| Distributed generation(DG), as an aid for traditional power supply, which localconsumes electricity, generates power efficiently, and reduces environment pollution,is widely utilized and playing an important role in power sys tem. When DGs areinstalled in the distribution network, the structure and operation mode of distributionnetwork will change. Moreover, the power flow, losses and the voltage of the networkwill be affected. Distribution network reconfiguration is the process of altering thetopological structures of distribution feeders by changing the open/closed status of itsswitches, which can help reducing losses and improving the voltage quality and actsas an important mean for the operation of the system economically. So it is importantto study the optimal allocation of DGs and make use of the network reconfiguration torealize the economical operation of the distribution network with DGs.The problems of optimal allocation of DGs and reconfiguration of distributionnetwork with DGs are mainly researched in this paper.Based on the analysis of operation and control of several typical DGs, such aswind power, solar power, fuel cell and microturbine, different types of DGs areclassified into PQ, PI, PV and PQ(V) type nodes. Then the models of four type nodesin power flow calculation are presented. A back/forward sweep distribution load flowalgorithm with DGs based on the node-layer is realized. The stability of the algorithmand the capacity for calculating power flow of the network with different typ es ofDGs are also simulated and analyzed.The optimal allocation model of a double objective including active power lossand static voltage stability is presented. Then quantum particle swarm optimization isadopted to solve this model. Considering the installing location and capacity limit ofDGs, the area optimization idea is proposed, then the area optimal allocationconsidering multiple types of DGs is realized and simulations are carried onIEEE33-node system. In addition, two problems, one is the hybrid area optimalallocation with different types of DGs considering the number limited, the other is theinjection capacity optimization of DGs under the condition of constant position aresimulated on the PG&E69-node system. To validate the effectiveness ofdual-objective optimization, the optimization results of the single objects and doubleobjects are compared. Through these simulations, the validity and effectiveness of the dual-objective area optimization method in this paper are proved.A model of distribution network reconfiguration is applied in the network withDGs. In the objective function, minimization of the real power losses is taken intoaccount. According to the radial structure of distribution network, the particles areadopted integer loop encoding. The interger coded quantum particle swarmoptimization has been applied to solve feeder reconfiguration of DGs, which improvesthe convergence speed and efficiency of optimization. For eliminating the infeasiblesolutions generated during the optimization, switch loop-node stratified infeasiblesolution judgment method is proposed in this paper. This judgment method adopts thenode-layer thought of the back/forward sweep distribution load flow algorithm basedon the node-layer, and it can completely eliminate all infeasible solutions in thereconfiguration. Several simulations related to network reconfigurations with orwithout DGs are tested on IEEE33-node and PG&E69-node systems. Moreover, theimpact of the rated voltage of PV node DGs on network reconfiguration is analyzed.To achieve the overall optimal performance of the distribution network, a jointoptimization of distribution network reconfiguration and injected power of DG isadopted for the dispatchable DGs. Through this process, the active loss of distributionnetwork with DGs will be further reduced. Moreover, the voltage level of the networkwill be further improved. |
PDF Full Text Request