Computational Intelligence And Its Application To Engineering

Computational intelligence is one kind of biological modeling computational method, which carries on the simulations and the researches of intelligent behaviors from the bottom of biology. It has developed the traditional computation pattern, with characteristics of self-study,self-organization,self-adapted and merits of simple, general, high-robustness, suitable for parallel processing and so on.. It provides a new approach to solve many complex problems and has obtained extensive researches and applications. Based on the research of the main three computational intelligent methods: Artificial neural network, Genetic algorithms, Fuzzy system, this article has done some researches on the engineering applications. The main studies including the following three aspects:1. An engineering solution to Taylor scries expansion coefficients based on BP neural network. Taylor series plays a very important role in solving non-linear problems. Using the high non-linear mapping and ability of self-study, a new solution, artificial neural network, which is suitable for engineering applications, is proposed. Neural network model of Taylor series is established in this article. Via the experiments of given functions, the results showed that it is viability and quickly. In the methods, we only need the sample space of the objective function. So the method has the value of application in industries.2. Genetic algorithm applied in the optimization of ordering and transporting of steel tube. Aim at the optimization problem, a second time programming model for the ordering and transporting of steel tube is developed. Genetic algorithm to solve the model has been discussed. On the basis of analyzing the disadvantages of genetic algorithm approach, a new method has been proposed, which is called genetic simulated annealing. The experimental result shows that this method, which is called genetic simulated annealing approach, is feasible and efficient for finding the optimum parameter in the ordering and transporting of steel tube and provides a good way to solve the problem.3. The speed adjusting approach for DC motor based on fuzzy control is designed and realized in this paper. On this base, the principle and strategy of the comprehensive control that combines fuzzy control with classic PID control method and the realization plan is given. A set of simulations and the corresponding research about the fuzzy control and fuzzy adaptive PID control is carried on by using Matlab/Simulink. The results show that the fuzzy adaptive PID designed in this paper has an improved control performance in real-time, robustness, high accuracy and so on. It provided a method to get a better control performance for a complex DC adjustment system.