Research On Viral Evolution Based RNA Genetic Algorithms And Applications

Genetic Algorithm (GA) is a stochastic global optimization technique which simulates natural evolution process, and it provide a general framework to solving complex optimization problems without depending on the specific problem area. It is widely employed in the field of Pattern Recognition, Image Processing, Data Mining, Control Systems, Chemical Process, etc. However, traditional genetic algorithms have some drawbacks, such as poor local searching ability and premature convergence etc. Inspired by RNA molecular coding and operation, viral evolution based RNA genetic algorithms and applications are studied. The main contributions of this thesis are as follows:(1)In the latter generations of the evolution, population individuals are becoming more and more similar with evolution, which is prone to premature convergence. In order to overcome this drawback, a viral evolution based RNA genetic algorithm VERNA-GA is proposed. The performance of this proposed RNA genetic algorithm is validated by some typical benchmark functions.(2)Combining VERNA genetic algorithm with sequential quadratic programming (SQP), a hybrid optimization algorithm is proposed to solve complex nonlinear optimization problems. Two new crossover operators and a dynamic mutation are designed. The numeral results of some typical benchmark functions show the efficiency of the proposed algorithm.(3)The hybrid optimization algorithm is applied to the short-term gasoline blending scheduling problem. The scheme can quickly find out a solution with higher profit.