Abstract The financial stability of new energy vehicle (NEV) firms has become a central concern amid rapid digitalization and the transition toward low-carbon economies. Traditional financial risk models assume fixed structures and linear relationships, which makes them ineffective for handling the complex, time-varying, and policy-sensitive nature of energy firms. To address this challenge, a deep learning–based early warning framework is developed using the Modular Energy Accountability Lattice (MEAL) and the Semantic Policy Fusion Strategy (SPFS). These components jointly encode financial flows and policy dynamics to predict systemic risk. Experiments on real-world NEV firm data show that the proposed model achieves higher precision, recall, and interpretability than conventional methods, offering a scalable and policy-aware tool for sustainable financial monitoring.
Zou et al. (Fri,) studied this question.