Effective modeling and management are critical in wastewater treatment facilities since the aeration process accounts for 65–70% of the overall energy consumption. This study assesses control strategies specifically designed for different sizes of WWTP, analyzing their economic, environmental, and energy-related effects. Small WWTPs see advantages from the utilization of on/off and proportional–integral–derivative (PID) control methods, resulting in 10–25% energy savings and the reduction in dissolved oxygen (DO) levels by 5–30%. Cascade control and model predictive control (MPC) improve energy efficiency by 15–30% and stabilize DO levels by 15–35% in medium-sized WWTPs. Advanced WWTPs that utilize technologies such as MPC integrated with artificial intelligence (AI) and machine learning (ML) can decrease energy usage by 30–40% and enhance DO levels by 35–40%. Life cycle assessment (LCA) demonstrates substantial decreases in greenhouse gas (GHG) emissions: 5–20% for small, 10–25% for medium, and 30–35% for large WWTPs. These findings illustrate the feasibility and expandability of these tactics in both controlled laboratory environments and real-world situations, emphasizing the significance of customized methods for improving energy efficiency and sustainability in wastewater treatment. Subsequent investigations should prioritize integrating renewable energy sources and resolving obstacles in developing nations to enhance wastewater treatment plants’ energy efficiency and sustainability.

Stamping is one of the most widely used processes in the sheet metalworking industry. Because of the increasing demand for a faster process, ensuring that the stamping process is conducted without compromising quality is crucial. The tool used in the stamping process is crucial to the efficiency of the process; therefore, effective monitoring of the tool health condition is essential for detecting stamping defects. In this study, vibration measurement was used to monitor the stamping process and tool health. A system was developed for capturing signals in the stamping process, and each stamping cycle was selected through template matching. A one-dimensional (1D) convolutional neural network (CNN) was developed to classify the tool wear condition. The results revealed that the 1D CNN architecture a yielded a high accuracy (>99%) and fast adaptability among different models.

This study focuses on the development of a novel polishing pad for SiC wafers. Fe and Al₂O₃ particles were impregnated in a polyurethane matrix, thus forming a fixed abrasive polishing pad. Four types of pads with different compositions of Fe and Al₂O₃ were fabricated. A combination of loose and fixed polishing methods was used for polishing with the fabricated pads and was investigated to improve the polishing process. The surface characteristics of the polished SiC wafer and the SiC removal rate during polishing using the designed pads were examined and compared with those for SiC polished with a conventional polyurethane pad. Experimental results showed that the removal rate for SiC in the case of polishing with the pads consisting 1 wt % Fe and 3 wt % Al₂O₃ particles was approximately 73% higher than that observed when polishing using the conventional polyurethane polishing pad. Additionally, the surface roughness of the resulting SiC wafers after polishing with the Fe and Al₂O₃-impregnated pads was identical to that when using the conventional polyurethane pad, without any surface damage. The results indicated that the Fe and Al₂O₃-impregnated pads can be effectively used for SiC wafer polishing. When the proposed process was employed for polishing single-crystal SiC, both the polishing time and cost were reduced. This novel design can facilitate the extensive use of single-crystal SiC wafers in the future.

With the development of intelligent machine tools, monitoring the vibration by the accelerometer is an important issue. Accelerometers used for measuring vibration signals during milling processes require the characteristics of high sensitivity, high resolution, and high bandwidth. A commonly used accelerometer is the lead zirconate titanate (PZT) type; however, integrating it into intelligent modules is excessively expensive and difficult. Therefore, the micro electro mechanical systems (MEMS) accelerometer is an alternative with the advantages of lower price and superior integration. In the present study, we integrated two MEMS accelerometer chips into a low-pass filter and housing to develop a low-cost dual-axis accelerometer with a bandwidth of 5 kHz and a full scale range of ±50 g for measuring machine tool vibration. In addition, a platform for measuring the linearity, cross-axis sensitivity and frequency response of the MEMS accelerometer by using the back-to-back calibration method was also developed. Finally, cutting experiments with steady and chatter cutting were performed to verify the results of comparing the MEMS accelerometer with the PZT accelerometer in the time and frequency domains. The results demonstrated that the dual-axis MEMS accelerometer is suitable for monitoring the vibration of machine tools at low cost.

Stamping processes remain crucial in manufacturing processes; therefore, diagnosing the condition of stamping tools is critical. One of the challenges in diagnosing stamping tool conditions is that traditionally, the tools need to be visually checked, and the production processes thus need to be halted. With the development of Industry 4.0, intelligent monitoring systems have been developed by using accelerometers and algorithms to diagnose the wear classification of stamping tools. Although several deep learning models such as the convolutional neural network (CNN), auto encoder (AE), and recurrent neural network (RNN) models have demonstrated promising results for classifying complex signals including accelerometer signals, the practicality of those methods are restricted due to the flexibility of adding new classes and low accuracy when faced to low numbers of samples per class. In this study, we applied deep metric learning (DML) methods to overcome these problems. DML involves extracting meaningful features using feature extraction modules to map inputs into embedding features. We compared the probability method, the contrastive method, and a triplet network to determine which method was most suitable for our case. The experimental results revealed that, compared with other models, a triplet network can be more effectively trained with limited training data. The triplet network demonstrated the best test results of the compared methods in the noised test data. Finally, when tested using unseen class, the triplet network and the probability method demonstrated similar results.

During the movement of a robotic arm, collisions can easily occur if the arm directly grasps at multiple tightly stacked objects, thereby leading to grasp failures or machine damage. Grasp success can be improved through the rearrangement or movement of objects to clear space for grasping. This paper presents a high-performance deep Q-learning framework that can help robotic arms to learn synchronized push and grasp tasks. In this framework, a grasp quality network is used for precisely identifying stable grasp positions on objects to expedite model convergence and solve the problem of sparse rewards caused during training because of grasp failures. Furthermore, a novel reward function is proposed for effectively evaluating whether a pushing action is effective. The proposed framework achieved grasp success rates of 92% and 89% in simulations and real-world experiments, respectively. Furthermore, only 200 training steps were required to achieve a grasp success rate of 80%, which indicates the suitability of the proposed framework for rapid deployment in industrial settings.