China 
Africa 
Explore chapters and articles related to this topic
Introduction
Published in Yi Chen, Yun Li, Computational Intelligence Assisted Design, 2018
Particularly, in 1992, Koza [Koza (1992), Koza (1994)], utilized GAs to evolve programs to perform certain tasks in a process called genetic programming (GP). GP is a technique of enabling a GA to search a potentially infinite space of computer programs rather than a space of fixedlength solutions to a combinatorial optimization problems. These programs often take the form of Lisp symbolic expressions, called Sexpressions. The idea of applying GAs to Sexpressions, rather than combinatorial structures, is due originally to the work of Fujiki and Dickinson [Fujiko and Dickinson (1987)] and was brought to prominence through the work of Koza [Koza (1992)]. Table 1.4 lists a few works on GP in recent years.Evolution Strategy
Electric Field Optimization
Published in Sivaji Chakravorti, Electric Field Analysis, 2017
Kitak et al. [45] described an algorithm for the design of medium-voltage switchgear insulation elements using numerical calculations on the basis of finite element method in connection with evolutionary optimization methods. Differential evolution and evolution strategy algorithms were used for optimization. The task of both optimization algorithms was to find an adequate capacitance of the voltage divider and the optimal distribution of electric field strength. The highlight of the work was simultaneous use of parametric representation in geometry, a novel mesh generator, numerical computation with finite element method (FEM) and a genetic optimization algorithm. The proposed methodology thus represented a generalized method of optimization for various objective functions of switchgear elements.
Evolutionary Mechatronic Tool
Published in C.W. de Silva, Mechatronic Systems, 2007
To solve a problem by GAs, the following four elements have to be incorporated: A representation scheme is needed for possible solutions. It should have a chromosome-like structure, so that each part of it can be viewed as a gene. A particular gene in a chromosome stores some characteristics of that chromosome. If this gene is transferred to the next generation, the particular characteristic will be inherited in that next generation.Reproduction operations have to be developed so that each reproduced individual inherits some specifications of its parents.A fitness evaluation methodology is necessary to reflect how well a trial solution satisfies the predefined performance requirements.An evolution strategy is necessary to insure that good solutions with higher fitness have a greater chance to participate in the reproduction operations in creating the next generation.
A possibilistic finite element method for sparse data
Published in Safety and Reliability, 2018
A. Dridger, I. Caylak, R. Mahnken, E. Penner
For calculating such quantities of interest considering optimality conditions with simple bounds (Bertsekas, 1982; Mahnken, 2017). Furthermore, combinations of the evolution strategy, the gradient methods and the Monte Carlo method are frequently used in order to solve the alpha-level optimisation problems (Möller & Beer, 2004). Moreover, if available, a monotonicity of the mapping model with respect to the design parameter vector may also restrict the location of the optimum points. For this reason, it may be advantageous to exploit the monotonicity in order to avoid costly optimisation methods (Möller & Beer, 2004) as can be seen in the representative example in Section 5.
Optimal thrusters steering for dynamically reconfigurable underwater vehicles
Published in International Journal of Systems Science, 2019
Maxence Blond, Daniel Simon, Vincent Creuze, Olivier Tempier
Hence solutions may be searched for using ‘global optimisation’ algorithms likely to find the global extremum in the whole search space by avoiding being trapped in local extrema. This is, for example, the case of Simulated Annealing algorithms where a controlled random search around a guessed solution allows for escaping local extrema. Among others, Genetic Algorithms (GA) are currently popular algorithms for searching a global optimum in a large multi-variable non-convex optimisation problem. They rely on a bio-inspired model based on an evolution strategy, where stochastic mutation and selection rules allow for iteratively searching better solutions from a population of initial candidates.
Antenna Array Pattern Synthesis Using Metaheuristic Algorithms: A Review
Published in IETE Technical Review, 2023
Population-based metaheuristic algorithms are further classified into evolutionary, swarm-based [35], and other algorithms as shown in Figure 6. Evolutionary algorithms are inspired from the biological evolution phenomena of nature. GA, biogeography-based optimization (BBO), and evolution strategy are some of the well-known evolutionary algorithms. Swarm intelligence algorithms are inspired from collective intelligence of group of agents such as birds, fish, ants, bees, etc. These algorithms are less complex and have few parameters to adjust in comparison to evolutionary techniques.