Digital simulation of continuous processes using limited computer resources.

The thesis deals with the use of a small digital computer to model dynamic continuous systems, with particular reference to problems arising in the chemical industry. The block structured simulation program developed for this work is known as Simulated Hybrid Analogue Computer (SHAC). Digital computers have largely superseded analogue computers for general purpose simulation of medium sized problems, the analogue being used mainly for teaching purposes or for very large, high speed problems. Historically, the digital computers used for this work have been the large service computers and the trend in simulation programs has been towards greater sophistication. This work has been in the opposite direction and it is demonstrated that, with carefully chosen techniques, a small digital computer may be used conveniently and economically to simulate medium sized 'analogue' models. This is particularly valuable where access to a service computer is poor, yet a small computer is readily available. The techniques used in designing the simulation program are carefully evaluated and a novel integration technique which is used is shown to be highly efficient yet simple to implement. A number of special facilities are described which facilitate the modelling of process control systems. The SHAC program is tested on three types of problem and is found to give reliable results at an acceptable speed. Another process control problem is run on SHAC and on three other programs; SHAC compares favourably on reliability, economy and speed. SHAC could readily be extended to carry out Real Time Simulation (using actual equipment) and automatic optimisation; techniques for doing this are discussed. Although the program has been written for a specific computer, the techniques are applicable to most process control and mini computers. The descriptions given in this thesis would enable a simulation program to be produced for another small computer very easily