Development of videogrammetry as a tool for gas ...

Journal of Environmental Management 203 (2017) 942e949

Contents lists available at ScienceDirect

Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman

Research article

Development of videogrammetry as a tool for gas-particle fluidization research S. Anweiler Faculty of Mechanical Engineering, Department of Environmental Engineering, Opole University of Technology, Poland

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 December 2016 Received in revised form 7 March 2017 Accepted 14 March 2017 Available online 22 March 2017

Many industries use fluidization of solid particles for energy efficiency or environmental friendly process development, and this paper introduces research techniques developed for investigating gas-particle systems At present there is plenty of room for refining gas-particle fluidization process. With the rapidly rising application of mathematical modelling, real time visualization of processes will be widely used for validation of those models in the near future. In presented research, photogrammetry, as a part of close range vision metrology, has been expanded to allow dynamic space and time analysis of the phase concentration distribution inside fluidization devices. A novel videogrammetry method was created with additional stochastic process analysis for detailed frequency and amplitude characteristics. Videogrammetry was used for the assessment of flow regimes, which were held in various types of fluidization apparatuses. Classic bubbling, jet-spouted and fast circulating fluidization processes were explored under the investigation. Videogrammetry is non-invasive flow regime recognition method, which enables detailed research of gas-particle fluidization phenomena. Until now, there were no comparative studies for three different types of fluidization processes with the use of one complex approach. Developed videogrammetric method consists of the flow structure visualization and dynamic image analysis. The analysed feature is the grey level of the image in time domain, and grey level signals were analysed with the use of autocorrelation function and power density function. The results are presented as images, plots and a flow map. Efficiency of the method was tested by comparison of real observed flow structures to the reconstructed flow structures and the recognition accuracy reached 92%. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Fluidization Visualization Flow pattern Dynamic image analysis Videogrammetry

1. Introduction Fluidization is a result of solid particles suspension in a stream of gas. This phenomenon can be observed in many industrial processes, such as coal, biomass and waste combustion or gasification, renewable energy production, chemical, petrochemical and metallurgical processes, granulation and drying (Di Maio et al., 2013). Fluidized bed technology appears to be effective, especially in municipal waste treatment systems (Peng and Lin, 2014). Deliverance of therapeutic agents to the lungs and airways in the pharmaceutical industry is also a part of fluidization technology (Pasquali et al., 2015). Electronic parts has just become serious environmental problem to solve. Therefore another up to date fluidized bed technologies for printed circuit board decomposition were reviewed by (Marques et al., 2013). Significant role of fluidized bed reactors in CO2 mineral sequestration was recently researched

E-mail address: [email protected]. http://dx.doi.org/10.1016/j.jenvman.2017.03.040 0301-4797/© 2017 Elsevier Ltd. All rights reserved.

by (El-Naas et al., 2015). Recent environmental difficulties with the polychlorinated biphenyls (PCBs) in contaminated soils and wastes are solved by destruction using circulating fluidized bed combustion (CFBC) technology, which was studied by (Desai et al., 2007). It has been demonstrated that the reaction efficiency, heat transfer and energy consumption during fluidization process depend on solid mixing and solidegas contact (Yurong et al., 2004). Major advantages of gas-fluidized bed reactors, such as efficient bed-tosurface heat transfer and temperature uniformity, derive from the motion of the particles, largely induced by interactions between voids and the dense phase (Tebianian et al., 2016). Solid mixing and solid-gas contact depend on solid/gas flow structure or solid/gas flow pattern (Li et al., 2014). For the reason it is essential to characterize the two-phase flow behaviours in gas-solid fluidized beds and monitor the fluidization processes for further control and optimization (Sun and Yan, 2016). Intensive research has been conducted to investigate the fluidization behavior experimentally
mez-Barea and Leckner, (Cloete et al., 2013) and numerically (Go

S. Anweiler / Journal of Environmental Management 203 (2017) 942e949

2010), and many models have been developed for optimizing reactor design and bed scale up. This includes even such unusual techniques like stereology (Masiukiewicz and Ulbrich, 2004). Shi et al., (2011) proposed the energy minimization multi-scale model (EMMS) to characterize the meso-scale structure of fluidization process. Still, gas-particle fluidization process is not fully described and there is plenty of room for refinement. Solid flow pattern is an important factor which affects the fluidized bed characteristics, such as the rate of heat and mass transfer, chemical reaction intensity, as well as the particle attrition and internal erosion. Therefore, understanding the mechanism, and knowing the rate of solid mixing are vital to control the product quality and heat flow (Askarishahi et al., 2015). Although computational fluid dynamics (CFD) has become a popular and efficient tool for process design in the past decades, development has been slower for multiphase processes the than for single phase processes. This is due to theoretical complications related to phase interactions and the greater requirements for computational capacity. The large size of industrial facilities further complicates application of multiphase CFD. Fluidized beds are no exception and only recently simulation of large industrial fluidized beds has become feasible. CFD algorithms which couples the macroscopic governing equations for gas phase and the second law of motions for individual particles have been done recently by (Koralkar and Bose, 2016). Computation times in simulations with the most accurate models are still too long however, and modifications and adaptations to the modelling approaches are therefore needed to apply them at commercial scale (Kallio et al., 2015). Numerical approach is expected to give new information on fluidization process and leads to improved on-line reactor diagnostics (Ramirez et al., 2017). With the rapidly rising application of mathematical modelling, visualization of real processes will be, in the near future, widely used for validation of numerical models. Non-intrusive measurement techniques and the current state of knowledge and experience in the characterization and monitoring of gas-solid fluidized beds was fully reviewed by (Sun and Yan, 2016). Due to latest advancements in high accuracy of CCD/CMOS devices, and commonly accessible digital technology of high processor capacity, many research centres take an interest in vision metrology. That causes appearance of wide range of methods and techniques for visualization, and analysis of many various processes. In the presented research, the novel method e photogrammetry, as a part of close range vision metrology, has been expanded to allow dynamic space and time analysis of the phase concentration distribution inside fluidization devices. With addition of stochastic process analysis for detailed frequency and amplitude characteristics - the method of videogrammetry was created. Videogrammetry is non-invasive flow regime recognition method and enables detailed research of fluidization phenomena. Next videogrammetry was used for the assessment of flow regimes, which occur in various types of fluidization apparatuses. In particular e classic bubbling, jet-spouted and fast circulating fluidization processes were taken under the investigation. Until now, there were no comparative studies for the three different types of fluidization processes with the use of one composite approach. Developed videogrammetric method consists of the flow structure visualization and dynamic image analysis. The analysed feature is the grey level of the image in time domain. Grey level signals were analysed with the use of autocorrelation function (ACF) and power density function (PDF). The results were presented as images, plots and a flow map. The efficiency of the method was tested by comparison of real observed flow structure to the reconstructed flow structure. Photogrammetry is the science and art of making precise and reliable measurements from images. While the term ‘photogrammetry’ may evoke the notion of ‘photographic’ images, the

943

technique as such is by no means restricted to those but incorporates all sorts of non-photographic images as well. The term ‘digital’ in ‘digital photogrammetry’ emphasizes the use of digital images, in whatever form, be it from the X-ray, visible, near and far infrared range, or microwave portion of the electromagnetic spectrum. Even ultrasound and electron-microscopic images and the like could be included (Gruen, 1997). In pharmaceutic industry for example dry powder inhalers (DPI) deliver therapeutic agents to the lungs and airways in the form of a powder aerosol. To achieve efficient delivery of these agents to the lungs, studies of fluidization of the bulk of the powder are conducted (Pasquali et al., 2015). In the processes industry segregation of particles inside the reactors have significant impact on the reactions (Gilbertson and Eames, 2001). Segregated fluidization in atmospheric conditions occurs when a bed has a broad particle size range, forming a horizontal interface that separates the lighter (flotsam) and the heavier particles (jetsam) (Gilbertson and Eames, 2001). The bed thus fluidizes non-homogeneously with increase of flow rate (Kumar et al., 2014). Regardless of the rapid development of numerical methods there is still need for experimental validation in process engineering, whether it is a dense phase of fluidized bed (Deen et al., 2007) or a dispersed phase in pneumatic conveying (Borsuk et al., 2016). Although industrial fluidized bed dryers have been used successfully for the drying of wet solid particles for many years, the development of industrial fluidized bed dryers for any particular application is fraught with difficulties such as scaling-up, poor fluidization and non-uniform product quality. Scaling-up is the major problem and there are very few good, reliable theoretical models that can replace the expensive laboratory work and pilotplant trials (Daud, 2008). Due to latest advancements in high accuracy CCD/CMOS devices, and commonly accessible digital technology with high processor capacity, many research centres are taking an interest in vision metrology, creating a wide range of methods and techniques for visualization and analysis. 2. Measurement setup Investigations were carried out in three, so called, twodimensional models of fluidization columns. Two-dimensional models of reactors were described by (Dyakowski and Jaworski, 2000) as an easy and useful, two-phase flow investigation method. Schematic view of such columns are shown in Fig. 1. Twodimensional columns are often used for modelling and visualization of fluidization process in laboratory conditions. They are made of transparent materials, and filled with solid particles, which creates a bed of solid particles. Gas flows upwards, through the bed, causing appearance of gas bubbles, plugs and other turbulences, which are known as two-phase flow patterns, and sometimes are interchangeably called flow structures. The overall view of the measurement stand is presented in Fig. 2 (Anweiler and Ulbrich, 2004). The measurement setup was developed to be flexible, which means that it is modal, so that different types of fluidization can be examined by modification of the side walls. With the use of different shaped side walls the desired kind of flow channel can be created by placing guide plates between two flat, transparent walls. In this way, any type of fluidization apparatus model can be build. Three different fluidic apparatuses were put to the test. First e classical fluidization column e commonly used for bubbling fluidization, which is a rectangular chamber with perforated bottom, is shown in Fig. 1a. Second e jet-spouted fluidization column e which is a conical chamber with axially mounted riser, for the purpose of particle flow stabilization, is presented in Fig. 1b. Third e fast fluidization column e often called circulating fluidized bed,

944

S. Anweiler / Journal of Environmental Management 203 (2017) 942e949

Fig. 1. Schematic view of two-dimensional columns made of transparent materials: rectangular and semicylinrical. Fluidization columns used for two-phase flow visualization: a) classical (bubbling), b) jet-spouted, c) fast (circulating).

which is long vertical, rectangular riser of relatively small equivalent diameter, is shown in Fig. 1c. All dimensions of described columns are given in Table 1. Above columns were filled with spherical, solid particles, with average diameter of 1 mm, 6 mm or 10 mm. The density of solid phase varied between 300 and 1000 kg/m3. Those particles were fluidized with the stream of compressed air, which velocity (wG0) varied between 1,4 and 16,7 m/s, and was dependent on the type of examined apparatus and particles used for filling the bed. Visualization of two-phase flow was possible due to the application of transparent materials, for building the apparatus models and high speed, digital video camera, connected with portable PC, equipped with frame grabber card. Mentioned device allows acquiring high resolution pictures (1024
1024 pixels) with high frequency (up to 1800 frames per second [fps]). The values of such important parameters as heat, mass and momentum transfer are strongly dependent on actual hydrodynamic condition of the bed, which are firmly connected with the flow pattern. The flow pattern of two-phase mixture is found to be a basic factor for estimation of mentioned values, for different types of fluidization process, held in columns of various shape. It was also stated that achievement and confinement of particular flow structure is crucial from the physical conditions point of view, and its estimation is necessary for correct process conducting. 3. Methods The subject of investigation was two-phase, involving gas-solid

Table 1 Dimensions of examined fluidization columns. Fig. 2. Overall view of the measurement setup for two-phase flow visualization: 1-air mains, 2-rotameter, 3-flexible air conduct, 4-fluidization chamber, 5-high speed camera, 6-PC for image acquisition, 7-lighting system, 8-light putter.

Fluidization Column

Height (mm)

Width (mm)

Depth (mm)

Classical Jet-Spouted Fast

1200 1650 2500

300 300 300

30 30 30

S. Anweiler / Journal of Environmental Management 203 (2017) 942e949

945

Fig. 3. Digital dynamic analysis of the grey level value fluctuation of the digital images sequence e investigation is conducted in space and time domains.

mixture flow pattern, and its estimation in the apparatus influence aspect. Geometrical shape of the fluidization column certainly have an influence on the flow structure, therefore spatial phase distribution is also dependent on the construction type of the apparatus. The method used for evaluation and recognition of the two-phase flow pattern was already mentioned e videogrammetry.

In this research work author took advantage of dynamic image analysis technique for recognition of two-phase flow pattern, with the use of spatial analysis. The research was conducted for different types of fluidization columns, and the newly developed method is called videogrammetry. Simply it can be described as measurements taken on video sequence of recorded process. Basically this

Fig. 4. Example images of different fluidization types: a) classical (from the left: bubbles, plugs, turbulent), b) jet-spouted, c) circulating.

946

S. Anweiler / Journal of Environmental Management 203 (2017) 942e949

Fig. 5. Comparison of example results of dynamic image analysis. Fluctuation of grey level value of the image for: A-bubbles, B-transition state between bubbles and plugs, C-plugs, jet-spouted and fast non-uniform fluidization, D-turbulent or fast uniform fluidization and pneumatic transport.

actions provide the receiving of so called “videogramme”. The videogramme can be compared to other time courses of specific activities e for example electrocardiogram of the human heart workflow. Videogrammetry is a digital image processing and analysis method, a part of great domain of vision metrology, evolved from photogrammetry, and aided with stochastic signal analysis in order of time analysis (Fraser, 1997). It consists of several steps, which are based on different ways of analysis. First e getting a sequence of images, representing particular process e digital high speed visualization. Second e preparing the images for analysis by specifying that features of the image, which allow correct flow pattern estimation. The final step involves automatic, digital, dynamic image analysis. In the case of videogrammetry used here, only one property of the image has been distinguished e grey level value of the image, also called brightness level. Grey level of the image is a property, which is a good characteristic from the changeability dynamics point of view. With the use of the software, the grey level of the image for the whole sequence of images can be automatically analysed. The high speed video camera gives monochromatic images with 8-bit colour depth, which gives 256 grey levels. Grey levels can vary between 0 e which means white colour of the image represents 100% gas phase, and 255 e which means black colour of the image represents 100% solid phase. Stochastic analysis of the obtained brightness level fluctuation signal appears as a fourth step. The change course of the grey level is a stochastic and stationary signal, because in long period of time the average value of grey level of the image is constant. In addition, the character of these changes is random, especially for high velocity of two-phase mixture. Random signal analysis can therefore be used for additional characteristic of the process. The basic idea of the method is the digital analysis of the image's grey level value fluctuation dynamics. The analysis is conducted in

the space domain by specified areas, called measuring areas and in the time domain as the second axis. Generally, measuring areas of the image should be placed in the most dynamic brightness level value changes of the image. The idea of dynamic image analysis is presented in Fig. 3. There is one image of fluidization with plotted measuring areas set in three different places. Area A e at the bottom of the central riser, area B e in the middle of the riser, and area C e is placed at the end of the riser. The image is analysed only in those three, small fields. Although whole image can be analysed, this method is based on averaging the grey level in the measuring areas, so the dynamics of the changes for the whole image will be extremely low and should oscillate around a few grey levels. For the whole image analysis, amplitudes of the changes are in range from 5 to 10 grey levels. On the basis of such low fluctuations analysis is impossible. At least 10 level difference in amplitude is needed. Therefore the measuring areas reduction and placement over zones with greater dynamics of grey level changeability is necessary. The analysis of Fig. 3, clearly shows that the fluctuation of grey level value is both time and space dependent. Small peeks on the output function made from area A, indicates small bubbles which forms at the beginning of the riser. The average amplitude in this case is in range from 10 to 20 grey levels. Graph B is particularly interesting and presents important properties. The maximum amplitude of the signal is almost 200 levels (average is about 100 levels) and from the width of the peak it can be easily stated that between 2nd and 4th second of the sequence, big plug of gas has moved through the measuring area B. In area C, the dynamics of the fluctuation of the grey level is small. In fact, there is no flow pattern because most the time this area is empty. Time analysis of the mean grey level value of the specified area of the image, can indicate the flow pattern. Sometimes fluctuations

S. Anweiler / Journal of Environmental Management 203 (2017) 942e949

are more frequent, as it takes place in faster processes, like fast fluidization or pneumatic transport. For such processes the graphs are fuzzy as it can be seen in Fig. 5D. In such cases stochastic process analysis with the use of basic statistic functions, as probability density and autocorrelation is very useful. These functions estimate the process in amplitude or frequency domain, and are useful in search for periodic events. Experienced researcher could detect and eliminate noises and use easy digital filtering of the signal with the use of random signal analysis. On the basis of the fluctuation of the grey level value of the image, such features of the process as: dynamics of the changes in specified measuring areas, the frequency of appearing of given structures, the duration of the structure, among other features. 4. Results The result of digital visualization is a sequence of digital images, acquired with high frequency of 200 frames per second (fps). During the research work, different types of fluidization and variety of flow regimes were obtained. Example images (single frames) of examined fluidization types are shown in Fig. 4. Digital image analysis give the result either as discreet string of averaged grey level values for further signal processing or in a graph form as shown in Fig. 4. Fig. 5 shows the analysis of different types of fluidization flow regimes. These graphs are matched to present the fluctuation of the grey level value of the image

947

sequence in variety of two phase flow patterns. The type of fluctuations vary depending on the type of flow regime. General types of flow patterns can be recognized by grey level fluctuation analysis only, and the recognition is mostly correct. In some cases, recognition of transitory types of flow patterns, require additional signal processing. Unfortunately, some results are similar even for very different types of flow structure, in extremely different constructed reactors. To avoid uncertainty in such case, stochastic analysis takes place, as explained earlier. Additional features appear when the transposition (mapping) function for the grey level fluctuation is. The use of mapping the grey level, with an appropriate transposition function, the grey level value of the image can be easily transformed into void fraction in time and space. The four most commonly occurring flow patterns inside almost every fluidization apparatus are shown in Fig. 5. On the grounds of the shape of plotted signals the discrimination of major two-phase flow regimes seems to be easy, and this is a great advantage of videogrammetry. Recognition of transitory states of the two-phase mixture is an important issue. In such cases, examination of the grey level fluctuation signal as a stochastic process that provides additional helpful information. The full videogrammetric and random signal analysis of the fluidization process, for two types of gas-solid flow structure, commonly appeared in fluidization devices are shown in Fig. 6. This analysis consists of dynamic image analysis in time and space, and stochastic analysis of the grey level fluctuation signal with the use of probability density function and

Fig. 6. Matching the results of dynamic image analysis and following stochastic analysis for two commonly appearing cases, that is, bubbling and plugging fluidization flow patterns as one case, and turbulent flow pattern as second. These two-phase flow regimes take place in all three types of fluidization.

948

S. Anweiler / Journal of Environmental Management 203 (2017) 942e949

for 100% void and additional filtering to remove the noise. In this stage of investigation the problem with lighting often appears. Videogrammetry is a sensible method, and calibration of lights, and the grey level of the background is indispensable. Fortunately, such calibration is not difficult, and has to be done before each series of measurements. Alternatively, it is possible to recalibrate measurement series by using image processing, although this is a time-consuming process. 5. Conclusions

Fig. 7. Example application of simplified transition function (1) on the data from graph 4B and shift of the grey level value to volume fraction of the gas phase in the flowing gas-solid mixture.

autocorrelation function. On the grounds of grey level analysis, especially its changes in time, there is the possibility of direct estimation of phase fraction distribution in desired place inside the apparatus. For the purpose of transforming grey level fluctuation function into the volume fraction, transformation function g(x,y) has to be defined. The simplest way to define such mapping function for grey level fluctuation of the image is to use grey level threshold (T) for binarization of the image. Threshold value is estimated on the basis of image's histogram. For most cases its value was set on 128. In case of T ¼ 128 the simplest transition function g(x,y) looks like Equation (1).

 gðx; yÞ ¼

255 for f ðx; yÞ  T 0 for f ðx; yÞ < T

) object ) bacground

(1)

If appropriate values of the grey level, as a transition function, is assumed, the definition of phase fraction distribution is possible. After this operation the graph B from Fig. 4 will not change in form, V only in values; the Y axis will be transposed to void fraction (RG ) instead of grey level value of the image. The function will look like Fig. 7. The transposing function, in this case, is very simple, just to explain the procedure. Many factors are not taken here into account, such as for example, the need of thresholding the minimum and maximum of the grey levels in order to get appropriate levels

A new measurement technique for two-phase flow of gas-solid mixture has been presented. It is called videogrammetry because of its connection with photogrammetry, which is a part of close range vision metrology (Ganci and Brown, 2000). Videogrammetry allows dynamic space and time analysis of the phase concentration distribution for gas-solid flow inside fluidization apparatus. The basis of this technique is dynamic analysis of digital image. It has been applied for estimation of the two-phase flow pattern, and evaluation of the geometry of the flow chambers influence on phase distribution. Stochastic process analysis has been applied for detailed frequency and amplitude characteristics of the process. Videogrammetry has been used for the assessment of different flow regimes, which were held in various types of fluidization apparatuses. The efficiency of the method was tested by comparison of real observed flow structure to the reconstructed flow structure. This was done by correlation of visualization results with measurement and recognition data. For each flow pattern stochastic functions and parameters were determined in time, space and gas velocity domain. Table 2 shows the method of accuracy analysis with the use of stochastic parameters technique for classical fluidization. The conformity of the digital recognition method is high, and gives 82% reliability. Until now there were no comparative studies for three quite different types of fluidization processes, with the use of one measurement technique. Classical, jet-spouted and fast circulating fluidization processes were taken under investigation. Process was held in so called two-dimensional, laboratory models of fluidization columns with transparent walls. This allowed digital high speed visualization and application of dynamic image analysis. Simple feature of the image (average grey level) was examined. It has been found that average grey level fluctuations can be useful indicator for the flow regime. The final conclusion of conducted research is that each kind of two-phase gas-solid flow pattern has its own and specific grey level

Table 2 Example of the accuracy test for elaborated optical estimation method for two-phase flow pattern e correlation of observed (visualized) flow pattern and identified flow structure with the use of dynamic image analysis e videogrammetry.

S. Anweiler / Journal of Environmental Management 203 (2017) 942e949

change function. With the use of stochastic process analysis it is possible to distinguish even transitory flow patterns. On the grounds of completed experiments and recently published publication reviews, videogrammetry can described as a complex investigation technique for fluidized bed reactors. It is an ideal method for evaluation and recognition of the two-phase flow patterns and estimation of different apparatus' geometry influence on phase distribution. This method allows comparison of various regimes of fluidization. It can also be used for verification of numerical methods, which are widely used for the fluidization process modelling. References Anweiler, S., Ulbrich, R., 2004. Badania struktury przepływu w aparatach fluid
znej _ alnych ro konstrukcji. Chem. Process Eng. 25 (3), 577e582. Askarishahi, M., Salehi, M.S., Godini, H.R., Wozny, G., 2015. CFD study on solids flow pattern and solids mixing characteristics in bubbling fluidized bed: effect of fluidization velocity and bed aspect ratio. Powder Technol. 274, 379e392. Borsuk, G., Pochwala, S., Wydrych, J., 2016. Numerical methods in processes of design and operation in pneumatic conveying systems. In: 22nd International Conference on Engineering Mechanics, Svratka, CZECH REPUBLIC, MAY 09e12. Cloete, S., Zaabout, A., Johansen, S.T., van Sint Annaland, M., Gallucci, F., Amini, S., 2013. The generality of the standard 2D TFM approach in predicting bubbling fluidized bed hydrodynamics. Powder Technol. 235, 735e746. Daud, W.R.W., 2008. Fluidized bed dryersdrecent advances. Adv. Powder Technol. 19 (5), 403e418. Desai, D.L., Anthony, E.J., Wang, J., 2007. A pilot-plant study for destruction of PCBs in contaminated soils using fluidized bed combustion technology. J. Environ. Manag. 84 (3), 299e304. Deen, N.G., Annaland, M.V.S., Van der Hoef, M.A., Kuipers, J.A.M., 2007. Review of discrete particle modeling of fluidized beds. Chem. Eng. Sci. 62 (1), 28e44. Di Maio, F.P., Di Renzo, A., Vivacqua, V., 2013. Extension and validation of the particle segregation model for bubbling gas-fluidized beds of binary mixtures. Chem. Eng. Sci. 97, 139e151. Dyakowski, T., Jaworski, A.J., 2000. Application of Tomographic techniques for imaging fluidized beds. In: Proceedings of 3rd Israeli Conference for Conveying and Handling of Particulate Solids, Dead Sea, Israel,15.1e15.10. El-Naas, M.H., El Gamal, M., Hameedi, S., Mohamed, A.M.O., 2015. CO2 sequestration using accelerated gas-solid carbonation of pre-treated EAF steel-making bag house dust. J. Environ. Manag. 156, 218e224. Fraser, C.S., 1997. Automation in digital close range photogrammetry. In:

949

Proceedings of First Trans Tasmin Surveyors Conference, April, pp. 12e18. Ganci, G., Brown, J., 2000. Developments in non-contact measurement using videogrammetry. In: Proceedings of Boeing Large Scale Optical Metrology Seminar, Long Beach, February. Gilbertson, M.A., Eames, I., 2001. Segregation patterns in gas-fluidized systems. J. Fluid Mech. 433, 347e356.
mez-Barea, A., Leckner, B., 2010. Modeling of biomass gasification in fluidized Go bed. Prog. Energy Combust. Sci. 36 (4), 444e509. Gruen, A., 1997. Fundamentals of videogrammetryda review. Hum. Mov. Sci. 16 (2), 155e187. Kallio, S., Peltola, J., Niemi, T., 2015. Analysis of the time-averaged gasesolid drag force based on data from transient 3D CFD simulations of fluidized beds. Powder Technol. 274, 227e238. Koralkar, N.V., Bose, M., 2016. Performance of drag models for simulation of fluidized beds with Geldart D particles. Adv. Powder Technol. 27 (6), 2377e2398. Kumar, A., Hodgson, P., Gao, W., Das, S., Fabijanic, D., 2014. Investigating the effect of segregation of particles and pressure gradient on the quality of fluidisation at sub-atmospheric pressures. Powder Technol. 254, 137e149. Li, Y., Fan, H., Fan, X., 2014. Identify of flow patterns in bubbling fluidization. Chem. Eng. Sci. 117, 455e464. Marques, A.C., Cabrera, J.M., de Fraga Malfatti, C., 2013. Printed circuit boards: a review on the perspective of sustainability. J. Environ. Manag. 131, 298e306. Masiukiewicz, M., Ulbrich, R., 2004. Estimation of gas-liquid flow structure in a vertical riser with quantitative stereoscopy and digital image processing. Inzynieria Chem. i Proces. 25 (3), 1303e1308. Peng, T.H., Lin, C.L., 2014. Influence of various chlorine additives on the partitioning of heavy metals during low-temperature two-stage fluidized bed incineration. J. Environ. Manag. 146, 362e368. Pasquali, I., Merusi, C., Brambilla, G., Long, E.J., Hargrave, G.K., Versteeg, H.K., 2015. Optical diagnostics study of air flow and powder fluidisation in Nexthaler®dPart I: studies with lactose placebo formulation. Int. J. Pharm. 496, 780e791. Ramirez, E., Finney, C.E., Pannala, S., Daw, C.S., Halow, J., Xiong, Q., 2017. Computational study of the bubbling-to-slugging transition in a laboratory-scale fluidized bed. Chem. Eng. J. 308, 544e556. Shi, Z., Wang, W., Li, J., 2011. A bubble-based EMMS model for gasesolid bubbling fluidization. Chem. Eng. Sci. 66 (22), 5541e5555. Sun, J., Yan, Y., 2016. Non-intrusive measurement and hydrodynamics characterization of gasesolid fluidized beds: a review. Meas. Sci. Technol. 27 (11), 112001. Tebianian, S., Dubrawski, K., Ellis, N., Cocco, R.A., Hays, R., Karri, S.R., GarciaTrinanes, P., 2016. Solids flux measurements via alternate techniques in a gasfluidized bed. Chem. Eng. J. 306, 306e321. Yurong, H., Huilin, L., Qiaoqun, S., Lidan, Y., Yunhua, Z., Gidaspow, D., Bouillard, J., 2004. Hydrodynamics of gasesolid flow around immersed tubes in bubbling fluidized beds. Powder Technol. 145 (2), 88e105.