`Pseudodiagonalisation¿ and the inverse-Nyquist array method

A new approach to the design of linear multivariable control systems using the inverse Nyquist array method is proposed and applied to two examples. The technique generalises diagonalisation at zero frequency to an arbitrary finite point on the D contour by minimising the sum of the squares of the moduli of the off-diagonal terms in each row of the inverse open-loop transfer-function matrix. This approach is simple to implement and is especially suitable for interactive computer-aided design. As a large range of possible solutions can be evaluated quickly, this leads either to the choice of a matrix compensator conforming to the requirements of the pertinent stability theorem or to the practical conclusion that the method does not yield a solution to the form considered.