Chaiyan Jettanasen

In this study, a double-sheet detection system for automated metal stamping is developed and modeled. This study aims to address the issue of die damage caused by overlapping workpieces (double sheeting), a significant contributor to production line stoppages. The proposed solution involves a control system utilizing a programmable logic controller (PLC), combined with inductive proximity sensors installed on the die, for real-time processing of workpiece status. The system is designed to immediately halt machine operation upon detecting any abnormalities. The experimental results demonstrate that the developed system successfully reduced double-sheet incidents from one occurrence to zero and eliminated production downtime (reducing it from 4 days to zero), resulting in a total of 32,972.26 USD saved in potential damage costs per incident. Furthermore, the system reduced the production cycle time from 11 s to 9.5 s per piece, increasing the production capacity by 1240 pieces per day. The economic assessment indicates that an initial equipment investment of only 483.2 USD yielded a return on investment (ROI) of 6723.73%, a payback period of 5.35 days, a net present value (NPV) of 89,333 USD, and an internal rate of return (IRR) of approximately 6830%. These figures demonstrate a very high level of economic feasibility. Therefore, this system is a cost-effective, reliable, and practical approach to enhance productivity and sustainably support the Smart Factory concept in the metal stamping industry, aligning with the automotive parts manufacturing industry.

This study proposes a programmable logic controller (PLC)–based energy management system integrated with an off-grid photovoltaic (PV) system and battery storage to reduce residential peak demand. The proposed system dynamically manages power supply between the distribution grid, PV generation, and battery storage based on real-time power demand measurements. When the measured power exceeds a predefined threshold, stored renewable energy is utilized to support high-load conditions and mitigate peak demand. Experimental results obtained from a residential-scale test system demonstrate that the proposed off-grid PV system with PLC control can reduce peak demand by up to 29.68% and 15.57% under office-working and work-from-home scenarios, respectively, compared with conventional grid supply and on-grid PV systems without storage. In addition, the system achieves electricity cost reductions of up to 13.63% under Time-of-Use tariffs and up to 11.42% under normal electricity rates, depending on load behavior. These results indicate that integrating PLC-based control with off-grid PV and battery storage can effectively mitigate residential peak demand and reduce electricity expenses under realistic operating conditions.

Most unpaved roads are subjectively classified as Class D roads. However, significant variations exist across different sites and environments (e.g., mining areas). A major challenge in the engineering field is how to quickly correct the Power Spectral Density (PSD) of the unpaved road in question using existing equipment and limited sensors. To address this issue, this study combines real-vehicle test data with a suspension dynamics simulation model. It employs time-domain reconstruction via Inverse Fast Fourier Transform (IFFT) and wavelet processing methods to construct an optimized model that fuses time-frequency domain features. With the help of a surrogate optimization method, the model achieves the best approximation of the actual road surface, corrects the PSD parameters of the unpaved road, and provides a reliable input basis for vehicle dynamics simulation, fatigue life prediction, and performance evaluation.

Effective communication between deaf–mute and visually impaired individuals remains a challenge in the fields of human–computer interaction and accessibility technology. Current solutions mostly rely on single-modal recognition, which often leads to issues such as semantic ambiguity and loss of emotional information. To address these challenges, this study proposes a lightweight multimodal fusion framework that combines gestures and micro-expressions, which are then processed through a recognition network and a speech synthesis module. The core innovations of this research are as follows: (1) a lightweight YOLOv5s improvement structure that integrates residual modules and efficient downsampling modules, which reduces the model complexity and computational overhead while maintaining high accuracy; (2) a multimodal fusion method based on an attention mechanism, which adaptively and efficiently integrates complementary information from gestures and micro-expressions, significantly improving the semantic richness and accuracy of joint recognition; (3) an end-to-end real-time system that outputs the visual recognition results through a high-quality text-to-speech module, completing the closed-loop from “visual signal” to “speech feedback”. We conducted evaluations on the publicly available hand gesture dataset HaGRID and a curated micro-expression image dataset. The results show that, for the joint gesture and micro-expression tasks, our proposed multimodal recognition system achieves a multimodal joint recognition accuracy of 95.3%, representing a 4.5% improvement over the baseline model. The system was evaluated in a locally deployed environment, achieving a real-time processing speed of 22 FPS, with a speech output latency below 0.8 s. The mean opinion score (MOS) reached 4.5, demonstrating the effectiveness of the proposed approach in breaking communication barriers between the hearing-impaired and visually impaired populations.

Aiming at the characteristics of limited actuation capability of the semi-active control system and strong nonlinearity of the hydro-pneumatic suspension, a constrained nonlinear control strategy of a semi-active hydro-pneumatic suspension system is proposed. According to the mathematical model of nonlinear hydro-pneumatic suspension, the static stiffness and linear damping coefficient based on the equivalent energy are calculated, and then the control-oriented dynamic equation whose expression minimizes the nonlinear term is constructed. Combined with actuation capacity constraints, an optimization model with constraints is established to minimize the deviation between the actual overall control force and the expected optimal control force, and the optimal approximation from nonlinear control to linear quadratic optimal control is realized. The control simulation results of various methods show that the nonlinear control with constraints of the semi-active hydro-pneumatic suspension system, which effectively combines the actuation capacity constraints and nonlinear characteristics of the system, achieves a good comprehensive control effect for the nonlinear suspension control with constraints.

Sign language plays a crucial role in bridging communication barriers within the Deaf community. Recognizing sign language sentences remains a significant challenge due to their complex structure, variations in signing styles, and temporal dynamics. This study introduces an innovative sign language sentence recognition (SLSR) approach using Hybrid Graph Embedding and Adaptive Convolutional Networks (HGE-ACN) specifically developed for single-handed wearable glove devices. The system relies on sensor data from a glove with six-axis inertial sensors and five-finger curvature sensors. The proposed HGE-ACN framework integrates graph-based embeddings to capture dynamic spatial–temporal relationships in motion and curvature data. At the same time, the Adaptive Convolutional Networks extract robust glove-based features to handle variations in signing speed, transitions between gestures, and individual signer styles. The lightweight design enables real-time processing and enhances recognition accuracy, making it suitable for practical use. Extensive experiments demonstrate that HGE-ACN achieves superior accuracy and computational efficiency compared to existing glove-based recognition methods. The system maintains robustness under various conditions, including inconsistent signing speeds and environmental noise. This work has promising applications in real-time assistive tools, educational technologies, and human–computer interaction systems, facilitating more inclusive and accessible communication platforms for the deaf and hard-of-hearing communities. Future work will explore multi-lingual sign language recognition and real-world deployment across diverse environments.

Gesture is a natural and intuitive means of interpersonal communication. Sign language recognition has become a hot topic in scientific research, holding significant importance and research value in fields such as deep learning, human–computer interaction, and pattern recognition. The sign language recognition process needs to ensure real-time performance and ease of deployment. Based on these two requirements, this paper proposes an improved YOLOv5s-based sign language recognition algorithm. Firstly, the lightweight concept from ShuffleNetV2 was applied to achieve lightweight characteristics and improve the model’s deployability. The specific improvements are as follows: The algorithm achieved model size reduction by removing the Focus layer, using the ShuffleNetv2 algorithm, and then channel pruning YOLOv5 at the head of the neck layer. All the convolutional layers and the cross-stage partial bottleneck layer with three convolutional layers in the backbone network were replaced with ShuffleBlock, the spatial pyramid pooling layer and a subsequent cross-stage partial bottleneck layer structure with three convolutional layers were removed, and the cross-stage partial bottleneck layer module with three convolutional layers in the detection header section was replaced with a depth-separable convolutional module. Experimental results show that the parameters of the improved YOLOv5 algorithm decreased from 7.2 M to 0.72 M, and the inference speed decreased from 3.3 ms to 1.1 ms.

The sign language recognition system has placed an important role in disabled people’s lives. The researchers utilize various methods to fulfill disabled people’s requirements. However, the methods fail to access their sign at a reasonable cost with minimum computational difficulties. The improper sign access causes a reduction of the sign language recognition accuracy. Therefore, this study uses the sensor of wearable devices to capture people’s motions and actions to identify sign language. The collected information is fused into a competition level that understands disabled people’s requirements. After that, fused information is processed by the Pareto Optimized Hypertuned Deep Elman Neural Model (POHDENM). In addition, the data augmentation process is incorporated to train the data, which helps to recognize the large volume of sentences. Here, the fused information is split into words processed by the neural model that recognizes the sign language features. The extracted features are analyzed by an optimized method that recognizes the language with maximum accuracy. During the analysis, network parameters are hyper-tuned with the help of the Pareto model, which reduces the misrecognition error rate. Then, the created system efficiency is evaluated using the experimental analysis. The POHDENM approach achieves a high accuracy of 97.86% and an F1-score of 98.08%. The model’s high performance is achieved by fine-tuning its hyperparameters using the Pareto optimization algorithm, which balances precision of 98.06% and recall of 98.12%.