Reactive Power Optimization Based On Improved Quantum-Inspired Genetic Algorithm

Reactive power optimization is not only an effective method to ensure the secure and economic operation of power system, but also an important measure to reduce network losses and improve voltage quality. With the development of power systems and the strengthening of networking, reactive power optimization is becoming more and more important. In essence, reactive power optimization problem is a multi-variable, multi-constrained, mixed non-linear combinatorial optimization scheduling problem with many local extrema and complex optimization process. The optimization algorithm plays a significant role in the quality and efficiency of optimization process for reactive power optimization control system. Consequently, reactive power optimization algorithms are researched to obtain good optimization performance, and reduce network losses and improve voltage quality. The main work and research fruits are as follows.1. Recent researches and developments of reactive power optimizations are introduced. And then a mathematic model of reactive power optimization, regarding the minimization of active power losses as objective function, is established.2. Basic genetic algorithms (GA) and its application to reactive power optimization are introduced. Also, quantum-inspired genetic algorithms (QGA) and its flowchart of reactive power optimization are described in detail. Experiments carried on IEEE30-bus system using GA and QGA show that the active power loss of QGA is lower than that of GA.3. To overcome the shortcoming of QGA's liability to fall into the local extreme value in solving reactive power optimization problem, a reactive power optimization method based on improved catastrophic quantum-inspired genetic algorithm (ICQGA) is proposed. Experiments carried on IEEE-6 bus system and IEEE-30 bus system show that ICQGA can drag the search out of local minima by introducing the strategies of catastrophe into QGA. The results show that ICQGA can get lower active power loss than QGA.4. To enhance the local search capability of quantum-inspired genetic algorithm (QGA), a quantum-inspired genetic algorithm with local search (LSQGA) is presented. This technique uses two layers quantum-inspired genetic algorithm search optimal solution. Outer layer quantum-inspired genetic algorithm is used for searching global solution. In the process of searching global solution, if the searched best solution is not improved in certain successive iterations, local search by using inner layer quantum-inspired genetic algorithm is applied to explore the neighboring domain of the solution. Experiments carried out on complex functions and IEEE30-bus system show that the performance of LSQGA is improved compared with QGA and ICQGA, in terms of search capability and convergence.5. After researching on LSQGA, a novel memetic algorithm based on real-observation quantum-inspired genetic algorithm (MArQ) is proposed. MArQ is a hybrid algorithm combining rQGA (real-observation quantum-inspired genetic algorithm) with local search techniques. In this algorithm, rQGA is used to explore the whole solution space and tabu search is embedded in rQGA as an evolutionary operator to explore the neighboring areas of solutions. In the process of tabu search, one of the optimization variables is changed in the neighborhood of the searched best solution of rQGA and then several new solutions are generated. Also, the neighborhood radius decreases as the evolutionary generation increases. In MArQ, chaos method is applied to generate the initial population with good distribution in whole search space. Several bench complex functions and an application example of reactive power optimization in power systems were applied to test the MArQ performances, results show that MArQ has got lowest active power loss compared with ICQGA and LSQGA, which offers a new thought for solving reactive power optimization problem.This paper was supported by the National Natural Science Foundation of China (Grand No. 60702026) and the Open Foundation of Engineering Research Center of Transportation Safety of the Ministry of Education of China (2008).