Sparsity constrained wavelet neural networks for robust soft sensor design with application to the industrial KIVCET unit. (March 2022)

• Record Type: Journal Article
• Authors: Varanasi, Santhosh Kumar; Daemi, Atefeh; Huang, Biao; Slot, Graham; Majoko, Primo
• Is Part Of: Computers & chemical engineering. Volume 159(2022)
• Publication Date: 2022-03
• Subjects: Wavelet neural networks -- Soft sensor -- Input selection -- Sparse optimization -- Interior-point methods; Chemical engineering -- Data processing -- Periodicals; 660.0285
• DOI: 10.1016/j.compchemeng.2022.107695
• Languages: English
• ISSNs: 0098-1354

Abstract

Highlights: A sparsity constrained wavelet neural network model for soft sensor design is proposed. The problem of redundant inputs and over-fitting is addressed by a zero-norm constraint. The proposed model can identify optimal number of nodes in hidden layer. The problem is solved using the Greedy Sparse Simplex algorithm. Robustness and performance are demonstrated by a numerical and an industrial case study. Abstract: The main objective of this paper is to develop a soft sensor for predictions of the residue element concentrations of feed mixture in a KIVCET furnace, since such measurements are not available online and are crucial in achieving the objective of real-time optimization. To realize this, a wavelet neural network is considered owing to its high accuracy in approximating functions with a few functional components over networks with sigmoid/tanh activation functions. The idea of wavelet decompositions is used to address the challenges associated with random initialization in such networks. A one-norm penalty on the data-fit term is considered to make the approach robust to outliers, and a sparsity constraint on the parameter vector in terms of zero-norm is considered to handle the issues of over-fitting and input redundancy. The efficiency of the proposed method is demonstrated through an industrial KIVCET unit and a simulation study.