Maintaining a normal sintering condition is vital to ensuring the quality of nonferrous metals in a rotary kiln. Identification of the sintering conditions is a crucial component of a condition control system. In order to exploit the dynamic information of the flame combustion process, i.e., multiview of flame. This article presents a generalized dynamic feature extraction method to improve the sintering condition identification accuracy. Compared to most of the current methods, our approach introduces the concept of multiview subspace clustering that reveals the manifold structure of the data through the symmetric positive definite (SPD) manifold termed invariant SPD manifold representation multiview subspace clustering (IMMSC). Moreover, our method is generalized to extract dynamic features for most static features. With the manifold metric's affine invariance property, we demonstrate that IMMSC can extract dynamic features to enhance classification accuracy. The proposed method can be efficiently optimized by manifold optimization. Experimental results using real datasets and coil-20-proc dataset demonstrate that the proposed method for recognizing sintering condition are effective and robust.

The quality of training data plays a decisive role in the establishment of intelligent models. Since raw data obtained from the real world are usually entwined with noise due to variety of causes, noise filtering has become an important aspect of machine learning techniques. In contrast with the extensive research conducted on noise elimination for classification purposes, papers addressing this problem for regression tasks are rather scarce. In this paper, we propose a novel noise filter to clean noisy instances with real-valued label noise. Aiming at the deficiency of the existing noise determination criterion, a new adaptive threshold-based method is first proposed. It allows a noisy instance to be adaptively defined according to the fitting difficulty levels of different datasets, and areas with different densities. Embedded with this criterion, an effective noise filtering procedure is also designed. An ensemble filtering scheme and an iterative filtering process are combined to detect as many potential noisy samples as possible from the original training set. According to the acquire noise detection information, a noise score for evaluating the noise level is specifically developed. The potential noisy samples whose scores exceed a reasonable threshold are further filtered, which can compensate for the possible errors incurred during the previous procedure, and contribute to more reliable filtering results. The validity of the proposed method is studied in exhaustive experiments. We discuss reasonable hyperparameters, and compare the developed method with several state-of-the-art noise filters. The outcomes show that the prediction accuracy of the utilized regressor can greatly benefit from preprocessing the given raw dataset by using our method. Simultaneously, the method is able to acquire a good balance between the elimination of noisy samples and the retention of clean samples, and consistently achieves a better noise filtering performance.

Combined the modified AdaBoost.RT with extreme learning machine (ELM), a new hybrid artificial intelligent technique called ensemble ELM is developed for regression problem in this study. First, a new ELM algorithm is selected as ensemble predictor due to its rapid speed and good performance. Second, a modified AdaBoost.RT is proposed to overcome the limitation of original AdaBoost.RT by self-adaptively modifying the threshold value. Then, an ensemble ELM is presented by using the modified AdaBoost.RT for better accuracy of predictability than individual method. Finally, this new hybrid intelligence method is used to establish a temperature prediction model of molten steel by analyzing the metallurgic process of ladle furnace (LF). The model is examined by data of production from 300t LF in Baoshan Iron and Steel Co., Ltd. and compared with the models that established by single ELM, GA-BP (combined genetic algorithm with BP network), and original AdaBoost.RT. The experiments demonstrated that the hybrid intelligence method can improved generalization performance and boost the accuracy, and the accuracy of the temperature prediction is satisfied for the process of practical producing.

The aim of this paper is to develop a nonlinear orthogonal signal correction (OSC) algorithm using kernel-based technique, termed as kernel OSC (KOSC), and investigate its effects on multivariate calibration. As a nonlinear data pretreatment, the proposed KOSC method can better analyze the nonlinear relationships between descriptor and response variables and remove from process measurement those undesirable variations not correlated with process property from a nonlinear point of view, which well prepares the corrected process trajectory for the subsequent calibration modeling. Two data sets are employed in illustration experiment. It is found that nonlinear OSC plus nonlinear calibration algorithm seems to have the superiority over other methods to improve the interpretation ability of regression model when process data nonlinearly vary with quality.