An Intelligent Automatic Generation Control For A Hydro-Power System

No Thumbnail Available



Journal Title

Journal ISSN

Volume Title


Addis Ababa University


In an interconnected power system, any sudden small load perturbation in any of the interconnected areas causes the deviation of area frequency andtie line power. Automatic generation control (AGC) plays a very important role to maintain system frequency and tie line power flow at their scheduled values during any load perturbation.Classical controllers like PI and PID controllers are very simple for implementation and give better dynamic response. But, they have larger response time, lack of efficiency, poor handling of system nonlinearities and are not suitable for complex, uncertain, high order and time delay systems. Intelligent controllers can provide a high adaption to changing conditions and have ability to make decisions quickly by processing imprecise information. They can also perform effectively even with nonlinearities. This thesis work presents the application of an intelligent AGC such as artificial neural networks (ANN) and fuzzy logic controller (FLC) for a hydropower system. The ANN techniques are used for AGC of interconnected hydro power systems. The feed forward neural network controllers are developed and trained using Lavenberg-Marquardt (LM) back propagation algorithm under supervised training method with adequate amount of data that are generated by using optimal control strategies. The programming was done in MATLAB and the resultshows an improved performance in terms of rise time, settling time, steady state error and overshoot. The designed FLC gave a response with settling time 45 sec and a frequency deviation of 2.3per unit (p.u). This is a better result than that of a PID controller with settling time and frequency deviation of 125 sec and 2.6p.u respectively. The designed ANN controller also gave a performance that is close to optimal controller and superior to the integral controller as expected.



Automatic Generation Control (AGC), Artificial Neural Networks (ANN), Fuzzy Logic Controller (FLC), Lavenberg-Marquardt (LM), Proportional Integral Derivative (PID)