الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 140 |  |  |
| | | from | 140 | | |
Abstract
Load frequency control LFC is considered as one of the main topics which need to be carefully studied for any power system. In an interconnected power system, as a load demand varies randomly, both area frequency and tie-line power intercharge also vary. The objectives of LFC are to minimize the transient deviations in these variables and to ensure their steady state errors to be zeros. When dealing with the LFC problem of power systems, unexpected external disturbances, parameter uncertainties and the model uncertainties of the power system pose big challenges for controller design.
In this thesis two types of controllers have been applied to the LFC problem. The first type is PID controller which has been applied to thermal single area power system including nonlinearities with three different cost functions. The PID has been tuned via ant colony optimization ACO technique which is considered one of the newest AI methods based on the nature of ant`slife and the method it searching for the food. Results have been compared with another PID based on Zeigler Nichols ZN method. The second control technique is the robust H controller which has applied to the same model but without nonlinearities and applied also to another hydro plant. Results has been compared wih ACO based PID tuned via the norm cost function. The control signals have been plotted on the same figure for easily comparison between them.
The first method has been applied to a hydro-thermal two area power sytem in presence of nonlinearities with the decentralization method which based on trial and error, while the second controller has been applied to the same model without nonlinearities using two methods, centralized and decentralized. The decentralized method based on te decomposition concept which divides the overall system into two subsystems and a controller is designed for each one separately and then recombined a gain.
Different cases had been applied to previous models to reveal the effectiveness and robustness of the proposed controllers. In this thesis we kept on applying realistic load disturbance nature and values to simulate what happens actually in real power systems.