HEURISTIC JUSTIFICATION AND DIFFERENTIAL EVOLUTION-BASED FINAL SELF-RESTORATION STATE OPTIMIZATION FOR URBAN POWER GRID AFTER BLACKOUT

Urban power grid has to be self-restored with a lack of power sources after blackout while waiting for external power supply. An optimal target system of self-restoration can instruct engineers or expert system to make an approximately optimal restoration plan. In this paper, a problem named final self-restoration state optimization (FSRSO) is put forward, and a simplified mathematical model of FSRSO is presented with the skeleton of power grid that can be given by engineers. A heuristic justification (HJ) is proposed to be integrated with differential evolution (DE) algorithm to solve the problem. HJ adjusts the active power output of the slack bus to be within limits, decreases constraints violations with great probability and therefore speeds up the evolution procedure of DE. Losses ratios of individuals of the last generation are utilized by HJ to evaluate power losses. According to tests results on IEEE 39-bus system, DE integrated with HJ gets better solution in much less iterations than the classical DE.