Research On Power System Reliability Evaluation Based On Monte Carlo Simulation

With the increasing size of the power system, the structure becomes more complex and the degree of automation is higher. The power system is facing new challenges of system security and reliability. It is an unfailing research topic in the field of power system to realize the reliability assessment of complex power system fast and accurately.Firstly, the common methods of power system reliability assessment are introduced. The time state sequence of elements in system is not considered in non-sequential Monte Carlo simulation method, which greatly reduces the simulation time, so that the method is widely used in large power system reliability assessment. However, the traditional non-sequential Monte Carlo simulation method is not sensitive to small probability events, which make the sampling speed in high reliability system greatly reduced. Under the condition of keeping a certain computational accuracy, sampling efficiency is inversely proportional to the sample variance in non-sequential Monte Carlo simulation method. Therefore, the technique of variance reduction is mainly researched in this article to improve the sampling efficiency of non-sequential Monte Carlo simulation method, such as importance sampling, Latin hypercube sampling, antithetic variable sampling and cross entropy importance sampling.Secondly, combined Latin hypercube sampling with importance sampling method, an improved Latin hypercube sampling method for power system reliability evaluation was proposed. The probabilistic distribution of current sample space of the system is changed by importance sampling, and then samples follow new distribution function are generated by Latin Hypercube sampling. Thus the sampling variance is reduced and massive repeated samples of normal system state are avoided.Thirdly, by combining antithetic variable sampling method with cross entropy importance sampling method, an improved sampling method, which suitable for power system reliability evaluation, is therefore proposed. An approximate function with zero variance probability, whose optimal parameters of component are obtained using cross entropy importance sampling density function of component, is constructed first. Then the variances of reliability indexes are further reduced through antithetic variable sampling according to the obtained optimal parameters.Finally, the two are proposed methods respectively applied to reliability evaluation of IEEE-RTS and modified IEEE-RTS test system. Comparison results show that the sampling efficiency is improved by the proposed method, compared with the other methods. The assessment results in high reliability system show that the improved methods are not only suitable for the general power system, but also suitable for the high reliability system. And its advantage is more obvious in high reliability system.