A MIP Model for Risk Constrained Switch Placement in Distribution Networks

Abstract

The great share of interruptions in distribution networks motivates distribution decision makers to establish various reliability enhancement strategies. Amongst these strategies, deploying remote controlled switch (RCS) can make a crucial contribution to the reduction of interruption costs. Nevertheless, the stochastic nature of contingencies affects RCS worth and imposes substantial financial risk to RCS deployment projects. This paper proposes a mathematical model to consider the risk in the optimal RCS deployment problem. The model determines the number and location of RCSs such that the expected profit is maximized while financial risk is minimized. The risk is modeled through conditional value-at-risk (CVaR) as one of the most applied risk indices. The risk preference of distribution companies is considered via integrating the weighted risk index and the expected profit in the objective function. The model is formulated in the format of mixed integer programming (MIP) which enables solving the problem in an effective runtime. The model is applied to the RBTS-Bus4 and a real distribution network. Various studies and sensitivity analyses reveal the importance of considering financial risk in the problem and the performance of the proposed model.