Differential Cross Genetic Algorithm And Its Application In The Optimization Design Of Structure And Inerter System

There are some characteristics for the general structural optimization problems,such as: the design variables are usually discrete,the constraint conditions are not easy to express using formula,and the objective function is non-derivative.Therefore,the general structural optimization problems cannot be effectively solved by traditional structural optimization methods.Genetic algorithm is a bionic intelligent algorithm which has wide range of adaptability,and it can overcome some deficiencies of conventional optimization methods.In this paper,based on the basic theory of genetic algorithm,combined with the characteristics of structural optimization design,the shortcomings of basic genetic algorithm were improved,and a new genetic algorithm,namely differential crossover genetic algorithm,was proposed.Excellent individual generation was guided dynamically by the differential crossover strategy of this algorithm,which speeds up the approach of feasible solutions to global optimal solutions.The probability of falling into the local optimal solution can be reduced and the global search ability was enhanced by the improved genetic algorithm.The main research content of this paper is as follows:(1)According to the implementation principle of the classical genetic algorithm,the shortcomings of the classical genetic algorithm were summarized,and the improvement of the genetic algorithm was put forward.Then,four classical optimization test functions are used to test the performance of the classical genetic algorithm,the binary genetic algorithm and the differential cross genetic algorithm proposed in this paper.Finally,the performance test results of the algorithm are analyzed.(2)An optimization model with the minimum weight of a truss structure as the objective function was established,and the section area of the truss structure was selected as the design variable.Combined with the improved genetic algorithm proposed in this paper,the plane truss structure and space truss structure were optimized,respectively.The optimization results of the improved genetic algorithm proposed in this paper were compared with those of classical genetic algorithm and binary genetic algorithm.(3)Based on the minimum weight of steel frame structures with the objective function of reducing cost,the optimal solution was provided.Then,the structural dimension of the discrete H-beam region was selected as the optimization parameters.The improved genetic algorithm is also used to optimize the frame structure in this paper.The optimization results of the improved genetic algorithm proposed in this paper were compared with those of classical genetic algorithm and binary genetic algorithm.(4)According to the constitutive equation of inerter system and the motion equation for the single-degree-of-freedom(SDOF)structure,the mechanical model of the inerter damping structure was established,and the expression of displacement mean square response of inerter damping structure under white noise excitation was derived.An optimization model with key parameters of inerter system as the objective function and the displacement mean square response of inerter damping structure as the constraint condition was established,and a maximum damping enhancement strategy was used,combined with the improved genetic algorithm proposed in this paper to optimize the inerter system.At the same time,the optimization results of the improved genetic algorithm proposed in this paper were compared with those of classical genetic algorithm and binary genetic algorithm.