Computational Intelligence Integrated Approaches Based On Swarm Intelligence And Neural Network

Since the invention of the computer, using a computer instead of the human brain deal with the problem has been the ultimate goal of artificial intelligence scholars.But the human brain is a complex system,we still can not fully understand many of its working mechanism today. So in order to achieve true artificial brain,there is still a long way to go. In the course of development of artificial intelligence, people's research is very broad.Although it can not accurately simulate the human brain, but we can still simulate the human's nervous system, the nature of biological evolution, characteristics of animal populations to take advantage of computer to solve problems.Computational Intelligence is an important area of artificial intelligence.It is a kind of simulation calculation method which do research and simulation in intelligent behavior from the bottom of biological point.It has many advantages, such as self-learning, self-organizing, adaptive, robust, simple, parallel processing, general and so on. So far,there are variety of computational intelligence methods,the two main methods of computational intelligence methods: neural network algorithms and swarm intelligence algorithm will research in this paper. Neural networks is a intelligent algorithm simulate the cerebral nervous system, and swarm intelligence methods is a intelligent algorithm simulate intelligent group behavior. These two methods now have been very widely used in the actual lives and solve previously intractable problems. Neural networks and swarm intelligence algorithm's main application areas are different.Neural network is often used in modeling and pattern recognition, and swarm intelligence algorithms are often used in optimization problems. However, in some comprehensive issues, we only use neural networks or only use swarm intelligence algorithms can not completely solve the problem.That requires us to combine the two together to achieve the purpose of problem-solving.This is called computational Intelligence integrated approaches based on neural networks and swarm intelligence algorithm.In this paper, we first give a brief overview of neural networks and swarm intelligence algorithms in their origin, history, development, research status and their respective advantages. Then we have a variety of neural network model details, and neural network modeling for microbial fermentation engineering.Then we give details of genetic algorithm (GA), particle swarm optimization (PSO) and the quantum behavior of particle swarm optimization (QPSO) and other swarm intelligence algorithms, and we proposed HSQPSO based on QPSO algorithm.In simulation experiments, we find that HSQPSO has obvious advantages compared to PSO and QPSO.After introduced the neural network and swarm intelligence algorithm, we propose a problem and it make us know what is computational Intelligence integrated approaches. Finally, we make the prospect of future development on neural networks and swarm intelligence algorithms.