Process Model Development For Computer Control

Formally, a mathematical model could be defined as: A set of equations whose solution will approximate the response of a physical process or system to a set of stimuli which appear as parameters in the equations. The purpose of a mathematical model is to permit the rationalization of experience to proceed by simulation rather than by physical experimentation. In the application of digital computer control to an industrial process, the mathematical model of the process (subsequently called the process model) is needed to calculate the present process performance from the measured variables and to predict the response of the process to a control change. Functionally, the cardinal element in process model development is a concise set of control objectives. These, besides implicitly defining the boundaries of the system to be modeled, will determine the accuracy demanded of the model its nature: dynamic or steady-state, and so forth, The objectives should explicitly define what process efficiencies, production rates, qualities are to lie within the scope of the computer control project, thereby providing a basis for simplifying assumptions in the development of the model. This requirement, a limited set of control objectives, is so often disregarded in formulating industrial control problems that its repetition can scarcely be avoided.