Combining multi-objective evolutionary algorithm and HΔ matrix for SCADA and PMU placement

Metering Systems Planning (MSP) for state estimation considering the investment cost is a multi-objective, combinatorial optimization problem that may require the investigation of many possible solutions. To solve this problem it is necessary to take into account the conflicting requirements of investment cost and the performance requirements (system observability during normal operating conditions and in the presence of contingency situations causing the loss of one or two measurements or even the loss of one RTU). This paper proposes a multi-objective formulation for the MSP problem and also a methodology to solve it. The methodology combines a Multi-Objective Evolutionary Algorithm based on subpopulation tables with the properties of the so-called H
_Δ
matrix and can be used to planning hybrid (with SCADA and PMU measurements) or conventional (only with SCADA measurements) metering systems. Simulations results with IEEE 14, 30, 57 and 118-bus test systems and with a real Brazilian system illustrate the efficiency of the proposed methodology.