Optimal Power Flow for Units with Non-Smooth Fuel Cost Using Population Based Methods

Abstract

The optimal power flow (OPF) is being used to find the optimal settings to operate the system. When the operating cost is minimized, the generation schedule is calculated by OPF. Traditionally, the cost function of each generator is represented by a simple quadratic function. However, thermal units are sometimes made to run on multiple fuels like coal, natural gas and oil. The multiple fuel options lead the cost function to be piecewise quadratic cost functions. The discontinuities are further added in the cost function when valve point loading effects are included. The optimal power flow with units having valve-point loading and ramp rate limits is represented as a non-smooth optimization problem having complex and nonconvex features.The work reported in this thesis is carried out with the objective to make use of population based methods (EP and PSO) for solving the optimal power flow (OPF) problem for units with non smooth fuel cost function. The effect of both the ramp rate limits and valve point loadings, which are the typical discontinuities, have been incorporated. The objective in the OPF problem has been decided as minimization of total cost of real power generation. The IEEE-30 bus system is considered to test the algorithms. Two case studies have been considered for different sets of cost curves and load demand. The effectiveness of the developed algorithms has been tested for quadratic, piecewise quadratics and quadratics with sine components