Input interaction and output fusion of adaptive interacting multiple load modeling

Abstract

Accurate load modeling with fluctuating loads has been one of the most challenging problems in power system. Input interaction and output fusion of the adaptive interacting multiple load modeling (AIMLM) method is introduced in this paper. Different load models switch through a Markov probability transfer matrix and filter in parallel through the extended Kalman filters (EKF). After the state estimation, model probabilities are updated according to the evaluated errors. Finally, load models are fused and output based on the probabilities, thus generating a synthetic load model. Simulation results show that interaction and output fusion of the AIMLM method are of high accuracy and can effectively solve the time-varying problem of load characteristics.