Method of simulating an industrial process, and use thereof for testing the operation of a controller

A simulation method for testing an automatic controller by simulating an industrial process, that is intended to be controlled by said automatic controller via control signals and state signals respectively emitted and received by the automatic controller, that causes the value of each physical magnitude characteristic of the state thereof to vary as a function of the state of said control signals, and that reports its state to said controller by means of said state signals emitted by sensors as a function of the values of the respective physical magnitudes that they monitor, the method being implemented in a simulator whose inputs and outputs are connected respectively with the inputs and outputs of the controller, via the control signals and state signals, wherein the method comprises the following steps:

• a) defining in the simulator a functional model of process monitoring for the industrial process with the state signals representatives of the physical magnitudes;
• b) defining a functional model of process control for the industrial process;
• c) initializing the state signals of the industrial process;
• d) calculating upon the state signals for each physical magnitude P.sub.i, a rate of change V.sub.i which is a function of the state of the control signals emitted by the controller;
• e) determining for each physical magnitude P.sub.i whose rate of change is not zero, a time interval .delta.t.sub.i during which the value of the physical magnitude can vary without any of the sensors monitoring said magnitude P.sub.i changing state;
• f) calculating the time interval .delta.t which is defined as the shortest of the time intervals .delta.t.sub.i determined in the preceding step;
• g) updating the value of each of the physical magnitudes P.sub.i, by incrementing each of them with the value of the product of the time interval .delta.t multiplied by the corresponding rate of change V.sub.i ;
• h) determining, as a function of the values of the physical magnitudes updated in the preceding step, which sensors change state, and updating for the controller the corresponding state signals;
• and
• i) waiting for a time lapse referred to as the waiting time, the value of which is parameterizable, prior to repeating the method from step d).