of conventional power generation units; ; For the Conventional power generation units in time Output power; For in time The charging power of the energy storage system; For in time The discharge power of the energy storage system;

[0036] Energy storage system capacity constraints:

[0037] ;

[0038] in, For in time The state of charge of the energy storage system; is the maximum state of charge of the energy storage system;

[0039] Energy storage system charging and discharging power limit:

[0040] ;

[0041] ;

[0042] in, is the maximum charging power of the energy storage system; is the maximum discharge power of the energy storage system.

[0043] As a further improvement of the technical solution, in S3, the genetic algorithm and the particle swarm optimization are combined to form a hybrid optimization algorithm, so as to use the hybrid optimization algorithm to solve the solution of the objective function and generate a real-time adjustment scheduling strategy. The specific steps are as follows:

[0044] S31. For the genetic algorithm, generate Random individuals , for particle swarm optimization, generate Random particles ;

[0045] S32. For each individual and particles , calculate its fitness value and ;

[0046] S33, for the genetic algorithm part, select individuals with good performance from the current population as the parents of the next generation, exchange genes between the two selected parent individuals with a preset probability to produce new offspring, and then change some genes of some individuals with a low probability;

[0047] S34, for the particle swarm optimization part, adjust the speed of each particle according to the optimal position of the particle itself and personal experience, and then adjust the position of the particle according to the updated speed;

[0048] S35. Merge the new individuals generated by the genetic algorithm and the particles updated by particle swarm optimization to form a new mixed population , select individuals and particles with high fitness from the mixed population as the initial population and particles of the next generation;

[0049] S36. For each generation of mixed population, sort them according to the non-dominated relationship, find the non-dominated individuals, that is, the Pareto optimal solution, calculate the crowding distance between the non-dominated solutions of the same level, and select the solution with the larger distance to enter the next generation;

[0050] S37, stop after reaching a preset maximum number of iterations or when the diversity of the population drops to a certain level or the change of the optimal solution is less than a certain threshold;

[0051] S38. Extract the Pareto optimal solution set from the final mixed population and select the optimal solution as the final solution based on actual conditions and preferences.

[0052] As a further improvement of the technical solution, the S33 is specifically as follows:

[0053] Select individuals with good performance from the current population as parents of the next generation based on roulette wheel selection, and calculate the relative fitness of each individual: , and then select individuals in proportion;

[0054] For two parent individuals and , at position To do a crossover:

[0055] ;

[0056] ;

[0057] in, is the first offspring individual generated after the crossover operation; is the second offspring individual generated after the crossover operation; is the first parent individual; is the second parent individual; is the gene length of the individual; is the intersection point;

[0058] Changing some genes of some individuals with low probability:

[0059] ;

[0060] in, The new individual after mutation is The value of each gene position; For the original individual The value of each gene position; is a random disturbance.

[0061] As a further improvement of the technical solution, the S34 is specifically as follows:

[0062] Adjust the speed of each particle:

[0063] ;

[0064] in, For particles In the The speed of generation; is the inertia weight; For particles In the The speed of generation; and is the learning factor; and For Random numbers in the interval; For particles The best position in history; For the The particle in The position of the generation; is the global best position;

[0065] Adjust the particle's position based on the updated velocity:

[0066] ;

[0067] in, For the The particle in The position of the generation.

[0068] As a further improvement of the technical solution, the S36 is specifically as follows:

[0069] For each solution , if there is no other solution Make and ,but It is a non-dominated solution. Sort the mixed population according to the non-dominated relationship and find the non-dominated individuals.

[0070] Calculate the crowding distance of each non-dominated solution :

[0071]

[0072] in, For the The crowding distance of a solution; For the The maximum value of the objective function, for The minimum value of the objective function; For the The solution is The value of the objective function; For the The solution is The value of the objective function.

[0073] Compared with the prior art, the present invention has the following beneficial effects:

[0074] 1. In the optimization and dispatching method of the new energy power system based on artificial intelligence, by introducing the turbulence model in wind power generation, it is possible to more accurately predict the power generation of wind turbines under different conditions, reduce equipment wear, extend its service life, reduce failures and damage caused by turbulence, and reduce maintenance frequency and costs. In solar power generation, the control strategy of optimizing the inverter is introduced to maximize the conversion efficiency and increase the power generation. The introduction of the energy storage system can store excess electrical energy and release it when needed, smooth the load curve, reduce the impact on the power grid, and improve the stability of the system. Through comprehensive optimization, solar energy resources can be better utilized, waste can be reduced, and the overall benefits of the system can be improved.

[0075] 2. In the artificial intelligence-based optimization and dispatching method for renewable energy power systems, the hybrid optimization algorithm uses a larger learning factor in the early stage of optimization, which helps particles to explore different areas more actively and increase the diversity of the search. As the number of iterations increases, the learning factor gradually decreases, that is, the particle's learning of its own historical best position and the global best position weakens, which helps to conduct local refined searches after finding potential high-quality solutions and improve the accuracy of the solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 The figure is a flow chart of the overall method of the present invention. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] Example

[0079] See also Figure 1 As shown, a new energy power system optimization scheduling method based on artificial intelligence is provided, comprising the following steps:

[0080] S1. Collect multi-source data in the power system and pre-process the collected data;

[0081] Multi-source data comes from various sensors, weather forecasts and historical operation data. Multi-source data includes at least power generation, load demand, weather conditions and grid status;

[0082] Specifically, it can obtain real-time data from smart sensors installed in power stations, substations, transmission lines, etc., which may include voltage, current, power factor, temperature, etc.; obtain weather forecasts for the future through meteorological satellites, ground meteorological stations, etc., especially for wind farms and solar photovoltaic power stations, wind speed, cloud coverage, sunshine intensity and other information are crucial; and there are also operation records of the power system over a period of time (such as weeks to years), including but not limited to power generation, load demand, fault conditions, etc.;

[0083] S2. Obtain the power generation of renewable energy based on the collected data and in combination with the needs of the power system; renewable energy includes wind energy and solar energy; in the process of obtaining the power generation of wind energy, the influence factor of turbulence is introduced for optimization; in the process of obtaining the power generation of solar energy, the inverter efficiency and the efficiency and capacity of battery energy storage are introduced for optimization;

[0084] The power generation of renewable energy in S2 is as follows:

[0085] Wind power generation:

[0086] ;

[0087] in, For wind turbines in time Output power; is the air density; is the area swept by the wind turbine blades; For in time wind speed; The power coefficient of the wind turbine indicates the efficiency of converting wind energy into mechanical energy, usually between 0.2 and 0.4; The efficiency of the generator, which indicates the efficiency of converting mechanical energy into electrical energy, is usually between 0.8 and 0.9;

[0088] In the process of obtaining wind energy, the influence of turbulence is introduced for optimization. Turbulence refers to the irregular, three-dimensional, disordered movement in the wind flow, as follows:

[0089] ；

[0090] in, The optimized wind turbine generator is Output power; It is the atmospheric pressure, which can be measured in real time through weather forecasts or sensors; is the molar mass of air; is the gas constant, approximately ; is the air temperature, which can be measured in real time through weather forecasts or sensors; is the power coefficient under non-turbulent conditions, usually between 0.2 and 0.4; is the correction factor for turbulence intensity, , Usually between 0.1 and 0.3; is the correction factor of the pitch angle, , is a constant that indicates the influence of the pitch angle on the power coefficient. is the optimal pitch angle; For the temperature The generator efficiency under , is the generator efficiency at reference temperature, usually between 0.8 and 0.9; is the temperature correction factor, is a constant that indicates the effect of temperature on the efficiency of the generator. Base temperature; for aerodynamic efficiency;

[0091] Turbulence can cause uneven distribution and instantaneous changes in wind speed, affecting the output power of wind turbines. It can also cause mechanical stress that can damage the structure of wind turbines. By introducing a turbulence model, it is possible to more accurately predict the power generation of wind turbines under different conditions, reduce equipment wear, extend its service life, reduce failures and damage caused by turbulence, and reduce maintenance frequency and costs.

[0092] Solar Power:

[0093] ;

[0094] in, For solar panels in time Output power; For in time The intensity of solar radiation incident on the solar panel at 1000 Hz; is the effective area of ​​the solar panel; is the photoelectric conversion efficiency of solar panels. The efficiency range of common solar panels is arrive ; The temperature coefficient indicates the ratio of the output power decrease for every 1°C increase in temperature, usually between 0.004 and 0.006; For in time The operating temperature of the solar panel; is the reference temperature, usually 25°C, which is the temperature of the solar panel under standard test conditions;

[0095] In the process of obtaining solar power generation, the efficiency of the inverter and the efficiency and capacity of the battery energy storage are introduced for optimization. After optimization, the specific results are:

[0096] ;

[0097] in, For optimized solar panels in time Output power; The efficiency of the inverter in converting DC power to AC power, usually between 90% and 98%; The efficiency of the battery during charging and discharging; is the maximum capacity of the battery; For battery in time The remaining power when This factor reflects the impact of the remaining capacity of the battery on the output power. When the battery is close to full load, this factor is close to 0, indicating that the battery cannot accept more energy input; when the battery is close to no load, this factor is close to 1, indicating that the battery can accept more energy input. The optimized solar power output power formula can more accurately reflect the actual system operation and improve the overall performance and reliability of the system.

[0098] Inverter efficiency directly affects the output power of the solar power generation system. By optimizing the control strategy of the inverter, the conversion efficiency can be maximized and the power generation can be increased. The energy storage system can store excess power and release it when needed, smooth the load curve, reduce the impact on the power grid, and improve the stability of the system. Through comprehensive optimization, solar energy resources can be better utilized, waste can be reduced, and the overall benefits of the system can be improved.

[0099] S3. Based on wind energy and solar energy, multiple objective functions and constraints are defined, and the genetic algorithm and particle swarm optimization are combined to form a hybrid optimization algorithm, so as to use the hybrid optimization algorithm to solve the solution of the objective function and generate a real-time adjustment scheduling strategy;

[0100] The specific objective function in S3 is:

[0101] Maximize the utilization of new energy:

[0102] ;

[0103] in, The objective function is to maximize the utilization rate of new energy; The time interval for optimization, usually one day or one hour, etc.

[0104] Minimize energy storage system costs:

[0105] ;

[0106] in, The objective function for minimizing the cost of the energy storage system; ; For in time Charging cost; For in time The charging power of the energy storage system; For in time The discharge cost; For in time The discharge power of the energy storage system;

[0107] By maximizing the objective function of new energy utilization , which can ensure the use of renewable energy such as wind and solar energy as much as possible, reduce dependence on traditional fossil fuels, help reduce carbon emissions and promote environmental sustainability; by minimizing the objective function of energy storage system cost , while ensuring the stable operation of the power system, it can effectively control and reduce the operating costs of the energy storage system. This not only helps to improve economic efficiency, but also enhances the competitiveness of the system.

[0108] The constraints in S3 include power balance constraints, energy storage system capacity constraints, and energy storage system charging and discharging power limits, specifically:

[0109] Power balance constraints:

[0110] ;

[0111] in, For in time electricity demand; is the base load; Load changes caused by demand response;

[0112] ;

[0113] in, is the number of conventional power generation units; ; For the Conventional power generation units in time Output power; For in time The charging power of the energy storage system; For in time The energy storage system discharge power ensures that at any time, the total output power of all power sources in the system (including new energy, conventional power generation units and energy storage systems) is equal to the total load. This helps to avoid grid instability caused by oversupply or undersupply; introduces demand response Incentives are allowed to adjust user electricity consumption behavior, smooth the load curve, reduce the pressure on the power grid during peak hours, and improve the overall operating efficiency of the system.

[0114] Energy storage system capacity constraints:

[0115] ;

[0116] in, For in time The state of charge of the energy storage system; It is the maximum state of charge of the energy storage system. It can prevent the energy storage system from overcharging or discharging, avoid battery damage, extend its service life, ensure the long-term stable operation of the system, and reasonably manage the state of charge of the energy storage system, so that electric energy can be stored during the low electricity price period and released during the peak period, which helps to reduce the peak and fill the valley, and further improve the utilization rate of new energy and economic benefits.

[0117] Energy storage system charging and discharging power limit:

[0118] ;

[0119] ;

[0120] in, is the maximum charging power of the energy storage system; It is the maximum discharge power of the energy storage system. It can prevent the energy storage device from being subjected to excessive current in a short period of time, reduce the probability of failure, and ensure safe operation. Clear upper limits on charge and discharge power help to respond quickly to sudden changes in demand or supply, maintain grid frequency stability, and improve the flexibility and reliability of the entire system.

[0121] In S3, genetic algorithm and particle swarm optimization are combined to form a hybrid optimization algorithm, so as to use the hybrid optimization algorithm to solve the solution of the objective function and generate a real-time adjustment scheduling strategy. The specific steps are as follows:

[0122] S31. For the genetic algorithm, generate Random individuals , for particle swarm optimization, generate Random particles ; It provides a diverse initial solution space for the algorithm, increasing the possibility of finding the global optimal solution;

[0123] S32. For each individual and particles , calculate its fitness value and ; It can quantify the quality of each solution and provide a basis for subsequent selection, crossover, mutation and speed and position update;

[0124] S33. For the genetic algorithm, select individuals with good performance from the current population as the parents of the next generation, exchange genes between the two selected parent individuals with a preset probability to produce new offspring, and then change some genes of some individuals with a low probability to increase population diversity; through selection, crossover and mutation operations, the genetic algorithm can simulate the natural selection process, retain excellent characteristics and introduce new characteristics, and promote the evolution of the population to a better solution. The selection operation ensures the inheritance of excellent individuals, the crossover operation promotes gene exchange, and the mutation operation increases the diversity of the population and avoids premature convergence;

[0125] S33 is as follows:

[0126] Select individuals with good performance from the current population as parents of the next generation based on roulette wheel selection, and calculate the relative fitness of each individual: , and then select individuals in proportion;

[0127] For two parent individuals and , at position To do a crossover:

[0128] ;

[0129] ;

[0130] in, is the first offspring individual generated after the crossover operation; is the second offspring individual generated after the crossover operation; is the first parent individual; is the second parent individual; is the gene length of the individual, that is, the number of genes contained in each individual; is the intersection point; express From the first gene to the part of a gene, express From Gene to part of a gene;

[0131] Changing some genes of some individuals with low probability:

[0132] ;

[0133] in, The new individual after mutation is The value of each gene position; For the original individual The value of each gene position; is a random perturbation, usually drawn from a distribution (such as a normal distribution), The size and distribution of can be adjusted according to the specific problem to control the intensity of variation;

[0134] For example: There are two parent individuals and , each individual has 5 genes, and the gene values ​​are as follows:

[0135] ;

[0136] ;

[0137] Assume cross , then the offspring individuals after crossover are:

[0138] ;

[0139] ;

[0140] Next, assuming that The fourth gene of , random perturbation ,but:

[0141] ;

[0142] After mutation for:

[0143] ;

[0144] S34. For the particle swarm optimization part, adjust the speed of each particle according to its own best position and personal experience, and then adjust the position of the particle according to the updated speed; by adjusting the speed and position of the particle, use the wisdom of the group to guide the search direction, and accelerate the convergence speed to the optimal solution area. The particle updates its speed according to its own historical best position and the historical best position of the group, which helps to quickly locate potential high-quality solutions in the search space;

[0145] S34 is as follows:

[0146] Adjust the speed of each particle:

[0147] ;

[0148] in, For particles In the The speed of generation; is the inertia weight; For particles In the The speed of generation; and is the learning factor; and For Random numbers in the interval; For particles The best position in history; For the The particle in The position of the generation; is the global best position;

[0149] In the early stages of optimization, a larger learning factor and It helps particles to explore different areas more actively and increase the diversity of search. As the number of iterations increases, the learning factor gradually decreases to and , the particle's learning degree of its own historical best position and global best position is weakened, which helps to conduct local refined search after finding potential high-quality solutions and improve the accuracy of the solution.

[0150] Adjust the particle's position based on the updated velocity:

[0151] ;

[0152] in, For the The particle in The position of the generation;

[0153] S35. Merge the new individuals generated by the genetic algorithm and the particles updated by particle swarm optimization to form a new mixed population , individuals and particles with high fitness are selected from the mixed population as the initial population and particles of the next generation; the advantages of the two algorithms are complementary, which not only ensures the global search capability but also improves the local search accuracy;

[0154] S36. For each generation of mixed population, sort them according to non-dominated relationships to find non-dominated individuals, i.e., Pareto optimal solutions. In order to maintain the diversity of solutions, calculate the crowding distance between non-dominated solutions at the same level, and select solutions with larger distances to enter the next generation; non-dominated sorting helps to identify Pareto optimal solutions, i.e., those solutions that cannot further improve any one goal among multiple goals without damaging other goals. Crowding distance calculation ensures the diversity of solutions, avoids the optimization results being too concentrated in a specific area, and helps to discover more potential high-quality solutions;

[0155] S36 is as follows:

[0156] For each solution , if there is no other solution Make and ,but It is a non-dominated solution. Sort the mixed population according to the non-dominated relationship and find the non-dominated individuals.

[0157] Calculate the crowding distance of each non-dominated solution :

[0158] ;

[0159] in, For the The crowding distance of the solution is the sum of the normalized gaps of all objective functions. The larger the crowding distance, the better the solution. The fewer solutions around a solution, that is, the sparser the distribution of the solution in the target space, by calculating the crowding distance, a solution with a more uniform distribution can be selected from the non-dominated solutions, thereby maintaining the diversity of solutions and the uniformity of distribution; For the The maximum value of the objective function, for The minimum value of the objective function; For the The solution is The value of the objective function; For the The solution is The value of the objective function;

[0160] This difference reflects the The distribution density of solutions around the solution. If this difference is large, it means that the The space around the solution is relatively sparse; if this difference is small, it means that the The space around a solution is relatively dense;

[0161] S37, stop after reaching the preset maximum number of iterations or when the diversity of the population drops to a certain level or the change in the optimal solution is less than a certain threshold; this can effectively control the algorithm running time, prevent meaningless additional iterations, and ensure that the algorithm can stop in time after finding a satisfactory solution;

[0162] S38. Extract the Pareto optimal solution set from the final mixed population. These solutions represent the best compromise between different objectives. Select the optimal solution as the final solution based on the actual situation and preference.

[0163] Finally, the optimal solution set obtained by the hybrid optimization algorithm is applied to the actual power system to generate dispatch instructions, which include the output plan of each wind turbine and solar photovoltaic panel, the charging and discharging strategy of the energy storage system, and the load distribution plan of the power grid. These dispatch instructions will be adjusted in real time to ensure the stable operation and efficient utilization of the power system, while maximizing the utilization rate of renewable energy, reducing operating costs, and reducing environmental pollution. By continuously monitoring system performance and collecting feedback data, the dispatch strategy is continuously optimized to improve the overall performance and reliability of the system.

[0164] For example, the unmet need is:

[0165] ;

[0166] if , indicating that the renewable energy generation is insufficient to meet the load demand, and it is necessary to start conventional generators or use energy storage systems to discharge; if , indicating that the power generation of renewable energy exceeds the load demand, it is possible to consider reducing the output of conventional generators or using energy storage systems for charging;

[0167] For each time step , according to the demand gap and the start-up or stop costs of conventional generators, decide which units to start or stop; consider the start-up cost and operating cost, and give priority to starting units with lower costs; ensure that the started units can continue to operate in the next time step to avoid frequent starts and stops; for conventional generators that have already started, Adjust the output to ensure the balance between system supply and demand; at the same time, consider the maximum or minimum output limit of the unit to avoid exceeding the operating range.

[0168] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A new energy power system optimization dispatching method based on artificial intelligence, characterized in that: The steps include: S1. Collect multi-source data in the power system and pre-process the collected data; S2. Obtain the power generation of renewable energy based on the collected data and in combination with the needs of the power system; renewable energy includes wind energy and solar energy; in the process of obtaining the power generation of wind energy, the influence factor of turbulence is introduced for optimization; in the process of obtaining the power generation of solar energy, the inverter efficiency and the efficiency and capacity of battery energy storage are introduced for optimization; S3. Based on wind energy and solar energy, multiple objective functions and constraints are defined, and the genetic algorithm and particle swarm optimization are combined to form a hybrid optimization algorithm, so as to use the hybrid optimization algorithm to solve the solution of the objective function and generate a real-time adjustment scheduling strategy; S4, applying the optimal solution set obtained by the hybrid optimization algorithm to the actual power system to generate a dispatch instruction; wherein the dispatch instruction includes the output plan of each wind turbine and solar photovoltaic panel, the charging and discharging strategy of the energy storage system, and the load distribution plan of the power grid; In S3, the genetic algorithm and particle swarm optimization are combined to form a hybrid optimization algorithm, so as to use the hybrid optimization algorithm to solve the solution of the objective function and generate a real-time adjustment scheduling strategy. The specific steps are as follows: S31. For the genetic algorithm, generate Random individuals , for particle swarm optimization, generate Random particles ; S32. For each individual and particles , calculate its fitness value and ; S33, for the genetic algorithm part, select individuals with good performance from the current population as the parents of the next generation, exchange genes between the two selected parent individuals with a preset probability to produce new offspring, and then change some genes of some individuals with a low probability; S34, for the particle swarm optimization part, adjust the speed of each particle according to the optimal position of the particle itself and personal experience, and then adjust the position of the particle according to the updated speed; S35. Merge the new individuals generated by the genetic algorithm and the particles updated by particle swarm optimization to form a new mixed population , select individuals and particles with high fitness from the mixed population as the initial population and particles of the next generation; S36. For each generation of mixed population, sort them according to the non-dominated relationship, find the non-dominated individuals, that is, the Pareto optimal solution, calculate the crowding distance between the non-dominated solutions of the same level, and select the solution with the larger distance to enter the next generation; S37, stop after reaching a preset maximum number of iterations or when the diversity of the population drops to a certain level or the change of the optimal solution is less than a certain threshold; S38. Extract the Pareto optimal solution set from the final mixed population, and select the optimal solution as the final solution based on the actual situation and preference; The S33 is specifically as follows: Select individuals with good performance from the current population as parents of the next generation based on roulette wheel selection, and calculate the relative fitness of each individual: , and then select individuals in proportion; For two parent individuals and , at position To do a crossover: ； ； in, is the first offspring individual generated after the crossover operation; is the second offspring individual generated after the crossover operation; is the first parent individual; is the second parent individual; is the gene length of the individual; is the intersection point; Changing some genes of some individuals with low probability: ； in, The new individual after mutation is The value of each gene position; For the original individual The value of each gene position; is a random disturbance.

2. The method for optimizing and dispatching a new energy power system based on artificial intelligence according to claim 1 is characterized in that: The power generation amount of renewable energy obtained in S2 is as follows: Wind power generation: ； in, For wind turbines in time Output power; is the air density; is the area swept by the wind turbine blades; For in time wind speed; is the power factor of the wind turbine; is the efficiency of the generator; Solar Power: ； in, For solar panels in time Output power; For in time The intensity of solar radiation incident on the solar panel at 1000 Hz; is the effective area of ​​the solar panel; is the photoelectric conversion efficiency of the solar panel; is the temperature coefficient; For in time The operating temperature of the solar panel; is the reference temperature.

3. The method for optimizing and dispatching a new energy power system based on artificial intelligence according to claim 2 is characterized in that: In S2, the influencing factors of turbulence are introduced to optimize the process of obtaining wind energy generation, as follows: ； in, The optimized wind turbine generator Output power; is atmospheric pressure; is the molar mass of air; is the gas constant; is the air temperature; is the power coefficient under non-turbulence conditions; is the correction factor for turbulence intensity; is the correction factor of the pitch angle; For the temperature Generator efficiency under ; For aerodynamic efficiency.

4. The method for optimizing and dispatching a new energy power system based on artificial intelligence according to claim 3 is characterized in that: In S2, the inverter efficiency and the efficiency and capacity of the battery energy storage are introduced to optimize the process of obtaining the power generation of solar energy. After optimization, the specific results are: ； in, For optimized solar panels in time Output power; The efficiency of the inverter in converting DC power to AC power; The efficiency of the battery during charging and discharging; is the maximum capacity of the battery; For battery in time remaining power when 5. The method for optimizing and dispatching a new energy power system based on artificial intelligence according to claim 4 is characterized in that: The objective function in S3 is specifically: Maximize the utilization of new energy: ； in, The objective function is to maximize the utilization rate of new energy; The time interval for optimization; Minimize energy storage system costs: ； in, The objective function for minimizing the cost of the energy storage system; ; For in time Charging cost; For in time The charging power of the energy storage system; For in time The discharge cost; For in time The discharge power of the energy storage system.

6. The method for optimizing and dispatching a new energy power system based on artificial intelligence according to claim 5 is characterized in that: The constraints in S3 include power balance constraints, energy storage system capacity constraints, and energy storage system charge and discharge power limits, specifically: Power balance constraints: ； in, For in time electricity demand; is the basic load; Load changes caused by demand response; ； in, is the number of conventional power generation units; ; For the Conventional power generation units in time Output power; For in time The charging power of the energy storage system; For in time The discharge power of the energy storage system; Energy storage system capacity constraints: ； in, For in time The state of charge of the energy storage system; is the maximum state of charge of the energy storage system; Energy storage system charging and discharging power limit: ； ； in, is the maximum charging power of the energy storage system; is the maximum discharge power of the energy storage system.

7. The method for optimizing and dispatching a new energy power system based on artificial intelligence according to claim 1 is characterized in that: The S34 is specifically as follows: Adjust the speed of each particle: ； in, For particles In the The speed of generation; is the inertia weight; For particles In the The speed of generation; and is the learning factor; and For Random numbers in the interval; For particles The best position in history; For the The particle in The position of the generation; is the global best position; Adjust the particle's position based on the updated velocity: ； in, For the The particle in Generation location.

8. The method for optimizing and dispatching a new energy power system based on artificial intelligence according to claim 1 is characterized in that: The S36 is specifically as follows: For each solution , if there is no other solution Make and ,but It is a non-dominated solution. Sort the mixed population according to the non-dominated relationship and find the non-dominated individuals. Calculate the crowding distance of each non-dominated solution : ； in, For the The crowding distance of a solution; For the The maximum value of the objective function, for The minimum value of the objective function; For the The solution is The value of the objective function; For the The solution is The value of the objective function.

Images