Summary

An artificial neural network is a network made of a large quantity of simple interconnected processing elements analogous to neurons in the human brain. This network is based on the understanding of the neuromorphic structure, functioning and information process capability of the brain. The research of neural network can be dated back to the 1940s when a psychologist and a mathematician summarized the basic neuromorphic characteristics of biological neurons and mathematically described a structural model of a neuron known as the M—P (McCulloch and Pitts) model. This effort heralded a new era of neural science research. The study of artificial neural network and its engineering application can be traced back respectively to the 1960s and 1980s. Neural network is now beginning to permeate through many engineering disciplines including structural dynamics and modal analysis. It has found applications in pattern recognition, intelligence control, signal and graphics process, system identification nonlinear optimization, vibration control, information process and market prediction.