Multi-scale Analysis Of Typical Flow Patterns' Signal In The Dense Phase Pneumatic Conveying Of Pulverized Coal Via The Gas-solid Interaction

The knowledge of statistical analysis,signal processing and mathematical modelling were comprehensively applied to investigate the influence of carrier gas and water types on the powder flowability,signal and multi-scale gas-solid hydrodynamics characteristics of typical flow patterns in the vertical pipe of dense phase pneumatic conveying of pulverized coal technology.1.From the particle scale,the water type in the pulverized coal and the influence of water addition were explored.The flowability criterion was further developed and the mechanism of water in powders was revealed.The influence of carrier gases on powder flowability was studied.Based on the analysis of particle force,the model was proposed to reveal the influence of carrier gases on the group A and C powders.2.Based on the classical gas-solid transportation phase diagram proposed by Zenz and combined with machine learning approach,a novel high accuracy(95.2%)flow pattern classification and prediction model was proposed,according to the relationship between superficial gas velocity,unit pressure drop and ratio of solid and gas.The advantages and disadvantages of different clustering algorithms and verification methods were discussed in the dense phase pneumatic conveying of pulverized coal.The optimal clustering algorithm and verification method of the system were given.3.The plug flow in the failure of above machine learning model was investigated.Combined with the pipeline electrical capacitance tomography technique,all the plug flow obeyed the same flow pattern change,like packed bed flow?annular flow?pure gas phase flow?annular flow?packed bed flow.What's more,there were five different solid velocity distributions.Based on the fluctuation characteristics of plug flow and system operation parameters,a new plug flow phase diagram was proposed and five different plug flows were successfully predicted.In an addition,static characteristics of the dynamic plug flow were revealed from the bed stress analysis and the reason of flow pattern change in the plug flow was given.4.The cross-sectional distribution characteristics of four typical flow patterns,such as packed bed flow,plug flow,annular flow and less dense phase flow were statistically analyzed.The pressure,particle velocity and solid concentration signal were investigated.The particle energy spectrum characteristics and the principal scale of typical flow patterns were obtained by wavelet analysis.By extracting the principal scale parameters,the coherent structure characteristics of typical flow patterns were identified,and the hydrodynamic mechanism of typical flow patterns was revealed.Combined with wavelet analysis and detrended fluctuation analysis,a new time series of solid concentration model was proposed.5.Based on wavelet analysis and fractal analysis,a multi-scale signal decomposition and coupling method for typical flow pattern signals in the dense phase pneumatic conveying of pulverized coal was proposed.In the frequency domain,according to the fractal characteristics of the conveying signal,the signals were decomposed into microscopic,mesoscopic and macroscopic scale.A multi-scale gas-solid interaction model,including in the gas turbulence,particle-wall friction and particle-particle collision,was proposed to reveal the structure-function relationship between gas-solid interaction and multiscale frequency.Moreover,the corresponding signal values of gas-solid interaction were measured through relative experiments.According to the multi-scale gas-solid interaction,the multi-scale interaction mechanism and the dominant gas-solid interaction of typical flow patterns were revealed.