Determination of the solar fraction for optimized parabolic concentrator solar collector networks

Studies show that the solar fraction of most solar thermal plants that supply heat to industrial processes ranges from 5-60%. The main limiting factor for increasing this value to 100% is the installation area. This work presents a study of the solar fraction parameter in optimized designs of systems using parabolic trough collector technology. An optimization process using Mixed-Integer Nonlinear Programming with 9 decision variables is proposed, employing the heuristic technique of Particle Swarm Optimization coupled with a transient thermohydraulic-economic model. The inlet temperature of the heat transfer fluid was varied from low to high, as well as the temperature and load required by the industrial process to obtain networks with optimized area, structure, and operating conditions. For predicting the solar fraction parameter and optimized network area, a regression using artificial neural networks was performed. It was found that it is possible to obtain flexible optimized networks capable of operating with average sf values of 1.05 throughout the year, and with a minimum area that supports changes in the operating conditions of temperature and thermal load of the process in the range of 70-400°C and 0.4-4 MW, respectively.