A second order cone based relaxation and decomposition algorithm for multi-period reactive power optimization considering uncertain PV integration in active distribution networks

To address daily operating times of discrete reactive power compensators, multi-period reactive power optimization (MPRPO) is set up to minimize energy loss over multiple periods. This paper proposes a second-order cone relaxation and decomposition (SOCRD) algorithm for solving MPRPO problem considering uncertain PV integration in active distribution networks to improve computational performance. To deal with uncertainty output of PV generation, an extended stochastic programming is employed in this method, coordinating with continuous and discrete reactive power compensators. Solution quality and computation performance are compared with deterministic model as well as traditional algorithms on IEEE-33 bus system, which illustrates efficiency and effectiveness of the proposed method.