The local search ability of arithmetic crossover operators common used in real coded genetic algorithms and their influence on the local search performance of the algorithms are analyzed based on the ability of the operators for tuning genes.

Analysis of Local Search Ability of Arithmetic Crossover Operators

算术杂交算子局部搜索能力分析

In chapter 4, the Simple Genetic Algorithm is used to realize the optimal design of taper roller bearing. In account of the defect of Simple Genetic, the real coding is used, the constraints is process by annealing dynamic function, Arithmetic Crossover operator , immunity operator and Non-uniform Mutation operator are used.

The tuning probability of various arithmetic crossover operators for genes (pair) in chromosome (pair) is given.

给出了各种算术杂交算子对染色体(对)的基因(对)的调节概率;

The relationship between the tuning ability of the arithmetic crossover operator for population and the elements in the algorithm such as the match method、the probability of crossover and the tuning probability of the arithmetic crossover operator for genes is revealed.

In this algorithm decimal coding (which is suit for multi-dimension search) is adopted, the arithmetic crossovering and adaptive mutation are also introduced.

On the basis of analyzing and summarizing the fundamental theory and research achievement of genetic algorithm, this paper proposed a new method of dynamic optimization of structure based on genetic algorithm. A new generalized genetic algorithm was presented in this paper and the corresponding program was designed. Many new theories such as integer code, real code, population isolation, optimum reserved strategy and adaptive random mutation were used.

By the evolution programs the method can search the best solution from many initial points simultaneously and obtain the total optimum solution of concrete creep model parameters with real chromosome, nonhomogeneous variation, simple crossover and arithmetical crossover operators.

Due to the linear search capability of algebra crossover operator and random search capability of mutation operator, the efficiency of real-coded genetic algorithm (RGA) is very bad.

For an individual being real coding, arithmetic crossover operator is adopted.

For this work, a floating-point encoded evolutionary algorithm with arithmetic crossover and mutation operations was applied.

To study the depths of seismic discontinuities in the upper mantle using the observed receiver function, this paper developed an inversion method of peeling genetic algorithm. The whole studied depth is divided into several sections from surface, and then inversion is made for each section. The effects of multiple reflected and reflect-transformed phases at the shallower discontinuities on the refract-transformed phase at the deeper one are peeled. Floating point coding, arithmetical crossover, and non-uniform...

To study the depths of seismic discontinuities in the upper mantle using the observed receiver function, this paper developed an inversion method of peeling genetic algorithm. The whole studied depth is divided into several sections from surface, and then inversion is made for each section. The effects of multiple reflected and reflect-transformed phases at the shallower discontinuities on the refract-transformed phase at the deeper one are peeled. Floating point coding, arithmetical crossover, and non-uniform mutation were used in the genetic algorithm in the inversion of each section. The method was checked using the model ispai'91 and applied to the inversion of velocity structure beneath HIA station.

An effective strategy based on real coded genetic algorithms was developed to optimize the operating conditions of simulated moving bed chromatography. The optimization algorithm's objection function was the normalized production intensity. In addition, the n exponent of purity was used as penalty function. The fitness of real coded genetic algorithm fitness was calculated with steady model of simulated moving bed chromatography. In order to satisfy the constrain conditions of simulated moving bed chromatography,...

An effective strategy based on real coded genetic algorithms was developed to optimize the operating conditions of simulated moving bed chromatography. The optimization algorithm's objection function was the normalized production intensity. In addition, the n exponent of purity was used as penalty function. The fitness of real coded genetic algorithm fitness was calculated with steady model of simulated moving bed chromatography. In order to satisfy the constrain conditions of simulated moving bed chromatography, the arithmetic crossover and completelynon uniform mutation were used as the operators of the real coded genetic algorithm. The parameters of real coded genetic algorithm were population size 60, probability of crossover 40%, probability of mutation 10%, and generation 1?000, respectively. With this algorithm, it was easy to find the optimization condition of simulated moving bed chromatography separating progress under nonlinear condition. Then, this optimization algorithm was used to optimize the separation condition of mother liquid of xylitol with simulated moving bed chromatography. It was found that the optimized separation conditions of different purity products could be gotten by changing the value of penalty function coefficient n. When n value changed from 5 to 80, the purity of products changed from 9465% to 9986% with optimized operating conditions. And the computer emulated results and the experimental results were compared when the n value equals to 20. The results showed that they were coincident. The purities of xylitol and xylose from experimental results were 9951% and 9926%, respectively.

After elaborating the theory of the genetic algorithm, including the coding,initial population, fitness function, genetic operation and controls parameter of arithmetic, this paper presents its application to feed diet of fishing in detail. It adopts real coding, linear ranking selection, arithmetical crossover and nonconforming mutation. The results show that this algorithm is better than the optimization of common application mathematics in the feed diet of fishing. It mainly embodies that the unitage cost...

After elaborating the theory of the genetic algorithm, including the coding,initial population, fitness function, genetic operation and controls parameter of arithmetic, this paper presents its application to feed diet of fishing in detail. It adopts real coding, linear ranking selection, arithmetical crossover and nonconforming mutation. The results show that this algorithm is better than the optimization of common application mathematics in the feed diet of fishing. It mainly embodies that the unitage cost price of the former is lower than that of latter. Due to resolving the long degree non-linear problem with more factors, the diet by the genetic algortithm more well and truly accords with the nutrition tontent criterion of corresponding fish.