Neural networks in control systems

D.H. Rao, M.M. Gupta, H.C. Wood
IEEE WESCANEX 93 Communications, Computers and Power in the Modern Environment - Conference Proceedings  
university of Saskatchewan, saskatoon, Cnnnnn. SRI ow0 , ABSTRACT The purpose of this papa is to provide an overview of neural network s t r u c~r e s used for system identification and control. Due to the complexity and diversity of the Propemes of biological neurons, the task of compressing their complicated characteristics into a model is extremely difficult. Towards this goal, an artificial neuron, also called a 'unit', was developed which receives its inputs from a number of other neurons
more » ... r of other neurons or from the external world. A weighted sum of these inputs constitutes the argument of an 'activation' function. This is a simple, but useful first approximation of a biological neuron. Using this model, many neural s t " r c s usually xeferred to as feedfarward neural networks have been reported in the literature. Many of these networks use only prcsent values of inputs, and are themfort called instantaneous or static systems. A natural extension of static networks is the dynamic or recurrent neural network which incorporates feedback in its structure. No general theory for dynamic nunal networks has yet developed like similar to that for static networks. With the parallel growth in the field of fuzzy logic, many neural models encompassing the principles of neural networks and fuzzy set theory are being dev+oped. In this paper we have made an attempt to provide the baslc concepts of static, dynamic and fuey neural struchrres.
doi:10.1109/wescan.1993.270588 fatcat:l4qvwbxfvfd65pwqzprg4wsxuy