Differential recurrent neural network based predictive control

An efficient algorithm to train general differential recurrent neural networks is proposed. The trained network can be directly used as the internal model of a predictive controller. The efficiency and effectiveness of the approach are demonstrated through a two-CSTR case study, where a multi-layer perceptron differential recurrent network is adopted.

• Type: Journal
• Year: 2008
• Where: CCE
• Authors: R. K. Al Seyab, Yi Cao
• Keywords: CCE 2007, CCE 2008, Efficient Algorithm, Trained Network, Two-CSTR Case Study