Emergency Data Scheduling Algorithm For Asynchronous And Multi-channel Industrial Wireless Sensor Networks

Industrial Wireless Sensor Networks(IWSNs), which consist of wireless sensors, controllers, actuators, communication relays, and gateways, are a special application of wireless sensor networks in industrial automation. IWSNs, can independently accomplish missions of industrial production, measurement, and control. Industrial wireless networks exhibit advantages of low cost, easy maintenance, and high flexibility. Consequently, IWSNs have become a novel research hotspot in fields of industrial measurement/control and wireless communication, and have wide application prospects. However, a great much of emergency data that have different priorities is broadly existed in industrial applications. These emergency data requires performances of hard real-time and high reliability. If the deadline of emergency data is exceeded, the industrial system will become instable and economic loss will come into being. Furthermore, the personal safety is even endangered. Aimed at the above application backgrounds and requirements, this thesis studies the scheduling methods for the emergency data in industrial scenarios. The contributions of this thesis are summarized as follows.1) This thesis gives an overview of industrial wireless networks and related compositions of the wireless sensor nodes. The advantages of industrial wireless networks are illustrated. In addition, the transmission scheduling problem in wireless sensor networks and the classifications of data in the application of industrial automation are introduced. The current research on the wireless scheduling methods for industrial emergency data are summarized and compared.2) The procedures of the beer filling production line are described. The design and implementation of wired beer filling production line system are illustrated, which include the system overall design, executing devices, the sensor selection and their usage, and the concrete design of the PLC control system, and the drawbacks of the wired PLC control systems are analyzed. A beer filling production line system based on the wireless sensor network is designed and implemented, which includes the system overall design and the introduction of wireless node, access point and gateway device. The designed system utilizes wireless modules to replace original wired sensors. In addition, access points and gateways are also deployed to form an industrial wireless system. The overall system realization diagram is proposed and the scheduling problem of this wireless system is compared with the original wired PLC system.3) Existing TDMA scheduling methods for industrial emergency data under the conditions of asynchronous and multi-channel medium have the problems of high delay, saturated control channel, and large energy consumption. This thesis proposes an Emergency data scheduling algorithm Oriented Asynchronous Multi-channel industrial wireless sensor networks(EOAM). EOAM is initiated by receivers and the problem of saturated control channel during asynchronous multi-channel scheduling is alleviated. A well-designed special channel together with the priority indication method is proposed to provide fast channel switch and real-time transmissions of emergency data. Additionally, the non-urgent data is allowed to occupy channel by a backoff-based mechanism indicated by the priority indication method, which can ensure the utilization of special channel. EOAM is suitable for both unicast and broadcast communication.4) The signal sequence and data features in the beer filling production line are analyzed, which concludes the two emergency data classes in production line. For the data transmission problem of the two emergency data classes, the proposed EOAM algorithm is realized in wireless beer production line monitoring system, and compared with the Distributed Control Algorithm(DCA) in terms of real-time, reliability, and energy consumption. The experiment results show that EOAM outperforms DCA in the network performances, and can meet the transmission requirements of industrial emergency data. Specifically, the minimum transmission delay of unpredictable emergency data is 160 microseconds, the reliability is above 0.95, and the energy consumption is reduced by 12.8%.