In this work,a combination of an acoustic emission (AE) technique and a machine learning (ML) algorithm (Random Forest (RF) and Gradient Boosting Regressor (GBR)) is developed to characterize the particle size distrib...In this work,a combination of an acoustic emission (AE) technique and a machine learning (ML) algorithm (Random Forest (RF) and Gradient Boosting Regressor (GBR)) is developed to characterize the particle size distribution in gas-solid fluidized bed reactors.A theoretical approach to explain the generation of acoustic emission signal in gas-solid flows is presented.An AE signal is generated in gas-solid fluidized beds due to the collision and friction between fluidized particles as well as between particles and the bed inner wall.The generated AE signal is in the form of an elastic wave with frequencies >100 KHz and it propagates through the gas-solid mixture.An inversion algorithm is used to extract the information about the particle size starting from the energy of the AE signal.The advantages of this AE technique are that it is a cheap,sensitive,non-intrusive,radiation-free,suitable for on-line measurements.Combining this AE technique with ML algorithms is beneficial for applications to industrial settings,reducing the cost of signal post-processing.Experiments were conducted in a pseudo-2D flat fluidized bed with four glass bead samples,with sizes ranging from 100 μm to 710 μm.AE signals were recorded with a sampling frequency of 5 MHz.The AE signal post-processing and data preparation for the ML process are explained.For the ML process,the AE frequency,AE energy and particle collision velocity data sets were divided into training (60%),cross-validation (20%) and test sets (20%).Two ensemble ML approaches,namely Random Forest and Gradient Boosting Regressor,are applied to predict particle sizes based on the AE signal features.The combination of these two models results in a coefficient of determination (R2) value greater than 0.9504.展开更多
Gas–solid fluidized beds have historically been employed in a variety of fields owing to the excellent mixing they provide,which can enhance chemical reaction rates and make the control of the reactor temperature eas...Gas–solid fluidized beds have historically been employed in a variety of fields owing to the excellent mixing they provide,which can enhance chemical reaction rates and make the control of the reactor temperature easier than other technologies.Due to this wide application,heuristic knowledge of their functioning has been accumulating over the years.This knowledge,however,is not always backed by a deep understanding of the physical phenomena occurring in such systems.While this heuristic knowledge is sufficient to operate fluidized beds,operation optimization and scale-up are much harder to perform.A range of diagnostic techniques have been applied over the years to draw information about the inner workings of fluidized beds.Among these,x-ray imaging techniques,especially x-ray digital radiography and x-ray computed tomography,stand out for the kind and quality of information they can provide.Their high penetrating power enables visualization of phenomena taking place in the bulk of a fluidized bed,without disturbing the bed hydrodynamics.Furthermore,x rays are generated by a source that can be switched off,making them inherently safer than other imaging techniques relying on radioactive sources,such asγ-ray computed tomography.This work gives an overview of the techniques themselves,of the quantities they can measure,and of some modern applications of gas–solid fluidized beds they have been applied to,such as waste treatment and thermochemical conversion of biomass.Overall,x-ray digital radiography and x-ray computed tomography are better suited for process understanding than for process monitoring and are extremely useful in the study of voidage distribution and macro structures,such as bubbles and jets.展开更多
基金support from Engineering and Physical Sciences Research Council,UK,through the PREMIERE Programme Grant(EP/T000414/1).
文摘In this work,a combination of an acoustic emission (AE) technique and a machine learning (ML) algorithm (Random Forest (RF) and Gradient Boosting Regressor (GBR)) is developed to characterize the particle size distribution in gas-solid fluidized bed reactors.A theoretical approach to explain the generation of acoustic emission signal in gas-solid flows is presented.An AE signal is generated in gas-solid fluidized beds due to the collision and friction between fluidized particles as well as between particles and the bed inner wall.The generated AE signal is in the form of an elastic wave with frequencies >100 KHz and it propagates through the gas-solid mixture.An inversion algorithm is used to extract the information about the particle size starting from the energy of the AE signal.The advantages of this AE technique are that it is a cheap,sensitive,non-intrusive,radiation-free,suitable for on-line measurements.Combining this AE technique with ML algorithms is beneficial for applications to industrial settings,reducing the cost of signal post-processing.Experiments were conducted in a pseudo-2D flat fluidized bed with four glass bead samples,with sizes ranging from 100 μm to 710 μm.AE signals were recorded with a sampling frequency of 5 MHz.The AE signal post-processing and data preparation for the ML process are explained.For the ML process,the AE frequency,AE energy and particle collision velocity data sets were divided into training (60%),cross-validation (20%) and test sets (20%).Two ensemble ML approaches,namely Random Forest and Gradient Boosting Regressor,are applied to predict particle sizes based on the AE signal features.The combination of these two models results in a coefficient of determination (R2) value greater than 0.9504.
文摘Gas–solid fluidized beds have historically been employed in a variety of fields owing to the excellent mixing they provide,which can enhance chemical reaction rates and make the control of the reactor temperature easier than other technologies.Due to this wide application,heuristic knowledge of their functioning has been accumulating over the years.This knowledge,however,is not always backed by a deep understanding of the physical phenomena occurring in such systems.While this heuristic knowledge is sufficient to operate fluidized beds,operation optimization and scale-up are much harder to perform.A range of diagnostic techniques have been applied over the years to draw information about the inner workings of fluidized beds.Among these,x-ray imaging techniques,especially x-ray digital radiography and x-ray computed tomography,stand out for the kind and quality of information they can provide.Their high penetrating power enables visualization of phenomena taking place in the bulk of a fluidized bed,without disturbing the bed hydrodynamics.Furthermore,x rays are generated by a source that can be switched off,making them inherently safer than other imaging techniques relying on radioactive sources,such asγ-ray computed tomography.This work gives an overview of the techniques themselves,of the quantities they can measure,and of some modern applications of gas–solid fluidized beds they have been applied to,such as waste treatment and thermochemical conversion of biomass.Overall,x-ray digital radiography and x-ray computed tomography are better suited for process understanding than for process monitoring and are extremely useful in the study of voidage distribution and macro structures,such as bubbles and jets.
