Selection for Flexibility Areas Using Probabilistic Machine Learning Under Measurement Uncertainty

Abstract

Coordination between transmission system operators (TSOs) and distribution system operators (DSOs) can support TSOs inusing the distribution system (DS) flexibility while ensuring feasible operation. Flexibility areas (FAs) can support TSO-DSOcoordination, aggregating the total feasible flexibility within the DS. However, existing real-time estimation approaches do notconsider the limited measurements within DS. This paper proposes a Bayesian neural network (BNN) to estimate the operatingconditions that bound the operational flexibility, including epistemic and aleatoric uncertainties. These uncertainties stem fromthe limited real-time measurements in DSs and the measurement noise. TSOs can select a threshold that confirms a probabilityof safety, considering uncertainty margins. The paper also provides FA estimation in DS topologies with two points of commoncoupling (PCC) with the transmission system. Case studies in the CIGRE and Oberrhein networks compare the proposed BNNsto baseline statistic-based approaches for forecast and measurement uncertainty in FAs. The case studies show the proposed FAestimation under various safety margins and systems with 2-PCC. Case studies also assess various measurement noise levels andevaluate model performance for different DS topologies