Catálogo de publicaciones - libros

Gene Expression Programming (GEP) has wide searching ability, simple representation, powerful genetic operators and the creation of high levels of complexity. However, it has some shortcomings, such as blind searching and when dealing with complex problems, its genotype under Karva notation does not allow hierarchical composition of the solution, which impairs the efficiency of the algorithm. So a new automatic programming method is proposed: Gene Estimated Gene Expression Programming(GEGEP) which combines the advantages of Estimation of Distribution Algorithm (EDA) and basic GEP. Compared with basic GEP, it mainly has the following characteristics: First, improve the gene expression structure, the head of gene is divided into a head and a body, which can be used to introduce learning mechanism. Second, the homeotic gene which is also composed of a head, a body and a tail is used which can increase its searching ability. Third, the idea of EDA is introduced, which can enhance its learning ability and accelerate convergence rate. The results of experiments show that GEGEP has better fitting and predicted precision, faster convergence speed than basic GEP and traditional GP.

This paper designs a new kind of structured population and evolutionary operators to form a novel algorithm, Organizational Evolutionary Algorithm (OEA), for solving constrained optimization problems. A simple and non problem-dependent technique is incorporated into OEA to handle the constraints. In OEA, a population consists of organizations, and an organization consists of individuals. All evolutionary operators are designed to simulate the interaction among organizations. In experiments, 4 well-studied engineering design problems are used to test the performance of OEA. The results show that OEA obtains good results both in the solution quality and the computational cost.

Multi-objective problems with parameter interactions can present difficulties to many optimization algorithms. We have investigated the behaviour of Simplex Crossover (SPX), Unimodal Normally Distributed Crossover (UNDX), Parent-centric Crossover (PCX), and Differential Evolution (DE), as possible alternatives to the Simulated Binary Crossover (SBX) operator within the NSGA-II (Non-dominated Sorting Genetic Algorithm II) on four rotated test problems exhibiting parameter interactions. The rotationally invariant crossover operators demonstrated improved performance in optimizing the problems, over a non-rotationally invariant crossover operator.

Two novel schemes of selecting the current best solutions for multiobjective differential evolution are proposed in this paper. Based on the search biases strategy suggested by Runarsson and Yao, a hybrid of multiobjective differential evolution and genetic algorithm with (N+N) framework for constrained MOPs is given. And then the hybrid algorithm adopting the two schemes respectively is compared with the constrained NSGA-II on 4 benchmark functions constructed by Deb. The experimental results show that the hybrid algorithm has better performance, especially in the distribution of non-dominated set.

In a repurchase program, different firms can obtain different gains when they announce a stock repurchase, so a firm needs to know whether announcement is an optimal choice. This paper presents a dynamic, two-player game model with imperfect information, analyzes its further equilibrium condition. The model shows that repurchase announcements have various effects on the firms. High-earnings firms choose to make announcements, whereas low-earnings ones are inclined not to announce.Finally, it gives empirical test for the model to validate the conclusion according to the data of China.

To handle the constrained multi-objective evolutionary optimization problems, the authors firstly analyze Deb’s constrained-domination principle (DCDP) and point out that it more likely stick into local optimum on these problems with two or more disconnected feasible regions. Secondly, to handle constraints in multi-objective optimization problems (MOPs), a new constraint handling strategy is proposed, which keeps infeasible elitists to act as bridges connecting disconnected feasible regions besides feasible ones during optimization and adopts stochastic ranking to balance objectives and constraints in each generation. Finally, this strategy is applied to NSGA-II, and then is compared with DCDP on six benchmark constrained MOPs. Our results demonstrate that distribution and stability of the solutions are distinctly improved on the problems with two or more disconnected feasible regions, such as CTP6.

A new strategy for parameter estimation of dynamic differential equations based on nondominated sorting genetic algorithm II (NSGA II) and one-step-integral Treanor algorithm is presented. It is adopted to determine the exact model of catalytic cracking of gas oil. Compared with those conventional methods, for example, quadratic programming, the method proposed in this paper is more effective and feasible. With the parameters selected from the NSGA II pareto-optimal solutions, more accurate results can be obtained.

The success (and potential success) of evolutionary algorithms and their hybrids on difficult real-valued optimization problems has led to an explosion in the number of algorithms and variants proposed. This has made it difficult to definitively compare the range of algorithms proposed, and therefore to advance the field.

In this paper we discuss the difficulties of providing widely available benchmarking, and present a solution that addresses these difficulties. Our solution uses automatically generated fractal landscapes, and allows user’s algorithms written in any language and run on any platform to be “plugged into” the benchmarking software via the web.

Interactive genetic algorithms depend on more knowledge embodied in evolution than other genetic algorithms for explicit fitness functions. But there is a lack of systemic analysis about implicit knowledge of interactive genetic algorithms. Aiming at above problems, an interactive genetic algorithm based on implicit knowledge model is proposed. The knowledge model consisting of users’ cognition tendency and the degree of users’ preference is put forward, which describes implicit knowledge about users’ cognitive and preference. Based on the concept of information entropy, a series of novel operators to realize extracting, updating and utilizing knowledge are illustrated. To analyze the performance of knowledge-based interactive genetic algorithms, two novel measures of dynamic stability and the degree of users’ fatigue are presented. Taking fashion design system as a test platform, the rationality of knowledge model and the effective of knowledge induced strategy are proved. Simulation results indicate this algorithm can alleviate users’ fatigue and improve the speed of convergence effectively.

Fuzzy mathematical programming (FMP) has been shown not only providing a better and more flexible way of representing the cell formation (CF) problem of cellular manufacturing, but also improving solution quality and computational efficiency. However, FMP cannot meet the demand of real-world applications because it can only be used to solve small-size problems. In this paper, we propose a heuristic genetic algorithm (HGA) as a viable solution for solving large-scale fuzzy multi-objective CF problems. Heuristic crossover and mutation operators are developed to improve computational efficiency. Our results show that the HGA outperforms the FMP and goal programming (GP) models in terms of clustering results, computational time, and user friendliness.