The objective of this project, ie, implementing Model Predictive Control in Single Board Heater System using Scilab was successfully achieved. Several experiments were successfully performed using the developed SCILAB MPC algorithm for both positive and negative step changes in both temperature-set-point and the disturbance variable (fan).
In addition to the above objective, we also tried studying the effect of weighting factors (tuning parameter) and control horizon parameter. We observed and concluded that increase in values of We (error weighting factor), increases oscillations and decreases settling time, while decrease in We leads to opposite effect. Wu (manipulated variable weighting factor), on the other hand has an opposite effect. It decreases oscillations and increases settling time with increase in its value. Hence, better control is obtained for high value of We and low value of Wu.
Thus, with this project, we were able to implement MPC successfully and also were able to comment on the general preferred tuning parameters (weighting factors for error and manipulated variable).
rokade
2017-04-23
