Stability in general refers to the ability of a system to continue in normal or almost-normal operation after the occurrence of an unanticipated event. In reference to electric power systems, such events may be a fault (short-circuit), opening of a transmission line, sudden or unforeseen loss of load or generating capacity, loss of synchronism by a generator or motor, or increase of load beyond a limit established by the system parameters.

In this research study, the stability discussions focused on the transient and dynamic periods. Transient stability studies deal with the system conditions within a few cycles after the occurrence of a fault. Dynamic stability refers to conditions existing following the transient stability phase, if the oscillations do not die out, that is the post-fault period. Hence, the study is made by starting with the system conditions as determined by a load flow study and changing a parameter such as the operating conditions of the plant, load variations or the different disturbances in the form of faults, to be introduced into the plant model.

From the previous results and discussions, the stability performance of the plant in terms of the transient period is improved with the P+LQG+D controller since this controller helps to speed up the transient period of the plant and thus, with a shorter transient period, lesser lossess will be incurred and the efficiency of the plant will be greatly increased. During the dynamic period, the LQG controller has proven to be a better controller since the parameters of the plant reached steady-state faster as compared to other controllers implemented.

The next aspect of the plant study that is of concern is the robustness of the plant system. Robustness usually refers to the effect of a change in the model represented using transfer function.

Generally in practical applications, the controller is designed before implementing it into the actual plant. Hence it is important to observe how well the controller can cope with the change in the operating conditions of the plant since different controllers usually have different degrees of robustness. In addition, the parameters of the plant might be inaccurate during modeling and thus, the controller designed based on a certain plant model may not be in accordance to the actual plant model. This further reinforces the importance of study the robustness of the controller.

In this research study, the parameters of the plant such as the state-space matrices, will be fine tuned using trial and error to study the effect of the different combinations of these matices. Using this method, a set of the state-space matrices will be derived and chosen that produces the most desirable degree of robustness.