Research On Dynamic Economic Dispatch Of Power System Based On Granular Computing

In recent years,the scale of the power system has gradually expanded.The dynamic economic dispatch of the power system is a high-dimensional,multi-constrained,nonlinear optimization problem,which is very difficult to optimize dispatch.After the wind power is connected to the system,the constraint conditions are increased,which further increases the difficulty of scheduling optimization.Granular computing is a typical method for dealing with complex problems,which is widely used in many important fields.Aiming at the problem of dynamic economic dispatch of power system and dynamic economic dispatch with wind power considering valve-point effects,transmission loss and porhibited operation zones.In this paper,the granular computing theory of hierarchical granulation is applied to dynamic economic dispatch.The main research contents are as follows:First,the basic theories involved in the article are analyzed in detail,and the particles,layers,and structures of the Granular Computing(Grc)model are discussed.By granulating the problem space,the problem dimension is reduced.The principles and solution steps of power system economic dispatch related optimization algorithms including Particle Swarm Optimization(PSO)and Differential Evolution(DE)are also discussed.And adjusted the relevant parameters and mutation strategy.Accelerate the convergence speed of the algorithm to facilitate the study of subsequent chapters.Then,the power system dynamic economic dispatch model and solution strategy based on granular computing are proposed.The model is divided into three layers.Study the dynamic economic dispatch of large power grids at different levels and different granularities.On the basis of the model,a method suitable for dynamic economic scheduling granularity division is proposed.Granulate according to the convergence characteristics at each moment to minimize the effect of decoupling on the overall calculation.The solution process of the granular computing model is discussed in detail.Through the Grc-DE algorithm and Grc-PSO algorithm,the dynamic economic dispatch of 5,10,30 and 100 units is studied and solved.Reduce the dimensions of the system andreduce the cost of power generation.It proves the superiority of granular computing in solving complex economic scheduling problems.Finally,a method for solving dynamic economic dispatch of power system with improved granular computing is proposed.On the basis of the proposed particle calculation method,the solution algorithm is improved.By combining the search diversity of the differential evolution(DE)algorithm and the memory mechanism of the particle swarm optimization(PSO)algorithm,a hybrid algorithm DE-PSO is proposed.And use the two-population strategy to solve.The hybrid algorithm is applied to granular computing.The Grc-DE-PSO algorithm is used to solve the problem of dynamic economic dispatch of wind farms.And comparative analysis with other algorithms.Effectively reduce the cost of power generation and calculation time.Further verify the superiority of granular computing to solve complex economic scheduling problems.