Optimal Power Dispatch of Renewable and NonRenewable Generation through a Second-Order Conic Model

Abstract

This work presents an extension of a second-order conic programming model (SOCP) to handle the multi-objective optimal power dispatch problem considering the probabilistic nature of some parameters related to power demand and the renewable energy sources (RES) generation, such as wind speed and solar irradiation level. Three objective functions are considered in this study: 1) costs of RES and non-RES generation; 2) active power losses in the transmission system; and, 3) emission pollutant gases produced by fossil fuel-based generating units. The stochastic nature of power demands and RES are developed through a set of representative operational scenarios extracted from historical data and via a scenario reduction technique. The results obtained in the SOCP model are compared with a nonlinear programming (NLP) model to check the robustness and precision of SOCP model. To this, both models are implemented and processed to simulate the optimal flow for the IEEE 57- and 118-bus systems.