A Distributed Charging Coordination Of Plug-in Electric Vehicles Considering Feeder Overload Constraint

In order to decrease the consumption of fuel and protect our environment, more and more people choose to drive Plug-in Electric Vehicles(PEVs). However, the popularity of PEVs also brings us some problems. Because of the insufficient power capacity of the distribution feeders, excessive load may break down the whole system. For that reason, it is crucial for us to limit the amount of power through each distribution feeder. Hence, it is of meaning to design an effective charging control algorithm to let the PEVs adapt to the existing distribution network.In this paper, firstly, we built the charging model for PEVs. We use the tree-structured distribution network with the distribution substation located at the root node of the tree. Each feeder has their own overload constraint. In the centralized charging model, the system cost consists of the total charging price cost and total battery degradation cost. In the distributed charging model, the individual cost consists of the individual charging price cost and individual battery degradation cost.Secondly, we design a centralized charging algorithm for PEVs cosidering feeder overload constraint, and simulations are presented to illustrate the performance of the proposed algorithms. After that, we design a distributed charging algorithm for PEVs cosidering feeder overload constraint. The algorithm is a serial iterative algorithm, each PEV update their own charging strategy to minimize their cost function without overloading the feeders according to the latest electric price and feeder overload information. After some step of iteration, the algorithm can obtain a local optimal charging control strategy.Finally, we use chart to compare the local optimal charging control strategy and the globel optimal charging control strategy, and draw the conclution that the local optimal charging control strategy and the globel optimal charging control strategy are nearly the same. Furthermore, we compared our distributed charging algorithm with other algorithm which consider feeder overload constraint as well, and showed the advantage of our algorithm in the number of iterations and other aspects.