Please use this identifier to cite or link to this item:
http://hdl.handle.net/10397/89634
Title: | Advanced heuristic optimization algorithms for optimal reactive power planning and dispatch in power systems | Authors: | Niu, Ming | Degree: | Ph.D. | Issue Date: | 2020 | Abstract: | This thesis develops three advanced heuristic optimization algorithms (HOAs) for power system reactive power planning and dispatch. Firstly, a comprehensive overview of the state-of-the-art HOAs applied for reactive power planning (RPP) and optimal reactive power dispatch (ORPD) is presented. It covers a number of HOA variants in the research field of RPP and ORPD problems, including genetic algorithm (GA), differential evolution (DE), particle swarm optimization (PSO), and evolutionary programming (EP), etc. A modified quantum-inspired differential evolutional algorithm (MQDE) with a novel reset strategy is developed for optimal RPP. The proposed MQDE is based on quantum mechanics combining with a competitive DE mutation scheme, i.e. DE/best/1/bin. It overcomes a major difficulty of DE techniques in ensuring the search diversity of the population when the algorithm is approaching the region of local optimum in the later stages of iteration process. A novel HOAs-adaptive range composite differential evolution (ARCoDE) algorithm is developed for ORPD that is one of the critical components in optimal power flow (OPF) study. Due to the nature of power dispatch, the ORPD problems need to be solved in a timely manner. This imposes a limitation on number of function evaluations. The proposed ARCoDE algorithm utilizes the concept of compositing different types of trial vector generation strategies, which makes possible a decent balance between the exploration and exploitation capabilities in the solution. In addition, a novel control parameter range adaptation mechanism is proposed to enable a highly efficient adaptive tuning of control parameters. These novelties support ARCoDE to deliver satisfactory solutions while fulfilling the stringent time requirements. Finally, an efficiency ranking-based evolutionary algorithm (EREA) is proposed aiming at directly obtaining the most efficient DMUs. A slacks-based measure (SBM) of efficiency and its super efficiency pattern are applied to yield a full ranking of relative efficiency of DMUs in each evolving generation, based on which the most efficient DMUs can be eventually found for the multi-objective formulation of ORPD problem. | Subjects: | Electric power distribution -- Mathematical models Electric power systems -- Management Mathematical optimization Hong Kong Polytechnic University -- Dissertations |
Pages: | xi, 94 pages : color illustrations |
Appears in Collections: | Thesis |
Access
View full-text via https://theses.lib.polyu.edu.hk/handle/200/11033
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.