Risk identification and influence analysis model for urban energy internet based on knowledge graph improved decision-making trial and evaluation laboratory

Abstract

The pressing issue of climate change has led to a growing concern for low-carbon energy transformation. Urban energy internet (UEI) provides a means to facilitate renewable energy consumption by leveraging modern energy grid, smart energy services, and cyber-physical systems. However, the implementation of UEI entails complex and unknown risks arising from energy supplies, cyber-attacks, and defective equipment. To address this challenge, we propose an improved decision-making trial and evaluation laboratory based on knowledge graph and complex network (KG-CN-DEMATEL). Specifically, we employ a knowledge graph with Gaussian embedding (KG2E) to vectorize text information related to the risks. The vectors are combined with experts’ scores, which evaluate the importance and cohesion of the complex network. We then analyze the significance and interrelations of the risks by computing their reason and central degree. Our results demonstrate that (1) the improved model offers higher accuracy and convenience than the conventional DEMATEL, (2) the 15 risks are ranked based on their importance, with R1 (Renewable energy’s uncertainty), R6 (Generation technology), and R2 (Cyber-physical degree) being the most significant, with central degrees of 2.913, 2.635, and 2.566, respectively, and (3) the risks can be classified into six causal and nine effect elements, with their interrelations analyzed.