Monitoring and Predictive System for Thermal Power Plant Using Wireless Sensor Network

Abstract

While the traditional approach to measure the high temperature in thermal plants is to use the wired temperature sensors, wireless measurement of temperature is a better approach. Wireless systems have the capability to reconfigure factory layouts without thinking about the rewiring task required to achieve the same. Wireless monitoring systems are deployed where the wired systems are quite complex and inefficient to be applied. Wireless Sensor Networks (WSN) consists of a central control node and many remote nodes which are connected to it. These remote nodes can allow data acquisition, provide actuating outputs or act as repeaters. It has been proven that WSNs can be used for a variety of applications such as remote sensing, industrial and domestic automation, and in control systems. It becomes cumbersome to measure the temperature of critical and sensitive areas like the furnace in thermal plants where the traditional approach of wired sensors becomes a challenging task. Here WSN plays an important role to accomplish the task to monitor the remote high temperature of the furnace/boiler wirelessly. In this thesis, several experiments have been conducted to carry out the research work. A real-time measuring and monitoring system based on WSN, has been developed for thermal power plant in which the temperature of furnace has been successfully measured, monitored and predicted. Monitored temperature is filtered (Noise free) data. Kalman filtering technique has been applied. Java and Java-Xbee libraries made it possible to remove the hardware part and thus making the system more economical. The system developed is economical, flexible and hardware independent and can be used with any type of thermocouple with cold junction temperature compensation. Thermocouple has directly been connected with remotely placed radio communication module without using any peripheral hardware. newline