Explore chapters and articles related to this topic
Evolutionary fuzzy systems
Published in Ali Zilouchian, Mo Jamshidi, Intelligent Control Systems Using Soft Computing Methodologies, 2001
R. Mohammad, T. Akbarzadeh, A.H. Meghdad
There are various approaches to evolutionary optimization algorithms including evolution strategies, evolutionary programming, genetic programming and genetic algorithms. These various algorithms are similar in their basic concepts of evolution and differ mainly in their approach to parameter representation [1]. Genetic algorithms (GA), in particular, is an evolutionary method which has performed well in noisy, nonlinear, and uncertain optimization landscapes typical of fuzzy systems. In this chapter, we will explore further why and how GA is used for optimization of fuzzy systems and, in particular, fuzzy controllers. Various issues such as determining the set of parameters, designing the transformation function for representing the parameter space in genetic domain, creating the initial population, and determining the fitness function will be discussed. Finally, application of GA will be illustrated in optimizing fuzzy control of a DC motor.
New Approaches to Mobile Ad Hoc network Routing: Application of Intelligent optimization Techniques to Multicriteria Routing
Published in Jonathan Loo, Jaime Lloret Mauri, Jesús Hamilton Ortiz, Mobile Ad Hoc Networks, 2016
Bego Blanco Jauregui, Fidel Liberal Malaina
There are different techniques included in this family of optimization algorithms. Evolution strategies work with vectors of values and emulate the mutation process over the possible solution population to explore the feasible space and to avoid holding up in a local minimum. Therefore, it is an abstraction of the evolution at the individual behavior level. Evolutionary programming is a stochastic optimization technique that focuses on the behavioral link between ascendants and descendants instead of replicating genetic operators observed in nature.
Cooperative coevolution of expressions for (r,Q) inventory management policies using genetic programming
Published in International Journal of Production Research, 2020
Rui L. Lopes, Gonçalo Figueira, Pedro Amorim, Bernardo Almada-Lobo
GP is a general purpose EA, proposed and popularised in the early 1990s by Koza (1994). This technique evolves computer programmes that solve problems, instead of evolving solutions to problems (of which are examples Evolution Strategies, and Genetic Algorithms). In this computational model, the programmes are represented by trees (an example of such representation is provided in Figure 1), upon which the variation operators act, and which are artificially selected according to their phenotypic effect (the programme execution result). The core procedure is similar to other EAs: (1) randomly define an initial population of solution candidates; (2) select, according to the fitness, some individuals for reproduction (sexual – crossover – and asexual – mutation) with variation; (3) define the survivors for the next generation; (4) repeat steps (2) and (3) until some termination condition is fulfilled.