A Knowledge-Based Real-Time Vision System for Monitoring the Inside of a Fluid Bed Heat Exchanger Chamber

Abstract

In this paper, we present a vision system with a special camera for knowledge-based real-time monitoring of the inside of a fluid bed heat exchanger (FBHE) chamber in a thermal power plant. With the proposed system, it is possible to monitor the internal flux condition and analyze the inner temperature of a chamber. Due to the fact that the mixture of coal and sand inside a chamber flows by very quickly, there is an immense amount of smoke and dust, which make it difficult to capture images and analyze an existing system. An adaptive average method is proposed here to observe the background internal environment of an FBHE chamber. The experimental results show that real-time monitoring is possible, even under hot and dusty conditions. Preliminary experimental results confirm expectations about the possibility and effectiveness of the developed device for commercialized real-time monitoring systems. They demonstrate that a single camera with embedded image processing software can concurrently analyze the degree of fluidization of a mixture and the temperature of the chamber inside, even in extremely harsh and hazardous conditions. We aim to eventually develop an image analysis system that combines image processing and knowledge engineering techniques.