Applications of modern optimal control theory to environmental control of confined spaces and life support systems: Part—5 optimality and sensitivity analysis

Abstract

The sensitivities of a cabin with temperature control which has been considered in the first four parts of this series are examined. These include the sensitivities to (1) parameter variation, (2) change in dimensions of mathematical models and (3) change in constraints on state variables are examined. The system is not sensitive to the variation of the parameter (the recycle ratio of air) when the magnitude of the variation is small. The effect, however, is noticeable when the variation is large. The effect of the change of the parameter (the ratio of the time constants of the system proper to that of the heat exchanger) is very small. The effect of variation of the parameter (the volume fraction of the first pool in the two CST's-in-series model) on the optimal conditions is substantial. There is a small but not negligible effect of the dimensional change in the system equations, which is caused by neglecting the time constant of the heat exchanger. The complexity of the model describing the system component has a definite effect on the predicted performance of the system. The effect is apparent for the particular models considered here, which are the one CST model and the two CST's-in-series-model.