Research On Bubble Distribution Characteristics Of Gas-liquid Two-phase Flow In Direct Contact Heat Transfer Process

Recycling and utilizing waste heat resources in industrial processes such as metallurgy can help reduce primary energy consumption such as coal and reduce carbon dioxide emissions,while Organic Rankine Cycle(ORC)can recover medium and low temperatures below 400 °C in metallurgical processes.The waste heat is used to generate electricity.The use of direct contact heat exchangers in the recovery of waste heat from the ORC system has attracted much attention due to the high heat transfer efficiency,and the mixing characteristics of the hot and cold fluids inside such heat exchangers are closely related to the heat transfer characteristics produced by direct contact..In this work,the experimental research on the direct contact heat transfer process of ORC was carried out,and four kinds of image analysis techniques based on computational homology and uniform design theory were proposed to accurately measure the fluids mixing state in the specific case of ORC direct contact heat exchanger.(1)In the analysis of bubble flow topology based on binary bubble flow image,although the Betti number method successfully realizes the quantitative analysis of the mixing uniformity and pseudo uniformity of two-phase flow or multi-phase flow,the calculation of Betti number depends on the image.The processing of segmentation,therefore,proposes a new idea based on image entropy theory to select the optimal image thresholding algorithm to achieve accurate calculation of Betti number.Then on this basis,the complexity evolution law of the bubble flow pattern and the local region characteristics of the bubble group are further analyzed.That is,on the one hand,a new objective function is proposed as the complexity evolution of the bubble flow pattern in the direct contact heat transfer process of ORC.On the other hand,the Repley's K function is introduced for the first time to study the shape characteristic parameters of the bubble group mixing transient.(2)In the case where the bubble is not completely adhered in the two-dimensional space,the two types of images in which the bubble group is in an aggregated state(or locally uniform in a two-dimensional space)and a dispersed state(or globally uniform in a two-dimensional space)may have The equal Betti number,which will lead to the Betti number method in this special case can not effectively judge the mixing uniformity of the bubble group.Aiming at this problem,by introducing the L2-star deviation in the uniform design theory,a uniform coefficient(information name Uniformity Coefficient,UC for short)for characterizing the uniformity of bubble group in a two-dimensional square or rectangular and circular observation area is proposed.To accurately identify and compare the uniformity of different mixing transients during the flow of a bubble group.(3)In order to further accurately extract the spatiotemporal characteristics of the gas-liquid two-phase flow mixing process in the ORC direct contact heat exchanger,the UC method based on the L2-star deviation is extended to the improved star deviation(ie,the centralization deviation and the rollable Deviation)UC method.The new UC method accurately locates the spatial position of the bubble in the form of coordinates to quantify the uniformity of bubble group mixing.Therefore,the calculation of the new UC no longer depends on the selection and processing of the local deviation function.In addition,by transforming the image data of the bubble group into a digital matrix,it is further proposed to use the centralization deviation and the rollable definition and the uniformity of the mixture to accurately extract the spatiotemporal uniformity characteristics of the multiphase hybrid system from the three-dimensional perspective.(4)For the gray level image of the bubble group,there is a problem that the bubble is difficult to be clearly identified,and the higher statistical analysis technique is introduced to analyze and process the gray level image data of the flow pattern of the recorded bubble group.That is,on the one hand,two types of hypothesis testing tools(ie Kolmogorov-Smirnov test and ?2 test)are introduced to determine whether each pair of bubble group flow images have the same or close gray distribution;on the other hand,a multiple linear regression model is introduced.To determine whether the illumination during the bubble group image acquisition is evenly distributed;then,using the nonlinear method to determine the direction and position of the fixed source,and based on the gray-level bubble group flow image,a new hybrid indicator is proposed for Reveal the evolution of the bubble group flow pattern.In this paper,four types of analysis methods based on bubble group flow images are proposed,which can be used to characterize and analyze the fluid topology and bubble group mixing uniformity in the internal gas-liquid two-phase flow mixing process of ORC direct contact heat exchanger.This study is not only important for the parameter testing and flow pattern identification of the gas-liquid two-phase flow mixing process in the direct contact of the ORC,but also provides a reference for improving the heat transfer performance of the ORC direct contact heat exchanger.