Chemical Engineering Transactions (Nov 2021)

Optimal Design of Solar-Aided Hydrogen Production Process Using Molten Salt

  • Wanrong Wang,
  • Nan Zhang,
  • Jie Li

DOI
https://doi.org/10.3303/CET2188142
Journal volume & issue
Vol. 88

Abstract

Read online

In this work, a machine-learning based optimisation framework is proposed for optimal design of solar steam methane reforming using molten salt (SSMR-MS) through machine learning techniques. The artificial neural network (ANN) is employed to establish relationships between total annualised cost (TAC), hydrogen production rate and molten salt duty, and independent input variables in SSMR-MS. A hybrid global optimisation algorithm is adopted to solve the developed surrogate model and generate the optimal design. The computational results demonstrate that a significant reduction in TAC by around 15 % can be achieved than the existing SSMR-MS. The lowest Levelised cost of Hydrogen Production (LCHP) is 2.43 $ kg-1 which is reduced by around 15 % compared to the existing process with LCHP of 2.85 $ kg-1.