ELD, LOAD FLOW ANALYSIS & OPTIMAL LOAD FLOW STUDIES USING INTELLIGENT OPTIMIZATION TECHNIQUE

Abstract

As protection of the environment gains more and more attention, the economic emission dispatch problem has emerged as an intriguing and crucial task in the power system. In essence, the EED problem is a multi-objective optimization problem, which concurrently reduces fuel costs and pollutant emissions while also satisfying certain system constraints like power balance and generating restrictions. The thesis developed a method based on the meta heuristic Particle swarm optimization for single objective, bi objective and multi objective power system optimization problems based on cost of fuel function, emission criterion function and operational constraints of the generating system. In this thesis comprehensive, a systematic and chronological effort has been made for literature review from 1983 to 2019 with historical development, addition of new parameters, tuning or refinement of parameters and its variants for different optimization problems with constraints, multi-objectives. In addition it also covers the parallel PSO, its hybrids, communication topology and for multi-objective problems strategy used for parallel computing are covered in detail. The new variant Perfectly Convergent Particle Swarm Optimization (PCPSO) developed is an intelligent algorithm which does not get trapped in local optima by using personal best value along with new parameters and new velocity update equation for better exploration in the search space. The velocity clamping effectively helped to control the maximum velocity of the particles from explosion state and align them towards the true global optimal with increased computational efficiency in less time. It has been implemented on uni-modal, multi-modal with local optima and noisy function. PCPSO technique was used to solve combined economic and multiple emissions dispatch scenarios with max-max price penalty factor using quadratic functions, while considering the implications of emissions. Implementing this method on three different standard IEEE test systems , such as the IEEE six-unit system, IEEE ten-unit system, and IEEE forty-unit system, and comparing the results with other meta - heuristic algorithms, allowed for the evaluation of this algorithm's effectiveness. Moreover same strategy is used for solving combined economic and multiple emissions dispatch problems while taking into account the impacts of various pollutants with seven price penalty factors using cubic functions. Cubic cost functions are more accurate and show the actual response of all thermal units. This algorithm has better search capabilities with strong convergence characteristics that minimizes the cubic cost and cubic multiple emissions functions at various load demands with minimum transmission losses for an IEEE 30 bus,6 generators test system. PCPSO was able to provide balanced exploration and exploitation in the search space. The suggested algorithm's effectiveness was tested on different separate test systems, both small and large, with differing degrees of complexity. In the realm of Multi Area Load Dispatch (MALD), this technique aids in the refinement of the global solution as well as local search. Energy transfer between locations and fossil fuel emissions from generating units are key concerns. As a result, the goal of MAPD is to minimize the overall generation cost of the areas while also lowering pollutant emissions. To appreciate the value of resolving the entire region into tiny regions, a comparison with the Single Area load Dispatch (SALD) method is made. The Price Penalty Factor (PPF) method is used to reduce a multi-objective optimization problem into a single-objective optimization problem while satisfying its different equality and inequality constraints. System security is ensured by keeping the tie line transfer limits between areas and the constraints in the multi area load dispatch. Various benchmark IEEE models were applied to this algorithm to test the developed PCPSO effectiveness and reliability.