Gas-Liquid Two-Phase Flow Characteristics And Intelligent Recognition In Wet Dust Scrubber

Gas-liquid two-phase flow state is an important factor affecting the performance of wet dust scrubber.However,the research on gas-liquid two-phase flow in wet dust scrubber is still not in-depth,and the qualitative understanding of flow state lacks basic analysis data.Because of the lack of this research,it is impossible to form effective guidance for technical innovation of performance enhancement of this kind of dust scrubber in industrial development.When the equipment is running,it is difficult to observe and objectively evaluate the internal operation status of the dust scrubber in real time,and to make timely adjustments in poor operation status,which will affect production safety and purification efficiency.Based on the research platform of strong convection wet dust scrubber,the high frequency pressure signal of the gas phase of the dust scrubber is collected as the original analysis data,and the detailed characteristic parameters closely related to the gas-liquid two-phase flow characteristics of the dust scrubber are extracted by using the theory of wavelet packet analysis and information entropy,recurrence analysis and complexity measure analysis.By comparing with the real-time image signals of gas-liquid two-phase flow,the macroscopic characteristics of gas-liquid two-phase flow,droplets,bubbles and other microscopic characteristics are studied and analyzed from multiple perspectives.Subsequently,according to the gas pressure signal of the dust scrubber,the corresponding statistical analysis and wavelet packet detail information characteristic parameters are selected,and an on-line monitoring system of gas-liquid two-phase flow pattern of the wet dust collector is designed and developed,which provides quantitative basis for real-time control of the operation of the dust collector in the high-efficiency working area and achieves the purpose of real-time strengthening the regulation and purification efficiency.A new index parameter,wavelet energy entropy,is proposed to measure the uniformity of energy distribution of gas pressure signal in different frequency bands.The uniformity of energy distribution of gas pressure in different frequency bands during gas-liquid two-phase flow in dust scrubber is studied.At the same liquid level,the value of wavelet energy entropy decreases with the increase of gas velocity,which indicates that the energy distribution is more uneven in different frequency bands during the gas-liquid two-phase contact process,reflecting the generation anddynamic distribution of different scale collectors in dust scrubber under different conditions;and the value of wavelet energy entropy under different flow patterns has a high discrimination degree,with the distinguishing efficiency reaching 92.5%.It can be used as an effective flow pattern recognition criterion.The multi-scale recurrence characteristics of the gas pressure signal of the dust scrubber are revealed.From a more detailed point of view,the changing process of gas phase pressure signals from low frequency to high frequency under different flow patterns is demonstrated,which further reflects the dynamic evolution of different scale dust collectors with frequency in the process of gas-liquid two-phase contact.Subsequently,it is found that the recurrence quantitative analysis features are not sensitive to the change of low-level gas-liquid resonance flow pattern of dust scrubber but have a strong distinction to the evolution of other flow patterns and are consistent with the results of wavelet energy entropy measurement analysis.It is proved that the detailed pressure information decomposed by wavelet packet can be used as a basis for flow pattern identification.Based on the research and analysis of sequence complexity measurement of gas pressure signals,the recurrence quantitative analysis results of two kinds of complexity indexes of gas with different flow patterns are further verified,showing a better corresponding relationship.This provides an auxiliary criterion for a deeper understanding of gas-liquid two-phase flow characteristics in dust scrubbers.An intelligent discriminant system for gas-liquid two-phase flow pattern of dust scrubber is developed.According to the gas pressure signals of the dust scrubber,the corresponding statistical analysis and wavelet packet detail information characteristic parameters are selected.Then BP neural network model,RBF neural network model and Elman neural network model are established to identify the gas-liquid two-phase flow pattern in the dust scrubber,and the recognition efficiency is 97.2%,98.2% and97.8%,respectively.Meanwhile,neural network fusion recognition rules are established.Based on the software platform of LABVIEW,an intelligent recognition system for gas-liquid two-phase flow pattern of wet dust collector is developed,which realizes real-time monitoring of flow pattern and enhances the process of dust collector running for a long time.