Neurocontrol: concepts and applications

The author provides a brief overview of neural network models, highlighting their distinctive features, and describes several ways in which neural networks have been used in control systems. In addition to their learning capabilities, artificial neural networks provide several features that are important for control applications. They can be used for process identification and control by emulating an existing controller, for direct and inverse process modeling, for direct adaptive control, for optimized nonlinear controller development, and for process structure and parameter identification