Solar heating as sustainable solution in dyeing industries: A compromise solution approach integrated machine learning

Dyeing mills are among the most energy-intensive industries, and their integration with solar technology can be a step toward net-zero carbon goals. This study conducts a comprehensive assessment of such integration. Initially, integration of solar collectors in the dyeing industry was examined from six perspectives: technical, economic, environmental, socio-political, and risk. For evaluation, a novel hybrid approach combining SWARA and COCOSO in Fuzzy uncertainty environment was employed. Furthermore, an integrated layout for the dyeing industry was proposed and modeled using the open-source SAM model. The layout designed based on multi-objective optimization (maximizing solar fraction and minimizing cost) driven by machine learning. The multi-criteria assessment indicated logistical costs, net present cost, and thermal energy potential had the highest effect on the development of solar water heaters. Additionally, among the predictive algorithms, the Random Forest (RF) demonstrated the highest accuracy, with an R2 value exceeding 99 %. The results also showed the optimal layout resulted in an energy savings of 218 MWh, accompanied by a reduction of 23,500 m3 natural gas and an annual decrease of 48 tons-CO2 emissions. The findings of this study can serve as a roadmap for developing countries, particularly those rich in fossil fuels, which suffer from air pollution problems.