A Novel Modeling Framework Based on Customized Kernel-Based Fuzzy C-Means Clustering in Iron Ore Sintering Process

This article proposes a customized kernel-based fuzzy C-means (CKFCM) clustering, which provides an original framework for data interpretation and data analysis and delivers an effective solution for an accurate division of actual run data under multiple operating conditions in an iron ore sintering process (IOSP). The improvement of CKFCM clustering is achieved by including an adjustment factor introduced into the kernel-based fuzzy C-means clustering. The adjustment factor only needs to consider a small amount of labeled data that are determined based on expert experience. Subsequently, the CKFCM clustering is applied to the modeling of the IOSP, and the Spearman’s rank correlation coefficient method is utilized to determine input variables of the model under different operating conditions. Furthermore, the broad learning model is employed to build the prediction model for each operating condition. Finally, we conducted an in-depth analysis of the presented clustering method. Its performance has been experimentally demonstrated by many publicly available datasets. Meanwhile, simulation results involving actual run data of the IOSP demonstrate the superiority and effectiveness of the developed model in carbon efficiency prediction. We show that the developed model outperforms the state-of-the-art prediction models in achieving a good balance between simplicity and accuracy.

I. Introduction

With the widespread applications of a distributed control system in complex industrial processes, lots of data have to be collected. There is an evident interest in process monitoring, modeling, optimization, and control realized on the basis of the data-driven methods [1]–[4]. For instance, Yuan et al. [5] proposed a universal data-driven end-to-end framework based on deep learning technology to detect and even predict faults and wearing conditions. Cheng et al. [6] proposed a novel data-driven framework to predict the remaining useful life of bearings by using deep convolutional neural networks. Xue et al. [7] introduced an integrated algorithm to accurately predict the remaining useful life of lithium-ion battery. By analyzing the collected data, one is able to employ them for process analysis and real-time monitoring of production states. The accurate process model is not only important to understand the characteristics of industrial processes, but also the key to the optimization and control aimed at improving the product quality and control performance. This article is concerned with a typical industrial process that is the iron ore sintering process (IOSP).