A pseudo measurement modeling based forecasting aided state estimation framework for distribution network

Abstract

Modern large-scale active distribution networks have complex and dynamic operating circumstances, which pose major difficulties to state estimation (SE) technology. This study suggests a novel forecasting aided state estimate (FASE) framework based on an enhanced pseudo measurement modeling approach to overcome these issues and boost management and control decisions. Specifically, the suggested framework employs an advanced kind of pseudo measurement modeling that builds a model that is consistent with the distribution network’s real operation by using a support vector machine (SVM) with an upgraded kernel function. Furthermore, we introduce a numerical stability enhanced FASE algorithm that enhances the accuracy and efficiency of the estimation process. Through the application of measurement transformation and trustworthy pseudo measurement data as input, the FASE algorithm attains high-precision operational parameter awareness of the distribution network. Ultimately, the case study illustrates the benefits of the suggested framework over existing methods in terms of estimation accuracy, efficiency, and numerical stability compared to existing methods.