Outlier Reconstruction Based Distribution System State Estimation Using Equivalent Model of Long Short-term Memory and Metropolis-Hastings Sampling

The accuracy of distribution system state estimation (DDSE) is reduced when phasor measurement unit (PMU) measurements contain outliers because cyber attacks or global positioning spoofing attacks. Therefore, to enhance the robustness DDSE outliers, approximate target Metropolis-Hastings (MH) sampling, and judge prediction long short-term memory (LSTM) network, this paper proposes an outlier reconstruction based method using equivalent model LSTM network MH sampling (E-LM model), motivated by characteristics chronological correlations PMU measurements. First, derived a kernel density function. Subsequently, reasons advantages E-LM are explained analyzed from mathematical point view. proposed LSTM-based can decrease number futile iterations. Moreover, MH-based forecasting each prediction, which independent its true value. Finally, simulations conducted evaluate performance integrating into DDSE.