RESUMEN

This paper introduces a novel method for enhancing fault classification and diagnosis in dynamic nonlinear processes. The method focuses on dynamic feature extraction within multivariate time series data and utilizes dynamic reconstruction errors to augment the feature set. A fault classification procedure is then developed, using the weighted maximum scatter difference (WMSD) dimensionality reduction criterion and quadratic discriminant analysis (QDA) classifier. This method addresses the challenge of high-dimensional, sample-limited fault classification, offering early diagnosis capabilities for online samples with smaller amplitudes than the training set. Validation is conducted using a cold rolling mill simulation model, with performance compared to classical methods like linear discriminant analysis (LDA) and kernel Fisher discriminant analysis (KFD). The results demonstrate the superiority of the proposed method for reliable industrial process monitoring and fault diagnosis.

RESUMEN

Dewatering is critical to oily cold rolling mill (CRM) sludge treatment. Therefore, finding an efficient, energy-saving, and applicable dewatering technology for oily CRM sludge is still urgent. This study investigated the performance of quicklime as a conditioning agent for oily CRM sludge conditioning and dewatering. The interactive effects of quicklime dosage, temperature, and time on filter cake's specific resistance to filtration (SRF) and the dewatering rate of oily CRM sludge were studied by response surface methodology (RSM). The optimal parameters for conditioning oily CRM sludge were quicklime dosage of 18.7%, temperature of 54 °C, and time of 43.3 min, which resulted in filter cake SRF of 0.50 × 1010 m/kg and dewatering rate of 61.2%. The viscosity of oily CRM sludge could be reduced by 90% after conditioned with quicklime, which caused by the neutralization or hydrolysis of high viscosity organic matter in the oil phase with quicklime to produce low viscosity organic matter. The study indicated the excellent performance of quicklime as a conditioning agent for oily CRM sludge treatment and provided an effective route for the recycling of the oily CRM sludge for steel production.

Asunto(s)

Óxidos , Aguas del Alcantarillado , Compuestos de Calcio , Frío , Filtración , Aceites , Agua , Eliminación de Residuos Líquidos/métodos

RESUMEN

High-strength non-oriented electro-technical steels with a low thickness possess excellent isotropy of electromagnetic and mechanical properties which is highly required in the production of high-efficiency electric motors. The manufacturing process of this type of steel includes very important and technologically complex routes such as hot rolling, cold rolling, temper rolling, or final heat treatment. The final thickness is responsible for the decrease in eddy-current losses and is effectively achieved during cold rolling by the tandem rolling mill. Industrial production of thin sheets of high-strength silicon steels in high-speed tandem rolling mills is a rather demanding technological operation due to the increased material brittleness that is mainly caused by the intensive solid solution and deformation strengthening processes, making the dislocation motion more complex. The main objective of this work was to investigate the distribution of local mechanical strains through the thickness of high silicon steel hot bands, generated during the cold rolling. The experimental samples were analysed by means of electron back-scattered diffraction and scanning electron microscopy. From the performed analyses, the correlation between the material workability and the nucleation of cracks causing the observed steel strip failure during the tandem cold rolling was characterized. Specifically, the microstructural, textural, misorientation, and fractographic analyses clearly show that the investigated hot band was characterized by a bimodal distribution of ferrite grains and the formation of intergranular cracks took place only between the grains with recrystallized and deformed structures.

RESUMEN

The data contained in this article was acquired from the automatic gauge control system for a steel cold rolling mill production line in Nigeria. Accuracy is the one of the most important indices of productivity during a milling process. A total of 486 data points were obtained from selected feedback sensors located on the rolling mill machine via the control panel Human Machine Interface (HMI). The selected rolling parameters were gathered at different time intervals for different sample coils strips during the different milling stages. The data shows parameters such as actual thickness measured, x-ray gauge temperature, mill speed at both entry and exit and the mill power. This dataset could be used to analyze and improve the accuracy of the Automatic gauge control system and reduction in error in thickness variation.