Wang Jingchun

A Modified PSO Algorithm with the Concept of Intrinsic Repulsive Field

An improved Particle Swarm Optimization Algorithm is proposed in this article. Inspired by the existence of Intermolecular Forces in physics, we bring particles that are influenced by repulsion in this Algorithm to keep the population active at all the stages. Meanwhile, the convergence speed can be assured because of the existence of particles that implement Standard Particle Swarm Optimization. Experimental results show that this improved algorithm can effectively solve the local optimal problems.

A Modelling Method Based on the Wavelet Transform and Artificial Neural Networks

Abstract A new modelling method called Discrete Wavelet Transform Model Identification (DWTMI) algorithm is presented. First, the basic concept of the wavelet transfonn is introduced. It is adopted to process and reconstruct the input and output data of a low sample rate system. An Artificial Neural Network (ANN) model of this kind of system is established from the processed data, which can give out high sample rate data. The process control quality can be improved with this model. The model is also useful for unit operating optimization. Extensive simulation results show that the algorithm is effective and feasible. The method is also successfully applied to a practical example