Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Lane Department of Computer Science and Electrical Engineering

Committee Chair

Parviz Famouri


The imbalance between electrical loads and power supplied by the system generators causes the frequency deviations in a power system. Maintaining the frequency close to its nominal value as well as in its allowed deviation range is the first objective of the automatic generation control (AGC). Nowadays, in interconnected power systems, several control areas are connected to each other by tie-lines and power is transferred between control areas based on a specific schedule. The second objective of automatic generation control is to keep the tie-lines power flow close to their secluded values.;An accurate and realistic load frequency control (LFC) model is very essential to have an effective and adaptive AGC strategy. The first objective of this thesis is to present the importance of considering communication delay in LFC model missing in most of the studies investigating AGC and its performance using different methods and optimization techniques. The second objective of this thesis is to present a comprehensive LFC model, which contains all of the physical constraints such as governor dead-band, generation rate and delay of communication links. The third objective is to evaluate different controllers and performance index criteria used in conventional AGC. Finally, the last objective is to introduce an adaptive performance index criterion cable of defining settling time and overshoot which cannot be applied by other performance index criteria.;Different optimization methods have been used to optimize the performance of AGC such as genetic algorithm, fuzzy logic and neural networks. Genetic algorithm has been used widely in LFC studies so it is chosen to be employed in this study to optimize the performance of controllers in the utilized AGC scheme. Integrator controller is the most common controller employed in LFC studies because of its design simplicity, however, in this thesis proportional-integral-derivative (PID) controller is employed to obtain the best performance.;This study shows that without a precise and detailed LFC model, results of different techniques or strategies used in AGC will not be accurate and practical even when they are derived by optimization methods. Moreover, it is shown that PID controller has the best performance in comparison with other controllers used in LFC studies when physical constraints are not considered in the LFC model. Furthermore, a robust GA based control system is designed considering all of physical constraints for a three-area power system and the simulation results show that it can track the load change and restore the frequency of all control areas to the nominal value effectively. Different performance index criteria are evaluated and results show that in specific cases they cannot be completely accurate or reliable to assess the performance of AGC schemes. Finally, an effective and adaptive performance index is introduced and simulation results validate its effectiveness and reliability.