Data Mining in Nonlinear Probabilistic Load Flow Based on Monte Carlo Simulation

This paper presents a technique framework to evaluate the nonlinear a.c. probabilistic load flow using Monte Carlo simulation with data mining off-line. It provides the whole process of Monte Carlo simulation with data mining technique from which the probability-density curves of the injected reactive powers, voltages, angles, active and reactive power flows can be made. The indices to evaluate the severity of risk for power system static security assessment are presented. To speed up the algorithm, parallel computation is suggested according to the scale of the bulk power system. The IEEE 14-bus test system has been taken for case study. The case has shown that the method is effective and efficient.