Research on Photovoltaic Power Generation Based on Multi-dimensional Indicators and Models

Photovoltaic (PV) power generation plays an important role in sustainable energy transition and carbon-emission reduction. However, existing studies often focus on only one aspect of PV analysis—such as power forecasting, deployment optimization, or environmental impact—making it difficult to support integrated planning and decision-making. To address this gap, this study develops a comprehensive framework for PV power generation analysis that combines power supply forecasting, optimization, and carbon-emission impact assessment. The methodology includes the construction of a multidimensional indicator system, dimensionality reduction using principal component analysis (PCA) and t-SNE, and particle swarm optimization (PSO) for model optimization. The results indicate that a 1% increase in PV power generation could reduce total carbon emissions in China’s power sector by approximately 2.05% by 2035. The power supply prediction model achieves a high goodness-of-fit (R² = 0.9975), and the optimization module provides a long-term scenario estimate of PV generation for 2024–2060, with total generation reaching 452,838,023.71 kWh by 2060. Overall, the proposed framework provides data-driven decision support for regional energy planning and low-carbon development. Nevertheless, its long-term applicability may be affected by policy shifts and market dynamics, which should be incorporated in future improvements.