Title :
Multiple-prior-knowledge neural network for industrial processes
Author :
Haichuan, Lou ; Hongye, Su ; Lei, Xie ; Yong, Gu ; Gang, Rong
Author_Institution :
Nat. Lab. of Ind. Control Technol., Zhejiang Univ., Hangzhou, China
Abstract :
A novel multiple-prior-knowledge neural network for industrial processes is proposed. Diversity of priori knowledge from industrial processes are discovered and embedded into three-layer hybrid feedforward neural network in the form of penalty function or nonlinear constraints. Meanwhile, Sequential quadratic programming method is selected as the network learning algorithm. Compared with BP neural network and an industrial neural network (the Bounded Derivative neural network, BDNN), MPKN network not only satisfies a certain process mechanism and keeps the process safety, but also has better generalization. Its effective performance is validated with the modelling of a nonlinear process of Continuous-stirred tank reactor.
Keywords :
backpropagation; chemical reactors; embedded systems; feedforward neural nets; quadratic programming; BDNN; MPKN; bounded derivative neural network; continuous stirred tank reactor; feedforward neural network; industrial processes; multiple prior knowledge neural network; network learning algorithm; sequential quadratic programming method; Artificial neural networks; Feedforward neural networks; Fitting; Knowledge engineering; Predictive models; Steady-state; Training; Industrial processes; Multiple-Prior-Knowledge neural network (MPKNN); Nonlinear modeling; Sequential quadratic programming (SQP);
Conference_Titel :
Automation and Logistics (ICAL), 2010 IEEE International Conference on
Conference_Location :
Hong Kong and Macau
Print_ISBN :
978-1-4244-8375-4
Electronic_ISBN :
978-1-4244-8374-7
DOI :
10.1109/ICAL.2010.5585314