Assessment of Texaco Syngas Components Using Extreme Learning Machine Based Quantum Neural Network - Intelligent Computing for Sustainable Energy and Environment - page 221

Assessment of Texaco Syngas Components
Using Extreme Learning Machine
Based Quantum Neural Network
Wei Xu 1 , , Raofen Wang 2 , Xingsheng Gu 3 , and Youxian Sun 1
1 State Key Laboratory of Industrial Control Technology,
Zhejiang University, 310027 Hangzhou, China
[e-mail]
2 Department of Automation, Shanghai University of Engineering Science,
201620 Shanghai, China
3 Key Laboratory of Advanced Control and Optimization for Chemical Processes of
Ministry of Education, East China University of Science and Technology, 200237
Shanghai, China
Abstract. Quantum neural computing has nowadays attracted much
attention, and tends to be a candidate to improve the computational ef-
ficiency of neural networks. In this paper, a new quantum neural network
(QNN) is proposed based on quantum mechanics of superposition and
collapse, etc. Instead of gradient descent methods and evolutionary al-
gorithms, extreme learning machine (ELM) is introduced to analytically
identify the parameters of the QNN. The ELM-QNN model is applied
to the online and real-time assessment of the syngas components in a
Texaco gasification process. The application would effectively avoid the
problems of time delay and low accuracy which result from the man-
ual analysis. In order to eliminate the redundant information stored in
variables, principal component analysis (PCA) is adopted to reduce the
number of input variables of ELM-QNN. The results indicate that ELM-
QNN combined with PCA method has satisfied computational accuracy
and eciency. The PCA-ELM-QNN is very capable of being used for the
real-time measurement of Texaco syngas components.
Keywords: Extreme learning machine, Quantum neural network, Tex-
aco gasification, Syngas component, Principal component analysis.
1 Introduction
In recent years, ANN has been applied successfully to pattern recognition, auto-
matic control, system modeling, signal processing, etc. However, it has the draw-
backs of slow processing speed, limited memory storage and iterative learning.
Owing to these, many researchers began to integrate other theories for improv-
ing the performance of ANN. The superposition of quantum mechanics provides