SCS6: Energy Transition and Environmental Regulation

Accurate forecasting and multi-energy collaborative optimization are critical for the secure and economic operation of integrated energy systems (IES). However, traditional models exhibit significant limitations in unstructured multi-source data fusion, few-shot scenario generalization, and interpretability. To address these issues, this paper proposes a collaborative optimization framework integrating large language models (LLMs) with multi-time-scale intelligent algorithms. Initially, a forecasting architecture based on multi-source data fusion and knowledge graphs is constructed. It utilizes the DEMATEL - AISM model to mine factors affecting electricity consumption, introducing time-delay and coupling effects to transform non-stationary time series into Stacking ensemble and deep neural network models. This enables precise predictive inversion of user loads and spot electricity prices. Subsequently, through AI-based case reasoning and fuzzy C-means clustering, three-level typical planning schemes encompassing equipment selection and capacity configuration are extracted. These schemes are integrated with a three-stage multi-time-scale operational mechanism (day-ahead, cyclic, and monitoring). By leveraging heuristic algorithms like Genetic Algorithm and Particle Swarm Optimization, the framework generates equipment output strategies that optimally balance economic costs, carbon emissions, and comprehensive energy efficiency.Simulation tests across various working conditions demonstrate that the proposed scenario-mapping algorithms significantly improve iterative convergence accuracy and speed compared to traditional single-solver optimizations. Simultaneously, the multi-stage scheduling results ensure strict compliance with physical operational constraints. Relying on the Dify workflow engine, this method achieves an intelligent closed-loop process from natural language semantic parsing to multi-objective decision-making, significantly enhancing the system's adaptive capacity under extreme weather conditions and uncertain disturbance scenarios.

Wind and solar energy will become crucial technologies for achieving deep decarbonization of the energy system. However, as typical "climate-sensitive" power resources, ongoing climate change may significantly alter the resource endowment conditions of wind and solar energy, thereby affecting their power generation potential and economic viability. Therefore, this study takes China as a case to quantitatively assess the extent to which wind and solar resources are influenced by different climate change scenarios from both resource potential and cost perspectives. This work provide a novel analytical perspective and methodological tools for long-term expansion pathway planning and site selection decisions in wind and solar power technologies.

Coal mining regions have long supported industrial development in China and other countries. However, under climate change pressures, energy transition in these areas face serious challenges, especially regarding the reemployment of coal miners. Achieving just transition has become a key concern. This study focuses on the impact of photovoltaic (PV) deployment on coal miner reemployment in coal mining regions across major coal-producing provinces of China, including Inner Mongolia, Shanxi, Shaanxi, and Ningxia. Results show that coal miner unemployment may peak in 2030 in these provinces, with 290,000–570,000 coal miners affected. Shanxi will be the most severely impacted, with 240,000–420,000 job losses. Deploying PV in coal mining regions can effectively ease reemployment pressures. The effectiveness of this approach depends on the migration distance between coal miners and PV-related jobs: within 10 km, about 15,232–18,232 coal miners can be reemployed; at 200 km and 500 km distance thresholds, reemployment reaches 25% and 50% of affected coal miners, respectively. Moreover, considerable regional disparities exist in coal miner migration flows, with most coal miners relocating from Shanxi, Shaanxi, and Ningxia to Inner Mongolia, forming 11 major migration corridors such as the Yulin–Ordos route, the Shuozhou–Baotou route and the Datong–Xilingol route. Finally, this study offers relevant policy recommendations.

The energy transition is fundamentally reshaping how power systems organize capacity and ensure reliability, bringing cross-regional capacity sharing to the centre of adequacy governance. In power systems with high renewable penetration, wind and solar variability make it increasingly difficult for any single region to rely solely on local resources to provide firm capacity credit. This tension intensifies under extreme weather conditions, when renewable shortfalls may coincide with demand surges, placing stress on adequacy models that depend primarily on cross-regional resource allocation. At the same time, temporal staggering in peak demand and spatial heterogeneity in renewable endowments create structural complementarities across regions. Cross-regional sharing therefore emerges not only as a cooperative arrangement but as a structural requirement for sustaining adequacy in zero-carbon systems. Reflecting this shift, jurisdictions worldwide are advancing new cooperative adequacy mechanisms. In the western United States, the Western Resource Adequacy Program coordinates multi-region obligations. In China, a unified national electricity market and expanding interprovincial transmission infrastructure support increasingly integrated capacity sharing.
Yet theoretical complementarity does not guarantee operational reliability. Climate-driven shocks can induce strong spatial and temporal correlations in both electricity demand and renewable generation, eroding the diversity benefits that cross-regional cooperation assumes. During Winter Storm Uri in February 2021, multiple interconnected North American regions experienced simultaneous demand spikes and widespread generation outages, sharply constraining cross-regional support capability. Similarly, the 2022 extreme heat and drought event in Sichuan reduced hydropower output while increasing electricity demand, simultaneously weakening regional self-adequacy and export capability. These events underscore an unresolved question: under highly correlated extreme conditions, can cross-regional cooperation reliably translate structural complementarity into assured capacity adequacy? Systematically assessing the reliability boundaries of cross-regional sharing under different system states and climate scenarios therefore becomes essential.