Ambient Signals Based Load Modeling with Combined Gradient-based Optimization and Regression Method.

Xinran Zhang,David J. Hill,Chao Lu,Yue Song

2020-01-01

Abstract:Load modeling is an important issue in modeling a power system. The approachof ambient signals-based load modeling (ASLM) was recently proposed to bettertrack the time-varying changes of load models. To improve computationefficiency and model structure complexity, a hierarchical framework for ASLM isproposed in this paper. Through this framework, the hidden quasi-convexity ofload modeling problem is explored for the first time, and more complicatedstatic load model structures can be applied. In the upper stage, theidentification of dynamic load parameters is regarded as an optimizationproblem. In the lower stage, the optimal static load parameters are obtainedthrough linear regression for a given group of dynamic load parameters.Afterwards, the regression residuals are regarded as the objective function(OF) of the upper stage optimization problem. The proposed method is validatedby the case study results in Guangdong Power Grid. The results have shown thatthe OF is mostly quasi-convex after the transformation of induction motormodel, which provides the basis for the application of gradient-basedoptimization algorithm. The case study results also validate that the proposedapproach has better computation efficiency and model structure complexitycompared with the previous ASLM approach.