Stochastic Approach for Evaluating the Operation of Electric Power Distribution Systems

This paper introduces a stochastic approach to evaluate the performance of electric power distribution systems (EPDS). The daily load and generation curves at each node are generated in a correlated and random manner for each day of the week. These curves, spanning seven days, constitute a sample within the Monte Carlo Simulation (MCS) and form a synthetic representation of a week. This data from the synthetic week is then utilized to calculate power flows, yielding the metrics of interest for each day: energy losses and nodal voltage violations. These daily metrics are aggregated over a year, considering the distribution of days within that year. Once the MCS converges, statistical values for annual voltage violations and losses are derived. The methodology was applied to a 23-node distribution system with capacitor banks, voltage regulators, and distributed energy resources. The findings indicate that employing the MCS while considering the correlation between nodal power demands and generation offers a practical approach to studying EPDS operation. Disregarding this correlation results in underestimating operational indicators within the grid.