Comparative Study of the Techniques for Optimization of Fuel Cost in Thermal Power Plants in Nigeria

This paper presents a comparative study of optimization techniques for minimizing fuel cost in thermal power plants within the Nigerian 330 kV power system. Three classes of optimization techniques are explored: heuristic, hybrid, and enhanced hybrid. The heuristic class encompasses Pattern Search Techniques (specifically, the Generalized Pattern Search method and the Generating Set Search method) and the Genetic Algorithm. The hybrid class combines Pattern Search and Genetic Algorithm. Further enhancement is achieved by integrating the hybrid approach with either an Interior Point Solver or Sequential Quadratic Programming (SQP), forming the enhanced hybrid class. In this work, a hybrid algorithm and heuristic algorithms were applied to solve the economic load dispatch problem, considering practical constraints such as power balance (equality constraint) and plant generation capacity limits (inequality constraints). An optimal economic dispatch analysis was conducted for varying load demands of 2500MW, 3000MW, 3500MW and 4000MW. The techniques used for solving the economic load dispatch problem were implemented using the optimization toolbox contained in MATLAB 2017, version 9.2.0. Results demonstrate that the enhanced hybrid technique, specifically the Pattern Search/Genetic Algorithm with Sequential Quadratic Programming, achieves the best minimum fuel quantity, yielding optimal values of 83581.3024 MMBTU/hr, 84661.9081 MMBTU/hr, 89721.6362 MMBTU/hr, and 95694.4339 MMBTU/hr for the respective load demands. Furthermore, the cumulative fuel consumption (including losses) with this enhanced hybrid approach is consistently lower than that of the other compared techniques. Therefore, the Pattern Search/Genetic Algorithm with Sequential Quadratic Programming is considered the most effective of the techniques compared in this work for solving economic dispatch problems of generation, offering significant potential for fuel cost savings.