The production of textiles has undergone a considerable transformation, progressing from its primitive origins in hand-weaving to the implementation of contemporary automated systems. Weaving yarn into fabric is a crucial process in the textile industry that requires meticulous attention to output quality products, particularly in the tension control section. The efficiency of the tension controller in relation to the yarn tension significantly affects the quality of the resulting fabric, as proper tension control leads to strong, uniform, and aesthetically pleasing fabric, while poor tension control can cause defects and yarn breakage, leading to production downtime and increased costs. Maintaining the desired yarn tension during textile production is crucial, although it poses several problems, such as the continuous diameter change of the unwinder and rewinder sections leading to system change. Another problem faced by the industrial operation is maintaining proper tension on the yarn while changing the roll-to-roll operation velocity. In this paper, an optimized method for controlling yarn tension through the cascade control of tension and position, incorporating feedback controllers, feedforward, and disturbance observers, has been proposed to make the system more robust and suitable for industrial use. In addition, an optimum signal processor has been designed to obtain sensor data with reduced noise and minimal phase difference.

To achieve carbon neutrality and address global energy supply issues by 2050, there is active progress in the industrial sector for waste energy recovery and commercialization projects. It is necessary to consider both the energy recovery efficiency and economic feasibility based on the production volume for the resource utilization of waste energy, along with eco-friendly processing methods. In this study, a waste heat recovery system was designed to recover a large amount of thermal energy from high-temperature exhaust gases of gas engines for power generation by using biogas produced from organic waste in industrial complexes. Types and sizes of components for a waste heat recovery system that were suitable for various engine sizes depending on biogas production were designed, and the energy recovery efficiency was analyzed. The waste heat recovery system consisted of a smoke tube boiler that generated superheated steam at 161 °C under 490 kPa of pressure from the exhaust gas as the heat source, along with two economizers for heating both supply water and hot water. Heat exchangers that were suitable for three different engine sizes were configured, and their performance and energy flow were calculated. In particular, when operating two engines with a power output of 100 kW, the boiler showed the highest steam production efficiency, and the superheated steam production from high-temperature exhaust gas at 600 °C was designed to be 191 kg/h, while hot water at 58 °C was designed to be produced at 1000 kg/h. In addition, further research on the heat exchanger capacity ratio confirmed that it was within a certain range despite the difference in heat exchanger capacity and efficiency depending on the engine size. It was confirmed that the heat exchange capacity ratio of the boiler was important as an optimal-capacity design value for the entire system, as it ranged from 46% to 47% of the total heat exchanger size.

Meltblown nonwoven fabrics are used in various products, such as masks, protective clothing, industrial filters, and sanitary products. As the range of products incorporating meltblown nonwoven fabrics has recently expanded, numerous studies have been conducted to explore the correlation between production process conditions and the performance of meltblown nonwoven fabrics. Deep neural network algorithms, including convolutional neural networks (CNNs), have been widely applied in numerous industries for tasks such as object detection, recognition, classification, and fault detection. In this study, the correlation between the meltblown nonwoven fabric production process and performance was analyzed using deep neural network algorithms for classifying SEM images. The SEM images of meltblown nonwovens produced under various process conditions were trained using well-known convolutional neural network models (VGG16, VGG19, ResNet50, and DenseNet121), and each model showed high accuracy ranging from 95% to 99%. In addition, LRP (Layer-wise Relevance Propagation) and Gradient-weighted Class Activation Mapping (Grad-CAM) models were applied to visualize and analyze the characteristics and correlation of the SEM images to predict the meltblown nonwoven fabric production process.

In light of ongoing advancements in smart manufacturing, there is a growing need for intelligent fault diagnosis methods that maintain reliability under noisy, high-variability operating conditions. Conventional feature selection strategies often struggle when data contain outliers or suboptimal feature subsets, limiting their diagnostic utility. This study introduces a density-based feature space optimization (DBFSO) framework that integrates feature selection with localized density estimation to enhance feature space separability and classifier efficiency. Using k-nearest neighbor density estimation, the method identifies and removes low-density feature vectors associated with noise or outlier behavior, thereby sharpening the feature space and improving class discriminability. Experiments using roll-to-roll (R2R) manufacturing data under mechanical disturbances demonstrate that DBFSO improves classification accuracy by up to 36–40% when suboptimal feature subsets are used and reduces training time by 60–71% due to reduced feature space volume. Even with already-optimized feature sets, DBFSO provides consistent performance gains and increased robustness against operational variability. Additional validation using a bearing fault dataset confirms that the framework generalizes across domains, yielding improved accuracy and significantly more compact, noise-resistant feature representations. These findings highlight DBFSO as an effective preprocessing strategy for intelligent fault diagnosis in intelligent manufacturing systems.