Power system reliability assessment considering multiple dependent variables based on non-parametric disaggregation model

• 1、School of Electrical Enineering,Chongqing University,Chongqing 400044

Abstract：The correlated random variation of the bus loads, wind powers, etc., has posed a challenge to the power system operation security. How to construct the accurate dependence model becomes vital for power system reliability evaluation. Most currently used methods in estimating the multivariate joint probability density function (PDF) ignore the aggregation constraint ，To incorporate the aggregation constraint, a non-parametric disaggregation (NPD) model (i.e. multivariate conditional PDF by non-parametric estimation) which can randomly disaggregating an aggregate variable (e.g. system load) into the non-aggregate variables (e.g. bus loads) is introduced. Moreover, an effective three-stage conditional sampling method is also presented for the proposed non-parametric disaggregation model (NPD). The dependence can be modeled flexibly and accurately by the NPD model, and its validity is verified by the reliability evaluation of the modified RBTS and IEEE-RTS79.?????

Keywords： reliability assessment non-parametric disaggregation ；kernel density estimation dependence