Multi-objective optimization for customized solar business models considering technical-economic-environmental performance: A NSGA-II integrated TOPSIS method

To address the urgent need for a transformative shift to the post-feed-in tariff (FiT) era, this study focuses on the development of customized solar business models in Hong Kong, which intends to integrate existing policies (i.e., FiT and renewable energy certificates (REC)) with alternative mechanisms (i.e., net metering (NM) and power purchase agreements (PPA)), and simulate stakeholder interactions considering property owners, utility companies, and third-party developers. This study employs a multi-objective optimization approach using a Non-dominated Sorting Algorithm II (NSGA-II) integrated with the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method to maximize net present value (NPV), while minimizing electricity bills and carbon emissions. Key findings indicate that under the NM-FiT model, higher NPV leads to reduction in electricity bills and carbon emissions, while the PPA-FiT and PPA-REC models show a trade-off relationship between NPV and the other two optimization objectives. Furthermore, NPV changes significantly with fluctuations in the PPA rate and PPA proportion under the PPA-FiT and PPA-REC models, and is sensitive to changes in installed capacity under the NM-FiT model. These findings will enhance decision-making strategies and promoting sustainable energy goals in Hong Kong, which ultimately contributes to a more resilient and adaptive energy framework.