Voltage and reactive power control using approximate stochastic annealing

One of the objectives of smart grids is to optimize the performance of power devices to improve energy efficiency by utilizing additional data from smart meters. In particular, this is important for the transmission and distribution network, in which approximately 7% of the total energy generated is wasted. Efficient management of voltage profiles and reactive power in power distribution systems plays an important role towards this goal. In this paper, we consider voltage and reactive power control (VVC) problem with the objective to determine the proper settings of capacitor banks and transformer taps in power distribution systems to minimize daily energy losses. Voltage constraints and operation limits constraints on transformer load tap changers (LTCs) and shunt capacitors (SCs) are considered in our model. We propose a stochastic search algorithm called Approximate Stochastic Annealing (ASA) for solving this VVC problem. The algorithm searches the optimal control schedule by randomly sampling from a sequence of probability distributions over the space of all possible settings of LTCs and SCs. A Lagrangian RelaxationDynamic Programming (LR-DP) algorithm is also proposed to obtain upper and lower bounds on the performance of the optimal solution. Our testing results on the well-known PG&E 69-bus distribution network indicate that the ASA algorithm may yield solutions very close to optimum within a modest amount of computational time.