Self-Organising Particle Swarm Optimisation for Economic Generation Dispatch Incorporating Wind Power

Abstract

The focus of this dissertation is to explore the potential of energy savings through wind power source in coordination with thermal power generation to meet the losses and continuously varying load demand with due considerations to technical constraints imposed by units. The power output from thermal units is considered deterministic whereas wind power output is intermittent due to stochastic nature of wind speed. The intermittent wind power is modelled by discontinuous Weibull probability distribution function. The optimal generation allocation among thermal units and wind based units is based on the operating cost of thermal units and cost of wind power. The wind unit cost accounts for linearly incremental cost due to unused available wind power, penalty cost to account for overestimation and underestimation of wind power and direct cost pertaining to the issue of ownership of wind generators. The proposed optimization model is solved for ten and forty thermal units with valve point loading effect and two, five, one wind unit to meet 24 hour load demand using self-organizing particle swarm optimization technique that can handle the problem of convergence to sub-optimum solutions prematurely. B-coefficient method is used to include transmission losses. The results show that the total cost of system decreases as the wind power scheduling increases and hence the thermal fuel cost decreases. Self- organizing PSO (SOPSO) gives better results.