Toggle Main Menu Toggle Search

Open Access padlockePrints

Operation optimization of Shell coal gasification process based on convolutional neural network models

Lookup NU author(s): Kangcheng Wang, Dr Jie ZhangORCiD



This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND).


Coal gasification technology has gained increasing popularity in recent years, butthe optimization of operating conditions remains inefficient. The operation optimizationof the Shell coal gasification process (SCGP) is investigated in this paper usingan operation optimization model integrating data analytics and mechanism analysis.The objective function contains two important indicators. One is effective syngasproductivity and the other one is specific oxygen consumption. The optimization issubject to constraints on gasifier temperature and syngas yield. The objective functionand the constraints can be calculated by six key process variables through threeconvolutional neural network (CNN) models, which can additionally utilize the correlationsbetween process variables. Prior physical knowledge and simplified mechanisticmodels of SCGP are integrated with the development of CNN models. Theeffectiveness of the proposed model is validated by an industrial case study. Afterthe operation optimization, the objective function decreases by 28.3306% comparedwith its minimum value on historical process operation data, which outperforms theoperation optimization model developed by ANN models. The sensitivities of theobjective function and effective syngas yield are analyzed. The operating conditionof SCGP can be optimized by the proposed model.

Publication metadata

Author(s): Wang K, Zhang J, Shang C, Huang D

Publication type: Article

Publication status: Published

Journal: Applied Energy

Year: 2021

Volume: 292

Print publication date: 15/06/2021

Online publication date: 07/04/2021

Acceptance date: 16/03/2021

Date deposited: 19/03/2021

ISSN (print): 0306-2619

ISSN (electronic): 1872-9118

Publisher: Elsevier Ltd


DOI: 10.1016/j.apenergy.2021.116847


Altmetrics provided by Altmetric


Funder referenceFunder name
National Natural Science Foundation of China (Nos. 61673236 and 61873142)
No. P7-PEOPLE-2013-IRSES-612230