Optimizing solar photovoltaic farm-based cogeneration systems with artificial intelligence (AI) and Cascade compressed air energy storage for stable power generation and peak shaving: A Japan-focused case study

This study proposes a novel solar cogeneration system that integrates compressed air energy storage units (CAES) and gas turbines (GT) with a solar farm consisting of photovoltaic panels. The primary objective of this research is to address the instability of solar energy production and help during peak energy consumption by utilizing CAES. The proposed system is modeled using EES software, and its performance is optimized using advanced artificial intelligence (AI) methods, including artificial neural networks and intelligent algorithms. The analysis identifies five critical decision variables that significantly impact system performance: the number of photovoltaic panels, CAES pressure ratio, CAES inlet pressure, gas turbine efficiency, and compressor efficiency. The results demonstrate that the optimized solar cogeneration system can achieve exergy efficiency of 36.44 % and a cost rate of 13.76 $/hour. The exergy analysis of the system indicates that the most significant destruction is related to the solar farm, gas turbine, and compressors. Furthermore, this study investigates the effect of weather in eight Japanese cities on system performance, considering two operating modes: with and without using system electricity (PV mode). The results show that the proposed solar cogeneration system has significant potential for clean electricity generation and CAES applications to overcome the instability of the solar system and help during peak energy consumption throughout the year. The study's findings highlight the attractive potential of integrating CAES and AI technologies in solar photovoltaic systems for stable and efficient power generation in Japan.