AbstractThe transport of particulate in a gas flow or pneumatic conveying system 1s widespread in many areas of industry, for example chemical, food processing, cement industries and transportation of pulverised coal in coal-fired power plants. However, a simple and reliable method for monitoring the flow parameters, particularly the mass flow rate, velocity and size of particulate solids in the pipeline, has yet to be developed. This is mainly due to the fact that numerous problems, including insufficient signal generation, particle deposition in sensing vicinity, inhomogeneous particle and velocity profile, can be encountered by flow meters which may affect their readings. Being able to monitor the flow parameters, especially the particulate mass flow rate for example, allows accurate delivery of particulates and hence a better product quality in food processing industrials. In coal-fired power plants, being able to monitor and subsequently control the flow parameters will result in higher combustion efficiency and lower pollutants emission. Furthermore, optimum conveying conditions could also be set, which would result in reduced energy consumption and wear on equipment.
This thesis is concerned with the generation of the Acoustic Emission (AE) from particulate flow and an investigation of the potential of implementing AE for flow parameters, namely the solid feed rate, particle velocity and size monitoring. A series of experiments has been conducted to gather AE signals from a laboratory scale single flow-loop pneumatic conveying system. Initially, AE sensors were attached to two steel meshes, which were placed with a fixed axial distance in the pipeline to study the generation of the AE and subsequently the possibility of using those generated AE to determine particle velocity in the pipeline. Particle velocities measured from this approach were compared with theoretical predictions. The results indicated that more than 90% of the measured particle velocities fall within ±10% of the theoretical particle velocity predicted using the modified Hinkle correlation. Since time alone is measured, no calibration is required.
The generation of AE on five different sensor mounting locations was also studied. The results showed that sensors mounted on all those locations were able to respond to changes in the flow parameters. However, only two optimum sensor locations (mesh and outer bend) were chosen, based on the higher strength and repeatability, for further investigation. The final experimental results indicated that the AE features, namely Root-Mean-Square (RMS) and energy of the AE, are related to the changes in the flow parameters and good correlations were found. Good correlations between the RMS and energy of the AE with the momentum and kinetic energy of the particles, respectively, were also found. Ringdown count of the time domain signal and centroid frequency and energy ratio of the Power Spectral Density (PSD) are independent of variation in the solid feed rate and conveying air velocity. However, they varied significantly with changes in the mean particle size. This clearly marks the potential of the AE method to detect particle size variation inside pipes and hence the performance of the pulverising mill. Overall, all those features of AE have great potential in gas-solid two phase flow parameter monitoring.
|Date of Award||Aug 2008|
- Acoustic emission