Research On Enocomic Dispatch Of Power System Based On Granular Computing

The research on economic dispatch of power system is of great significance to reduce the cost of power generation and ameliorate the economy of power system operation.With the expansion of power system scale,when considering the valve-point effects,network transmission losses and many other practical problems,not only the cost function becomes nonconvex and discontinuous,but also the calculation dimension becomes very high,which makes it difficult to solve the economic dispatch of power system.In accordance with the granular computing theory,on the premise of ensuring the accuracy of the results,the complex problems can be decomposed through reasonable granular partition,and optimized by intelligent algorithm to realize the calculation dimension reduction and solve the results.In this paper,the granular computing theory is applied to the economic dispatch of power system,and the main work is follows:Firstly,the basic framework of power system economic dispatch research based on granular computing is constructed,and the traditional economic dispatch model is improved,then establish the granular computing model of power system economic dispatch.In this model,the complex and high-dimensional economic dispatch calculation is divided into coarse granular computing and fine granular computing,which can effectively reduce the calculation dimension.The process of differential evolution algorithm and the significance of control parameters used in this paper are discussed.Secondly,an economic dispatch method based on granular computing is proposed.In order to ensure the accuracy of the results after dimensionality reduction,a granular partition strategy based on the unit convergence characteristics is proposed.The strategy consists of two parts: unit convergence test and unit partition.Through the convergence test,the convergence characteristics of each unit in the optimization process can be obtained.By analysis and comparison,the units that are not easy to converge are clustered into the same fine granular for targeted calculation,and the rest units are evenly divided into other fine particles according to the parameter characteristics to complete the unit division.The feasibility of the proposed granular computing method is verified by numerical simulation.Finally,an economic dispatch method based on improved granular computing is proposed.Based on the granular calculation method,some improvement measures from three aspects are presented.In the initialization process,chaotic operation is introduced,and the initial population with more uniform distribution can be obtained.For the problem of parameter change simply,two nonlinear dynamic parameter adjustment schemes are proposed.In the calculation process,the fast search strategy is adopted to realize the dynamic change of the inner layer iterations with the search process to effectively shorten the calculation time.The simulation test of two improved granular computing schemes in different scale power system can get ideal results in a short time,which verifies the effectiveness of the improved granular computing method,and compares the impact of two parameter change schemes on the simulation results.