AUTHOR=Ma Fangyuan , Wang Jingde , Sun Wei 

TITLE=A Data-Driven Semi-Supervised Soft-Sensor Method: Application on an Industrial Cracking Furnace

JOURNAL=Frontiers in Chemical Engineering

VOLUME=Volume 4 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/chemical-engineering/articles/10.3389/fceng.2022.899941

DOI=10.3389/fceng.2022.899941

ISSN=2673-2718

ABSTRACT=Cracking furnace is the key equipment of ethylene unit. Coking in furnace tube is resulted from the generation of coke during cracking, which will compromise the heat transfer efficiency and lead to shape change of tubes. In order to keep cracking furnace operating economically and safely, the engineers need to decoke according to surface temperature of furnace tube. However, the surface temperature of furnace tube is difficult to obtain in practice. Due to the high level of instrumentation and control of the cracking furnaces, a large number of operation data have been collected, which makes it possible to predict surface temperature of furnace tube based on autocorrelation and cross-correlation within and among variables. Traditional prediction methods rely on labeled data samples for training, ignoring the process information contained in a vast amount of unlabeled data. In this work, a data-driven semi-supervised soft-sensor method is proposed. Considering the nonlinear and dynamic relationship among variables, Long Short-Term Memory Network (LSTM) autoencoder (AE), a deep neural network suitable for the feature extraction of long-term nonlinear series, is used for pre-training to extract process data features from unlabeled and labeled data. Then, Principal Component Analysis (PCA) and Mutual Information (MI) are applied to remove feature correlation and select features related to target variables, respectively. Finally, the selected data features are utilized to established a soft-sensor model based on Artificial neural network (ANN). Data from an industrial cracking furnace of ethylene unit is considered to validate the performance of proposed method. The results show that the prediction error of furnace tube surface temperature is about 1%, and successfully aid engineers in determining the optimal time for decoking.