A New Improved Group Search Optimizer Algorithm Based On Quantum

Optimization problem is a kind of common problems in the life aclouding power system,Swarm Intelligence Algorithm (SI) is an emerging evolutionary Algorithm in recent years. SIhas been proved that it has high robustness of the system, good system expansion and fastcalculation speed, so it has been widely applied in solving optimization problem. GroupSearch Optimizer (GSO）is a nem swarm intelligence algorithm which was proposed in recentyears, although GSO has shown its preponderance on solving large scale optimizationproblem with multi-dimenson and multi constrains, however, since the GSO is presented in ashort time, the application of GSO in power system and other field is still in the beginningstages so far. So developing the study of GSO and its improved algorithm, applying thesealgorithms to the optimization problem in power system, putting forward a new way to solvethe optimization problem of the in power system have important significance on enhancingthe stability and economy of power system operation.Firstly, the paper has introduced the basic knowledge of optimization problem、quantumtheory and swarm intelligence algorithm, making a summray about reactive poweroptimization and unit commitment problem in power system. Then on the basis of combiningGSO and quantum-inspired evolutionary algorithm (QEA), the paper presents a new GroupSearch Optimizer based on quantum theory (QGSO) and takes a detail study on the searchstrategy of each kind of group member. A complete algorithm frame and calculation flow arealso been put forward in the paper. The QGSO is progammed in matlab to disscuss itsapplication on power system.Then the paper selects the active loss minimum model as the objective function andgives the control variables and state variables constraints to apply QGSO on reactive poweroptimization problem of IEEE14node system and30node system respectively. Analysisresults show that system status has been improved and active power loss is obvious decreasedafter the optimization of QGSO. The results comparision between QGSO and other algorithmshow that the performance of QGSO is superior to other algorithms.Finally QGSO is applied to unit commitment problem in power system. The paperproposes a improved heuristic group initialization strategy to get the initial feasible solution fastly. A special handling strategy is also adopted in process of calculation. The simulationresults on unit commitment problem of10-100units system show that the results of QGSOare not the best in small scale system but QGSO has an obvious advantage in large scale unitcommitment problem. The results have a certain reference significance on solving the the unitcommitment problem of large-scale power grid in practical power system.