Citation: Li, Y.; Yang, C.; Sun, Y.
Sintering Quality Prediction Model
Based on Semi-Supervised Dynamic
Time Feature Extraction Framework.
Sensors 2022, 22, 5861. https://
doi.org/10.3390/s22155861
Academic Editor: Jiayi Ma
Received: 2 July 2022
Accepted: 3 August 2022
Published: 5 August 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Sintering Quality Prediction Model Based on Semi-Supervised
Dynamic Time Feature Extraction Framework
Yuxuan Li , Chunjie Yang * and Youxian Sun
State Key Laboratory of Industrial Control Technology, College of Control Science and Engineering,
Zhejiang University, Hangzhou 310027, China; yuxuanli@zju.edu.cn (Y.L.); sunyx@zju.edu.cn (Y.S.)
* Correspondence: cjyang@iipc.zju.edu.cn
Abstract:
In the sintering process, it is difficult to obtain the key quality variables in real time, so
there is lack of real-time information to guide the production process. Furthermore, these labeled
data are too few, resulting in poor performance of conventional soft sensor models. Therefore, a novel
semi-supervised dynamic feature extraction framework (SS-DTFEE) based on sequence pre-training
and fine-tuning is proposed in this paper. Firstly, based on the DTFEE model, the time features of the
sequences are extended and extracted. Secondly, a novel weighted bidirectional LSTM unit (BiLSTM)
is designed to extract the latent variables of original sequence data. Based on improved BiLSTM, an
encoder-decoder model is designed as a pre-training model with unsupervised learning to obtain
the hidden information in the process. Next, through model migration and fine-tuning strategy, the
prediction performance of labeled datasets is improved. The proposed method is applied in the
actual sintering process to estimate the FeO content, which shows a significant improvement of the
prediction accuracy, compared to traditional methods.
Keywords:
LSTM; semi-supervised learning; FeO content; soft sensor; encoder-decoder, dynamic
feature extraction
1. Introduction
The iron and steel industry is the basic industry of the country, and it is also an energy-
intensive process. The energy consumption and emissions of the steel industry account for
a high proportion of a country’s industry. However, the current iron and steel industry is
still at a low level of automation and informatization, and there are still many problems.
For example, in the production process, it is difficult to obtain key information, establish
process control models, and rely on manual experience. Therefore, for example, the product
quality is unstable and the working condition is unstable.
The sintering process is the first key production process in the iron and steel industry. It
provides raw materials for subsequent blast furnace ironmaking and determines the quality
basis of subsequent processes. If the quality of iron ore sinter is poor and fluctuates greatly,
the smelting process of blast furnace will be greatly affected, such as unstable working
conditions and poor molten iron quality. Moreover, the sintering process is one of the most
energy-intensive links in the iron-making process. The whole sintering process mainly
depends on coal and coke as fuel, which produce a large amount of carbon emissions. To
achieve the goals of improving quality, improving production efficiency, saving energy,
protecting the environment, and sustainable development, the intelligent sintering process
will become a research hotspot in academic and industrial circles in the future. Therefore, it
is necessary to study the quality prediction of the sintering process. A prediction model of
FeO is established to predict the quality in time, provide effective guidance information for
operators, and control the production process timely and accurately. An accurate prediction
model is helpful to improve the production quality and efficiency of the sintering process.
Sensors 2022, 22, 5861. https://doi.org/10.3390/s22155861 https://www.mdpi.com/journal/sensors
