The performance comparison in predicting n-decane pyrolysis process between three ANNs methods: MLP, RBFN and GRNN
- 11 February 2019
- journal article
- research article
- Published by Taylor & Francis in Petroleum Science and Technology
- Vol. 37 (9) , 1053-1058
Abstract
In this work, three kinds of ANNs are employed to predict the pyrolysis behavior of n-decane. Six-fold cross validation method is adopted to optimize the key parameter of ANNs and evaluate the performance of ANNs. The results indicate that RBFN has best performance in predicting the species distribution and all ANNs have good performance in predicting the bulk temperature. Meanwhile, these results reveal that all ANNs are more suitable for predicting products with high mass fraction than those with less mass fraction, whose errors are also acceptable.Keywords
Funding Information
- National Natural Science Foundation of China (91641121)
This publication has 5 references indexed in Scilit:
- The Scramjet Engine2009
- The Use of ANN and the Mathematical Model for Prediction of the Main Product Yields in the Thermal Cracking of NaphthaPetroleum Science and Technology, 2007
- Modeling of thermal cracking of LPG: Application of artificial neural network in prediction of the main product yieldsJournal of Analytical and Applied Pyrolysis, 2007
- MODELING THE HYDROCRACKING PROCESS USING ARTIFICIAL NEURAL NETWORKSPetroleum Science and Technology, 1999
- A general regression neural networkIEEE Transactions on Neural Networks, 1991