| Title |
Static vs. Dynamic Control Strategies of Double Skin Systems |
| Authors |
Yoon Seong-Hwan ; Park Cheol-Soo |
| Keywords |
Double-skin ; Optimal Control ; Performance ; Optimization ; Rule-based ; Exhaust Search ; Gradient |
| Abstract |
This study compares three different control strategies for double skin systems. The selected control strategies are (1) rule-based approach, (2) exhaustive search and (3) gradient based search. The fundamental principle of the rule-based approach is “if this, do that” under certain circumstances, and the rules are generally based on expert knowledge. The disadvantage of this approach is that it does not reflect the transient behavior of the system. The exhaustive search or “brute force” search tests all possibilities for the solution. The gradient method uses the derivative of the criterion function. This study aims to investigate the difference in the performance of the systems depending on the different control strategies. The control simulation uses the lumped calibrated model developed in the previous study and an optimization routine to solve for optimal control variables which minimize a cost function over the time horizon. It was found that the rule-based approach works significantly worse in terms of decreasing the cost function in heating and cooling modes. There is a negligible difference between the gradient and exhaustive methods. Considering significant computation time required for the exhaustive method the gradient method is suited for optimal control of double skin systems. |