Artificial intelligence (AI), machine learning (ML), and data science are leading to a promising transformative paradigm. ML, especially deep learning and physics-informed ML, is a valuable toolkit that complements incomplete domain-specific knowledge in conventional experimental and computational methods. ML can provide flexible techniques to facilitate the conceptual development of new robust predictive models for multiphase flows and reactors by finding hidden pattern/information/mechanism in a data set. Due to such emergence, we thereby comprehensively survey, explore, analyze, and discuss key advancements of recent ML applications to hydrodynamics, heat and mass transfer, and reactions in single-phase and multiphase flow systems from different aspects: (1) development of multiphase closure models of drag force, turbulence stresses and heat/mass transfer to improve the accuracy and efficiency of typical CFD simulations; (2) image reconstruction, regime identification, key parameter predictions, and optimization of multiphase flow and transport fields; (3) reaction kinetics modeling (e.g., predictions of reaction networks, kinetic parameters, and species production) and reaction condition optimization. These sections also discuss and analyze the key advantages and weakness of ML for solving the problems in the domain of multiphase flows and reactors. Finally, we summarize the under-solving challenges and opportunities in order to identify future directions that would be useful for the research community. Future development and study of multiphase flows and reactors are envisaged to be accelerated by ML and data science.

Accurately predicting the complex inhomogeneous heat transfer behavior in gas–solid fluidized beds is of fundamental importance. In this work, we constitute an enhanced filtered interphase heat transfer coefficient (IHTC) closure by systematically filtering the dataset from highly resolved three‐dimensional (3D) computational fluid dynamics–discrete element model simulations. Particularly, effects of several potential filtered variable markers on filtered IHTC predictions are examined by statistical analysis. We reveal the formulated filtered IHTC correction closure manifests a systematic dependence on filtered interphase temperature difference as an additional marker. The proposed closure shows good agreement with the filtered fine‐grid simulation data in an a priori analysis. Moreover, the difference of filtered IHTC corrections deduced from 3D Euler–Euler and Euler–Lagrange simulations is quantified. Finally, the comparative analysis between our proposed filtered IHTC formulation and those in literature is implemented. This work holds a potential to facilitate the development of thermal gas–solid flow modeling.

No abstract

This study presents conventional and artificial neural network‐based data‐driven modeling (DDM) methods to model simultaneously the filtered mesoscale drag, heat transfer and reaction rate in gas–particle flows. The dataset used for developing the DDM is filtered from highly resolved simulations closed by our recently formulated microscopic drag and heat transfer coefficients (HTCs). Results reveal that the filtered drag correction is nearly independent of filter size when including the filtered gas phase pressure gradient. We further find that the filtered HTC correction critically depends on the added filtered temperature difference marker while the filtered reaction rate correction shows weak dependence on the additional markers. Moreover, compared with conventional correlations, DDM predictions agree better with filtered resolved data. Comparative analysis is also conducted between existing HTC corrections and our work. Finally, the applicability of conventional and data‐driven models coupled with coarse‐grid computational fluid dynamics simulations for pilot‐scale (reactive) gas–particle flows is validated comprehensively.

This is the second part of a two-part paper. First, the design-optimization system based on the adjoint gradient solution approach as described in Part I is introduced. Several test cases are studied for further validation and demonstration of the methodology and implementation. The base-line adjoint method as applied to realistic 3D configurations is demonstrated in the redesign of the NASA rotor 67 at a near-choke condition, leading to a 1.77% efficiency gain. The proposed adjoint mixing plane is applied to the redesign of a transonic compressor stage (DLR compressor stage) and an IGV-rotor-stator configuration of a Siemens industrial compressor at a single-operating point, both producing measurably positive efficiency gains. An examination on the choice of the operating mass flow condition as the basis for the performance optimization, however, highlights the limitation of the single-point approach for practical applications. For the three-row compressor configuration, a near peak-efficiency point based redesign leads to a measurable reduction in the choke mass flow, while a near-choke point based redesign leads to a significant performance drop in other flow conditions. Subsequently, a parallel multipoint approach is implemented. The results show that a two-point design optimization can produce a consistently better performance over a whole range of mass flow conditions compared with the original design. In the final case, the effectiveness of the present method and system is demonstrated by a redesign applied to a seven-row industrial compressor at the design point, leading to a remarkable 2.4% efficiency gain. Downloaded From: http://asmedigitalcollection.asme.org/ on 12/13/2013 Terms of Use: http://asme.org/terms Journal of Turbomachinery APRIL 2010, Vol. 132 / 021012-9 Downloaded From: http://asmedigitalcollection.asme.org/ on 12/13/2013 Terms of Use: http://asme.org/terms