Deep Learning of Partially Labeled Data for Quality Prediction Based on Stacked Target-Related Laplacian Autoencoder.

Partially labeled data, which is common in industrial processes due to the low sampling rate of quality variables, remains an important challenge in soft sensor applications. In order to exploit the information from partially labeled data, a target-related Laplacian autoencoder (TLapAE) is proposed in this work. In TLapAE, a novel target-related Laplacian regularizer is developed, which aims to extract structure-preserving and quality-related features by preserving the feature-target mapping according to the local geometrical structure of the data. In addition, stacked TLapAE (STLapAE) is further constructed to extract deep feature representations of the data by hierarchically stacking TLapAE blocks. For model training, backward propagation equations are derived based on matrix calculus techniques to update the model parameters of the proposed TLapAE. The effectiveness of the proposed STLapAE is evaluated using the butane content prediction case in a debutanizer column, the silicon content prediction case in a blast furnace (BF) ironmaking process, and the ethane concentration prediction case in an ethylene fractionator. The results show that the proposed TLapAE model has significantly improved prediction accuracy compared to soft sensors using only labeled data and other partially labeled data modeling methods.

Multi-Gate Mixture-of-Experts Stacked Autoencoders for Quality Prediction in Blast Furnace Ironmaking.

The blast furnace is an energy-intensive and extremely complex reactor in the ironmaking process. To reduce energy consumption, improve product quality, and ensure the stability of blast furnace operation, it is very important to predict the quality indicators of molten iron accurately and in real time. However, most of the existing product quality prediction models, such as the stacked autoencoder (SAE) model, use a single-channel stack structure. For such models, when the working conditions of the blast furnace ironmaking process change, a large prediction error will occur. To solve this issue, this paper develops a novel deep learning model, called the multi-gate mixture-of-experts stacked autoencoder (MMoE-SAE), for predicting the quality variable in the blast furnace ironmaking processes. The proposed MMoE-SAE model is constructed based on a multi-gate hybrid expert structure, in which a series of SAE networks are selected as experts. The MMoE-SAE model inherits the advantages of MMoE and SAE, which can not only extract the deep features of the data but also have better adaptability to the changes of working conditions in the blast furnace ironmaking process. To verify the effectiveness and practicability of the proposed MMoE-SAE model, it was applied to predict the silicon content of molten iron in the blast furnace ironmaking process. The experimental results demonstrate that the proposed MMoE-SAE model outperforms other prediction models in prediction accuracy.