ANN Modeling of an ORC-Binary Geothermal Power Plant: Simav Case Study

Abstract

Due to recently developed technologies, there is now more availability to generate electricity. One of these technologies is the binary cycle system, a kind of Organic Rankine Cycle. Since the design of these technologies requires more proficiency and longer times within complex calculations, artificial neural network is a new tool used to make rapid decisions and modeling of the processes within the expertise. The most suitable algorithm was found to be Levenberg-Marguardt with 20 neurons in a single hidden layer for o2 and o3 type cycles, whereas it was found to be Levenberg-Marguardt with 22 neurons for the b3 type cycle.

Keywords:

• artificial neural network
• life cycle cost
• geothermal energy
• Levenberg-Marquardt
• Organic Rankine Cycle-Binary
• Pola-Ribiere conjugate gradient
• scaled conjugate gradient