| The amorphous alloy metal distribution transformer(AMDT)is an ideal energy saving products in the field of transformers,which has the characteristics of low no-load loss.However,the production costs of AMDT are much more expensive compared with the conventional silicon steel sheet core transformer.In addition,the traditional transformer design method has low automation,large design workload and long design period,so it is difficult to achieve high efficiency,energy saving and low cost design.In view of the problems existing in the design of AMDT,this thesis aims to reduce the main material cost and total loss of AMDT to optimize the AMDT with single-objective and multi-objective optimization by using intelligent algorithm.Based on the market demand,an optimization system of AMDT based on intelligent algorithm is designed to improve the design efficiency.The main research works of this thesis are as follows:(1)The traditional manual design method of AMDT is studied,and the design flow of AMDT is constructed.The engineering calculation method of electromagnetic,structural and performance parameters related to the core and winding of AMDT is introduced.It provides a reference for the selection of objective function and optimization variable.(2)A single-objective and multi-objective optimization models of AMDT are built,and the main material cost and total loss of the AMDT are selected as the optimization objective function.The basic theories of single-objective genetic algorithm(GA)and multi-objective fast and elitist non-dominated sorting genetic algorithm(NSGA-Ⅱ)are introduced respectively.The shortcomings of GA and NSGA-Ⅱ in the optimization design of AMDT are reviewed,and the corresponding improvement strategies are put forward.The realization method of single-objective and multiobjective optimization model in AMDT of Chaos Genetic Algorithm(CGA),Adaptive Genetic Algorithm(AGA),Chaos Adaptive Genetic Algorithm(CAGA)and improved NSGA-Ⅱ is proposed,which provides a reference for the development of software system optimization algorithm program.(3)Based on the theory of transformer electromagnetic design and optimization algorithm,combined with software system functions and user requirements,a set of AMDT optimization system based on intelligent algorithm is designed and developed by Visual Basic 6.0 software tools.The interactive connections among the optimization system and other softwares are realized through the interactive interface design of the software system.(4)The optimization ability and running speed of GA,particle swarm optimization(PSO)and differential evolution(DE)are studied.The simulation results show that GA is more suitable for the optimization design of AMDT.On this basis,CAG、AGA、CAGA and improved NSGA-Ⅱ are used to execute the single-objective and multi-objective optimization of SCLBH15-315/10 AMDT.The analysis of the optimization results shows that the CAGA optimization effect is the best in the single-objective optimization design.Moreover,CAGA can greatly reduce the main material cost of the AMDT and improve its loss performance.The improved NSGA-Ⅱ can obtain more Pareto front solutions on the basis of material saving and energy saving in multi-objective optimization design,which can provide more optimization schemes for transformer manufacturers.In addition,the practicability and efficiency of the optimization system are verified through the execution time and optimization effect of the optimization system... |
PDF Full Text Request