The Application of Acoustic Emission to Monitor Pulverised Fuel Flows

Abstract

There are a large number of industrial processes involving the transport of pneumatically conveyed solid including mineral processing, electrical power generation, steel and cement production. For coal-fired power plant, in particular, pulverised fuel (pf) is fed by pneumatic means where coal particles are transported by the primary air from each mill directly into furnace. The distribution of coal particles to each burner bank is normally split mechanically from larger pipelines into a smaller network of pipes connected to each of the burners. Despite the use of matched outlet pipes and riffle devices within the splitters, uneven distribution of the pulverised coal inevitably occurs. Incomplete combustion due to the non-uniform distribution of the pulverised coal between the burner’s feed pipes leads to a reduction in boiler efficiency. This also directly leads to an increase in slagging and fouling in the burner and increased NOx emission from the burner. Measuring can solve this problem and subsequently controlling the mass flow in each burner feed pipe and then adjusting the excess air to operate near the minimum. Over the past ten years or so, there has been increased interest in applying acoustic emission (AE) detection methods for process condition monitoring. The European Working Group for Acoustic Emission (EWGAE), 1985, defines AE as ‘the transient elastic waves resulting from local internal micro displacements in a material’. The American National Standards Institute defines AE as ‘the class of phenomena whereby transient elastic waves are generated by a rapid release of energy from a localised source or sources within a material, or the transient elastic waves so generated’. Therefore, in principle, any impulsive and energy release mechanism within a solid or on its surface, such as plastic deformation, impact, cracking, turbulence, combustion, and fluid disturbances, is capable of generating. Since these mechanisms can be associated with the degradation occurring within a particular process, it follows that AE has great potential in condition monitoring, for example, monitoring of tool wear, corrosion and process monitoring of the pneumatically conveyed solid. Unlike most of the other techniques, AE sensors are non-invasive so that their interruption with the flow within the pipe can be totally avoided. Furthermore, the frequency responses of AE sensors are normally very high (in the order of a Mega Hertz) so that they are immune to low-frequency environmental noises. The use of AE detection techniques is appropriate in this project since the frictional contacts between the flowing particles and the inner wall of the conveying pipe can effectively generate ‘elastic waves’ which propagate through the inner pipe wall and be detected by an AE sensor attached to the outer pipe wall. Consequently, the current research work aims to demonstrate the use of an AE to monitor the flow of particles in a conveying pipe. Preliminary results indicate that AE is generated and is highly repeatable for both variations in velocity for a fixed particle size and also for variations in mass flow rate at a fixed velocity.