NERC Compliant Adaptive Immune Based Load Frequency Control with Varying Wind Penetration

Energy demand is increasing rapidly. It is a vital factor that determines the civilization in 21st century. The rapid depletion of fossil fuels, their cost as well as environmental issues due to their emissions, call for speedy development of the renewable energy sources (RESs). These renewable sources provide diversification in energy and avoid the construction of large generating stations and new transmission lines.;Wind power has influential penetration over the years among different renewable energy sources. With a high penetration of wind power, power system undergoes severe frequency deviations. It causes the under and over frequency relays to operate leading to generation and load disconnections. In fact, traditional wind generators do not participate in the Automatic Generation Control (AGC) or the frequency regulation. Thus, the total inertia and the frequency robustness of the power system reduce when there is high wind penetration. The Area Control Error (ACE) variability also increases due to the increase in unscheduled flows which causes deterioration of the control performance standards. These different issues have led to increased research effort in this sector.;In this thesis, the design and operational features of the power system with conventional and non-conventional generation resources is studied in a two-area and a four-area system. The thesis develops an algorithm to mitigate the frequency control issues in a highly wind penetrated system and improves the performance of an AGC system. It is capable of complying with the North American Electric Reliability Council (NERC) control performance standards (CPS) as well as reducing the machine oscillations. These controllers are capable to deliver best operating conditions over a wide range of operating conditions as they generate real time gains dependent upon the load variations.;The thesis demonstrates the efficiency of the suggested algorithm in different test beds with variation in DFIG wind penetration levels and draws some important conclusions. To prove the effectiveness of the proposed algorithm, the thesis develops a genetic algorithm to generate the optimal gain of the test bed and compares the two. In addition, the proposed controller is also compared with the fixed gain controllers with high gains and low gains.