Abstract
Abstract Propylene purity is a very significant quality index in the operation and advanced control of propylene/propane splitter. In this paper, the principal component analysis is proposed to select the secondary variables in a soft-sensor and a propylene purity soft-sensing model is established by a BP neural network. The real-time estimation of propylene purity has been implemented in a refinery and the result shows good performance.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have