Load Models Considering Dependence And Their Applications In Reliability Evaluation Of Power System

Under the current social background conditions, with the increasing expansion of power system and the deepening of system interconnection degree, more and more attention have been paid to the reliability of power system. Accordingly, reliability evaluation technology of power system has being increasingly applied to provide quantitative reference in grid projecting and stable operation of power system.Load is one of the most important modules in power system. The commonly used load models are probabilistic model and time series model. The reference value of power system reliability evaluation is influenced by the veracity of load model. In a large scale power system network, there will not be only one load node. Due to the geographical position, load property and other factors, load data change between each load node has a certain correlation and has a certain influence on the reliability of power system. Therefore, in power system reliability evaluation, this correlation between each load node should be considered in the simulation of load sample data.In this paper, from the perspective of load spatial correlation and load temporal correlation, copula function method which is commonly used in correlation analysis has been used to study the effects of load correlation to the reliability of power system. Considering the bus load spatial correlation, the creation process of bus load joint probability distribution can be decomposed into two parts:The marginal distribution structure of single load sequence, and related structure identification between each load. Load marginal distribution is structured by the nonparametric kernel density estimation method, and related structure between each node is adopted by Copula function. Considering the bus load temporal correlation, the Copula function is applied to the construction process of continues Markov chain model, a CSMC (Continuous state Markov chain) model is obtained based on Copula function. Finally, a multi load CSMC model based on Copula function is obtained by the combination of the load spatial correlation model and the load temporal model. In this paper, according to the difference of bus load model, non sequential Monte Carlo method and sequential Monte Carlo method is adopted to calculate the corresponding system reliability indices. A case study of a revised IEEE-RBTS was used to explain the influence of bus load correlation on the reliability indices, and to validate the modeling method proposed.