Optimal design of a concentrated solar power plant with a thermal energy storage system using the Downhill Simplex Method

Abstract

In this work, a concentrated solar power (CSP) plant with a thermal energy storage system to produce 120 megawatts of electrical energy was designed using Thermoflex. As non-renewable resources are limited, to move towards sustainable development, the importance of developing alternative technologies such as optimized solar power generation systems becomes considerable. To eliminate the gap between electricity demand and power supply in solar power plants, it is necessary to use optimized thermal energy storage systems. So, in this study the Nelder-Mead algorithm as a novel optimization method was coupled with Thermoflex in the Excel environment to obtain an optimized techno-economic design. Finding the minimum of the Rosenbrock benchmark function confirmed that the simplex algorithm can find optimal solutions in a short time for complicated optimization problems. For plant optimization, twelve parameters that have the greatest effect on the objective function were selected as decision variables. The optimization results after 77 iterations demonstrated that in the optimal case, the objective function decreases from 4.48 to 4.38. This was equivalent to a 2.98 MWe power increase, while the cost of CSP components remained at an acceptable level. The total heat transfer for the solar boiler and feed water heater train sections before and after optimization is 391.816 MWth and 395.033 MWth, respectively.