Voltage Instability Risk Forecasting under Load Demand Probabilistic Model

This paper proposes a methodology for a daily projection (“Day Ahead Operational Forecasting”) of the voltage instability operative risk of a power system considering the probabilistic behaviour of the load. The method computes hour to hour the probability distribution function (fdp) of the modal behaviour of the Reduced Jacobian Matrix (JR) and based on this fdp computes the probability (risk) that the system undergoes a voltage instability. Each fdp settles down by means of statistical analyses of the critical eigenvalue of the JR behaviour, using different load conditions around the scheduled base case of operation by means of Monte Carlo simulation. The proposed methodology was evaluated in the 39 nodes New England system, showing the dependency of the voltage instability risk respect to the load demand level and the planned generation. Of another part, the behaviour of the risk based on the correlation crossed between the loads is evaluated.