Reliability evaluation based on modified latin hypercube sampling and minimum load-cutting method

As its sampling frequency is independent of the scale of the system, Monte-Carlo simulation is widely applied to reliability evaluation of generation and transmission system, which is a complex large system. However, it takes a relatively long period to obtain a higher accurate reliability index. This paper points out that, the combination of Importance sampling method and Discrete Latin Hypercube sampling method is able to reduce the variance of the sample space and uniformly sample it simultaneously, which improves the convergence of Monte-Carlo simulation. Besides, the combination of sensitivity analysis and linear programming in the process of overload correction is able to guarantee the solution optimality as well as improve the solving speed. Finally, the practicability of the proposed algorithm is verified with the case study in IEEE RTS79 and RTS96.

I. Introduction

The Reliability Evaluation of generation and transmission system is mainly divided into two kinds: the analytical method and the simulation method. Monte Carlo simulation is a typical simulation method, which determines the status of the power system element state with the sampling method and employs the statistical techniques to deduce reliability index [1]–[3]. Significantly, the probabilistic-model-based Monte Carlo simulation compensates for the shortage of N-1 test, and reflects the stochastic character of power system such as system behavior, user requirement and various faults on electrical equipment. Under this situation, the mathematical expectations and variances of the system's reliability evaluation are given, along with its probability distribution in all cases. In addition, the simulation method is capable to deal with huge numbers of uncertainties in the grid, as well as the random characteristic of loads and the control strategy in actual operation. Moreover, such method has the feature that sampling numbers is much smaller than that of state numbers, contributing it overwhelmingly suitable for the reliability evaluation of the continually developing large system, whose scale is also undergone a rapid expansion [4] [5]. In this sense, Monte Carlo simulation possesses attractive properties for power system reliability evaluation with the development and popularization of computer [6] [7]. Consequently, Monte Carlo simulation has been exceedingly applied to deal with completed large system that contains various factors such as correlative loads and operation control strategies.