Advanced adaptive control applications (in industry)

The implementation of an adaptive controller using an orthogonal function space to represent the plant transfer function is described. This approach for modeling continuous functions is similar to the common Fourier series orthogonal function set used to approximate periodic functions. It is shown that the use of orthogonal filters permits rapid transfer function identification with a minimum of prior process information while maintaining robust control under a wide range of operating conditions. The mathematical model, the development of a computer algorithm, and the results of several applications at industrial sites are presented. The basics of proportional-integral-derivative control and adaptive control are reviewed.<>