Effective detection of unexpected events in wide-area surveillance remains a critical challenge in the development of intelligent monitoring systems. Recent advancements in Narrowband Internet of Things (NB-IoT) and 5G technologies provide a robust foundation to address this issue. This study presents an integrated architecture for real-time event detection and response. The system utilizes the Constrained Application Protocol (CoAP) to transmit encapsulated JPEG images from NB-IoT modules to an Internet of Things (IoT) server. Upon receipt, images are decoded, processed, and archived in a centralized database. Subsequently, image data are transmitted to client applications via WebSocket, leveraging the Message Queuing Telemetry Transport (MQTT) protocol. By performing temporal image comparison, the system identifies abnormal events within the monitored area. Once an anomaly is detected, a visual alert is generated and presented through an interactive interface. The test results show that the image recognition accuracy is consistently above 98%. This approach enables intelligent, scalable, and responsive wide-area surveillance reliably, overcoming the constraints of conventional isolated and passive monitoring systems.
Simulated moving bed (SMB) is one of the most successful technologies in chromatographic separation, which is widely used in chemical industry and food industry. There are many parameters that can affect the separation effect of SMB, and the sensitivity is very strong, so the effective control is always a difficulty problem for SMB. In this paper, a self-tuning advanced fuzzy controller is proposed to control the purity of the SMB outlet, and the study finds it has smaller steady-state error and better adaptability of in SMB when compares with the traditional and advance fuzzy type controller.
Chromatographic separation plays a pivotal role in the manufacturing of chemical products and biopharmaceuticals. This technique exploits differences in distribution between stationary and mobile phases to separate mixtures, impacting final product quality. Simulated moving bed (SMB) technology, recognized for its continuous feed, enhances efficiency and increases production capacity while reducing solvent and water consumption. Despite its complexity in controlling variables like flow rates and valve switching times, traditional control theories fall short. This study introduces an intelligent fuzzy controller resembling an approximate neural network (NN) for SMB control. Simulation results demonstrate the controller’s effectiveness in achieving desirable outcomes for the SMB system.
25 March 2021 in Journal of Ambient Intelligence and Humanized Computing
Chromatographic separation is a critical technique in the manufacturing processes of both chemical products and biopharmaceuticals. Its principle relies on exploiting the differences in distribution between the stationary and mobile phases to achieve the separation of mixtures. The precision of substance concentration during separation directly impacts the quality and usability of the final product. Therefore, the development of an effective and precise separation control technique has long been a vital concern in the field of chromatographic separation control. Currently, the employment of Simulated Moving Bed (SMB) technology for chromatographic separation, which allows for continuous feed, has been recognized as a cutting-edge technique. SMB is a continuous process technology that enhances the efficiency of adsorbents within the bed. This sequential multi-column SMB technology not only increases production capacity but also reduces the consumption of solvents and water. It is acknowledged as one of the cleanest manufacturing technologies in the biopharmaceutical industry. However, multi-column SMB involves various variables such as flow rates in multiple sections and valve switching times, rendering its system control highly complex. Unlike traditional control objectives that aim to minimize control output errors, the control objective of SMB is to achieve a specific proportion of substance concentration. Consequently, designing and implementing traditional control theories for SMB is challenging, leading to SMB systems largely being controlled based on simple PLC controls. Moreover, these controls are often applicable only to single or a few-column SMB setups, limiting the effectiveness of high-capacity applications. To achieve effective control of SMB, this study employs an adjustable intelligent fuzzy controller with a structure like an approximate neural network (NN) for SMB control research. Simulation results demonstrate that the intelligent controller effectively achieves desirable control outcomes for the SMB system.