Research On Mixed Particle Swarm Algorithm In Optimal Scheduling Of Hydropower Station

No matter from all over the world or from China,the energy crisis has come.Since the harm to the environment was not considered before for the ofindustrialscience.andtechnology,we are ndeeply aware of the serious environmental problems.In view of the shortage of energy,we need tdevelop a new kind of energy tmeet the needs of daily life,and water resources are the first choice for our country to devnew energy,because our country is rich in water resources and is also out of a leading position in the world.However,duto the backwardness of hydropowedevelopment technology in China and the lack of reasonable and effective scienand technology to optimize reservoir operation so asoptimize its producted capacity.The main research works :(1)Considering comprehensively some factors that need to be satisfied in establishing the optimal operation model of hydropower station and some constraints for establishing the objective function;The essence of reservoir optimal operation is to impose some restrictions on the reservoir such as water quantity,power generation and water level on the premise of meeting the maximum economic benefits.(2)Research on the influence of inertia weight on optimization in standard particle swarm optimization algorithm.Research shows that constant inertia weight makes the algorithm easy to fall into local optimal solution and difficult to obtain global optimal solution.Therefore,this paper sets inertia weight to linear decreasing mode through experiments and finds that this linear decreasing distribution can better obtain global optimal solution.(3)There are two learning factors in standard particle swarm optimization algorithm,one is the embodiment of self-learning ability,the other is the embodiment of mutual learning ability between particles;Therefore,it is difficult to obtain the optimal solution if the two particles are kept unchanged.Therefore,we have obtained the change rule of the two learning factors through continuous research.If the two particles are changed asynchronously,there will be a maximum point to be satisfied,and the self-learning ability is the strongest and the learning ability between particles is also the strongest.(4)According to the previous research,it is found that simulated annealing algorithm has strong ability to jump out of local optimization in the later stage.Therefore,in this paper,we consider combining it with particle swarm algorithm to obtain a hybrid particle swarm algorithm called(SA-PSO)by the author.The algorithm uses the position and velocity of particles to bring into the objective function to optimize the solution,and then carries out annealing treatment.In this paper,several simple functions and complex functions are compared,and it is found that the new algorithm has good results in all aspects.(5)The new algorithm is applied to the optimal operation model of hydropower station and good results are obtained.