Intelligent manufacturing, defined as the integration of manufacturing with modern information technologies such as 5G, digitalization, networking, and intelligence, has grown in popularity as a means of boosting the productivity, intelligence, and flexibility of traditional manufacturing processes. The steel industry is a necessary support for modern life and economic development, and the Chinese steel industry’s capacity has expanded to roughly half of global production. However, the Chinese steel industry is now confronted with high labor costs, massive carbon emissions, a low level of intelligence, low production efficiency, and unstable quality control. Therefore, China’s steel industry has launched several large-scale intelligent manufacturing initiatives to improve production efficiency, product quality, manual labor intensity, and employee working conditions. Unfortunately, there is no comprehensive overview of intelligent manufacturing in China’s steel industry. We began this research by summarizing the construction goals and overall framework for intelligent manufacturing of the steel industry in China. Following that, we offered a brief review of intelligent manufacturing for China’s steel industry, as well as descriptions of two typical intelligent manufacturing models. Finally, some major technologies employed for intelligent production in China’s steel industry were introduced. This research not only helps to comprehend the development model, essential technologies, and construction techniques of intelligent manufacturing in China’s steel industry, but it also provides vital inspiration for the manufacturing industry’s digital and intelligence updates and quality improvement.

The quantitative analysis accuracy of calibration-free laser-induced breakdown spectroscopy (CF-LIBS) is severely affected by the self-absorption effect and estimation of plasma temperature. Herein, a CF-LIBS quantitative analysis method based on the auto-selection of internal reference line and the optimized estimation of plasma temperature is proposed. The internal reference line of each species is automatically selected from analytical lines by a programmable procedure through easily accessible parameters. Furthermore, the self-absorption effect of the internal reference line is considered during the correction procedure. To improve the analysis accuracy of CF-LIBS, the particle swarm optimization (PSO) algorithm is introduced to estimate the plasma temperature based on the calculation results from the Boltzmann plot. Thereafter, the species concentrations of a sample can be calculated according to the classical CF-LIBS method. A total of 15 certified alloy steel standard samples of known compositions and elemental weight percentages were used in the experiment. Using the proposed method, the average relative errors of Cr, Ni, and Fe calculated concentrations were 4.40%, 6.81%, and 2.29%, respectively. The quantitative results demonstrated an improvement compared with the classical CF-LIBS method and the promising potential of in situ and real-time application.

Different fault types of rolling bearings correspond to different features, and classical deep learning models using a single attention mechanism (AM) have limitations in capturing feature diversity. Therefore, a novel dual attention mechanism network (DAMN) with self-attention (SA) and frequency channel attention (FCA) is proposed for rolling bearing fault diagnosis. The SA mechanism is used to capture global relationships between the input features and fault types, and the FCA mechanism applies multi-spectral attention to learn the local useful information among different input channels. The results of the ablation study on the effects of FCA blocks showed that including a proper combination of multiple frequency components is helpful in achieving higher accuracy. Experiments were conducted to diagnose rolling bearings with multiple types of faults. The results show that, compared with current fault diagnosis models, the proposed DAMN has better comprehensive performance in terms of diagnosis accuracy and model convergence speed. It was also demonstrated that the backbone of DAMN based on a dual AM could achieve better performance than the backbone based on a single AM.

It is a challenge task to enhance the analysis accuracy of laser-induced breakdown spectroscopy (LIBS) in high temperature applications when certified standard samples used for building calibration curves at high temperature are limited or not available. A novel LIBS quantitative analysis method for alloy steel at high temperature via feature-based transfer learning (FTL) is proposed. The spectral data of calibration samples at room temperature and the spectral data of uncalibrated samples at high temperature are together transferred into a high-dimensional feature space using kernel function mapping where an LIBS regression model is trained and established. For testing samples, the measured spectra at high temperature are mapped into the high-dimensional feature space with the same kernel parameters used in the training process, and then the concentration results can be obtained by the regression model. Experiments on certified alloy steel standard samples were conducted, in which 12 samples with both the concentration information and the measured spectra at room temperature and 8 samples only with the spectra measured at high temperature were used to train the analysis model. The 8 samples at high temperature were used for testing. The experimental results of the Cr concentration showed that with feature-based transfer learning, the mean relative error decreased from 32.31% to 6.08%. The proposed method does not need the element concentration for samples at high temperature to build the regression model, which provides a feasible and effective approach for LIBS analysis of samples at high temperature, such as fast industrial measurements in iron and steel smelting production processes.

Data fluctuation in multiple measurements of Laser Induced Breakdown Spectroscopy (LIBS) greatly affects the accuracy of quantitative analysis. Our proposed method achieved better prediction accuracy and modeling robustness.