Published Article
Deep Chemometrics using One Dimensional Convolutional Neural Networks for Predicting Crude Oil Properties from FTIR Spectral Data
Souvik Ta, Shahla Alizadeh, Lakshminarayanan Samavedham, Ajay K. Ray
October 19, 2022
The determination of physicochemical properties of crude oils is a very important and time-intensive process that needs elaborate laboratory procedures. Over the last few decades, several correlations have been developed to estimate these properties, but they have been very limited in their scope and range. In recent years, methods based on spectral data analysis have been shown to be very promising in characterising petroleum crude. In this work, the physicochemical properties of crude oils using FTIR spectrums are predicted. A total of 107 samples of FTIR spectral data consisting of 6840 wavenumbers is used. One Dimensional convolutional neural networks (CNNs) were used with FTIR spectral data as the one-dimensional input and Keras and TensorFlow were used for model building. The Root Mean Square Error decreased from 160 to around 60 for viscosity when compared to previous machine learning methods like PLS, PCR, and PLS-GA on the same data. The important hyper-parameters of the CNN were optimised. In addition, a comparison of results obtained with different neural network architectures is presented. Some common preprocessing techniques were also tested on the spectral data to determine their impact on model performance. To increase interpretability, the intermediate neural network layers were analysed to reveal what the convolutions represented, and sensitivity analysis was done to gather key in-sights into which wavenumbers were the most important for prediction of the crude oil properties using the neural network.
Crude Oil Properties, FTIR, Neural Network architectures, One Dimensional Convolutional Neural Network
Suggested Citation
Ta S, Alizadeh S, Samavedham L, Ray AK. Deep Chemometrics using One Dimensional Convolutional Neural Networks for Predicting Crude Oil Properties from FTIR Spectral Data. (2022). LAPSE:2022.0089
Author Affiliations
Ta S: Western University
Alizadeh S: Western University
Samavedham L: Western University, National University of Singapore
Ray AK: Western University
[Login] to see author email addresses.
Journal Name
CSChE Systems & Control Transactions
First Page
Last Page
Publication Date
Version Comments
Original Submission
[Download 1v1.pdf] (381 kB)
Oct 19, 2022
Main Article
CC BY 4.0
Record Statistics
Record Views
Version History
[v1] (Original Submission)
Oct 19, 2022
Verified by curator on
Oct 21, 2022
This Version Number
Most Recent
This Version
URL Here
Original Submitter
Mina Naeini