An updating method for the dynamic MIMO model of a controlled technological object

This paper considers the degradation of MIMO models of controlled industrial processes. We propose a method for solving a nonlinear programming problem with an objective function formed by the least squares method according to technological object data. The method involves dynamic process modeling algorithms based on the imposition of the step response of the process. The advantage of this method is the possibility of using passive experiment data to construct an appropriate multichannel model of a technological object. The method is applied to update the multiparameter controller model for the Advanced Process Control system of butyraldehydes oxo synthesis. For real controlled technological objects, this method allows improving the accuracy of process modeling and the performance of automatic control, reducing the human factor, and increasing the overall economic efficiency of the production process.