Combined Artificial Neural Network/Fuzzy Modelling to Optimize the Prototype of Concentrating Solar Tower Using Analytic Hierarchy Process Technique

Abstract

The object of this paper is to simulate and optimize small scale concentrating solar power tower (CSP) built and operationalized at King Abdulaziz University, Jeddah, Saudi Arabia, through analytic hierarchy process (AHP) technique. The aim is to facilitate cost effective integration of solar power coupled with energy generation technologies subjected to challenging climatic conditions; and also to present the effects of changing in parameters such as receiver, heliostats, storage tanks or power generation subsystem on the cost and system performance. This study adopts the AHP technique to obtain the most appropriate receiver shape out of three possible shapes; spherical, cubic, and cylindrical. The used criteria in this in this optimization are reliability, manufacturing in the vicinity, manufacturing cost, service and maintain cost, lower operation risks, and high performance. Based on the results of AHP analysis, square shape is selected. A finite element analysis via ANSYS is performed to compute the through analytic division of temperature in the receiver. The highest temperature from the simulation is 503°C. The thermal power, dispensed by the molten is 12.52 kW during the heat exchanger. However, 13 kW is the design thermal power; while about 3.7% is the percentage error in the thermal power. The findings of this research will provide the needed knowledge and scientific background to assist the authorities concerned in the energy sector in establishing a commercial-scale plant. At the end, Artificial Neural Network algorithms/Fuzzy system is modeled to optimize the process.