Sustainable energy transition in cities: A deep statistical prediction model for renewable energy sources management for low-carbon urban development

Abstract

This study proposes a novel approach to sustainable energy transition in urban environments, focusing on the prediction and management of renewable energy outputs for low-carbon urban development. We introduce a new intelligent prediction model based on Gated Recurrent Unit (GRU) Networks to forecast the output power of both tidal and biomass units. The GRU model, known for its capability in capturing sequential dependencies, is employed to enhance the accuracy of renewable energy predictions in the context of urban sustainability. To address the interpretability challenges inherent in complex deep learning models, we leverage LIME (Local Interpretable Model-agnostic Explanations). By integrating LIME into our GRU-based prediction model, we enhance transparency and understandability. This approach facilitates the identification of key factors influencing predictions, providing stakeholders and decision-makers with valuable insights for effective sustainable energy management. The proposed hybrid model is applied to the digital model of real tidal and biomass sites in China, allowing for a statistical examination of renewable integration and energy management strategies. The results offer a comprehensive understanding of the dynamics between renewable energy sources and urban development, paving the way for informed decision-making in the pursuit of low-carbon and resilient cities. This research contributes to the broader discourse on sustainable urban development by offering a robust statistical framework for renewable energy integration and management in urban environments.