Available Transfer Capability Based On Improved Chaos Cloud Particle Swarm Optimization Algorithm

With the expansion of power system in China, electric power market is developingrapidly, the available transfer capability(ATC) in interconnected power systems becomes animportant measurement to assess the security and reliability of system. Its size reflects thesystem interconnection strength and structure. The available transfer capability not only beseen as a market signal, which is used to constraint commercial behavior of marketparticipants. It also can guide system dispatcher to schedule, ensure the safe and stableoperation of the system. Therefore, how to reasonably and efficiently calculate the availabletransfer capability between regional power grid has important research significance.On the basis of the existing research results about available transmission capability, anoptimal power flow model is established for available transfer capability under the staticsecurity constraints. The active power maximum of all load nodes in receiving area is taken asobjective function, the fast decoupled method is applied for power flow calculation. In thispaper, the particle swarm optimization algorithm, chaos algorithm and cloud model algorithmare researched deeply and a new kind of intelligent optimization algorithm is put forward.In order to balance the local and global search ability of particle swarm optimization,firstly, the golden section grouping criteria is introduced to divide particle swarm intostandard particles, chaotic particles, cloud particles three sub swarms accord to fitness value.Each population has different processing operations and updating modes.Secondly, for thestandard particles, in order to solve the premature convergence in particle swarm optimization,the dissipation operation is added in the velocity update. For the chaotic particles, in order toimprove the search efficiency, fine search in a certain range by random search in theneighborhood of the current optimal solution after each step of chaotic search. For the cloudparticles, X condition cloud generator is used to adapt inertia weight dynamiclly.Finally, the chaotic cloud particle swarm algorithm based on the golden section criteriawas simulated in IEEE-30and IEEE-39node system to caculate available transfer capabilityvalue. Then compared results with the cloud-PSO and chaos-PSO algorithm.The simulationresults show that the algorithm is obviously improved in global search ability and precision ofoptimization. It is more suitable for processing such large scale nonlinear optimizationproblems.