Gasification

The MFS team supports the NETL Advanced Gasification program in both multiphase simulation and experimentation programs. The purpose of this research is the development and application of validated models for advanced gasification technology so that new concepts can be developed with less empirical testing, and yielding improved performance.

The objective is to develop accurate simulation models that can be used for the development of advanced gasification technologies. Large-scale, reacting Multiphase Flow with Interphase eXchanges (MFIX)-based models, including MFIX-Discrete Element Model (MFIX-DEM) and MFIX MPIC will be developed for application to gasification processes. These models will employ complex chemical reactions in a representative gasifier geometry – while working to reduce execution speed. Verification, Validation, and Uncertainty Quantification (VV&UQ) are all key aspects of ongoing model development and use. While techniques are well-established for single phase CFD, National Energy Technology Laboratory (NETL) efforts will continue to extend these methodologies to reacting, multiphase CFD. MFIX-based models will be applied to laboratory and industrial scale processes, which is key to demonstrate model viability and utility. Comparison of model predictions to physical data is the basis for model validation.

Well-characterized laboratory-scale data will be generated to aid in understanding physical processes for development of mathematical models and for use code validation. These efforts support the development, operation, and maintenance of laboratory-scale cold-flow experiments for data generation. A range of small-scale, cold flow test units will be created and exercised to generate the data. This approach will help to provide cost-effective, well-characterized data in a timely fashion. The experimental units will cover the range of fixed, moving, bubbling, turbulent, and transport fluidized bed.

Development of Reacting Multiphase Models for Advanced Gasification Processes
  1. Model Development and Optimization: In this work a large-scale, reacting MFIX-DEM model employing complex chemical reactions, a realistic geometry, and millions of reacting particles will be developed. This effort requires that the (1) complex boundary specifications for MFIX-DEM be rigorously tested and stability ensured for parallel computing; (2) the parallel computing routines must be generalized to accommodate thermochemical properties of discrete solids; and (3) code optimization is continuously performed to reduce time-to-solution. Development work will also produce a large-scale, reacting MFIX-MPIC model employing complex chemical reactions and realistic geometry.  Work is underway to achieve a significant speed-up over current MFIX-Two-Fluid Model (MFIX-TFM), MFIX-DEM, and MFIX-MPIC models on new computer chip architectures such as the Intel MIC architecture. A factor of 5-10 speed up over current speed for these models is planned to be demonstrated.
  2. Verification, Validation, and Uncertainty Quantification Tools and Processes for Multiphase Models: Verification, Validation, and Uncertainty Quantification (VV&UQ) are all key aspects of ongoing model development and use. While techniques are well-established for single phase CFD, NETL efforts will continue to extend these methodologies to reacting, multiphase CFD. Verification techniques include good software engineering practices, estimating the error introduced by the numerical methods solving the governing equations, and comparing model predictions to exact solutions if available.  Code validation is the process to determine how accurately the code represents the physical world. This is done by comparing experimental data with model output obtained by simulating the experimental conditions. Simulations will be performed using the MFIX suite of models and compared to well-characterized experimental data produced from NETL experiments.   The effect of uncertainty and variability in model inputs on the simulation output will be characterized for the MFIX suite of codes. The statistical Bayesian Calibration technique will also be developed for multiphase flow simulation and used to assess the model discrepancy due to various sources of errors that are inherent in CFD simulations. By knowing this discrepancy, the Bayesian technique can then be used to calibrate of model parameters, such as kinetic reaction rate parameters, to maximize the accuracy of the predictions. These calibration techniques will be developed and validated for the full MFIX suite of codes and a VVUQ toolset will be released to the MFiX user community.
  3. Physical Sub-Models, Pre- and Post-Processing Tools for Model Enhancement:   Gasification is a complex physical process, including gas and particle flow for a broad range of particle sizes and shapes, gas-phase and solid-phase chemical reactions, interparticle and particle-wall interactions that can lead to attrition and agglomeration, and all modes of heat transfer. This work supports the efforts to identify and/or develop and adopt new and improved physical submodels to accurately model these processes and incorporate them into the MFIX model suite.  This work continues the development and improvement of C3M capabilities and extending applicability of the toolset to other particle-phase reactions and gasifier feedstock.
  4. MFiX Model Application and Validation: This work includes the application of the MFiX Suite to advanced  transport reactor gasification systems to validate the code against pilot and industrial scale data and to demonstrate the utility of MFiX for design, scale up, and optimization.  Extensive simulations of the small-scale model validation units in the MFS laboratory facilities are also underway to aid in model validation.
  5. Experimentation for Model Development and Validation: Well-characterized laboratory-scale data will be generated to aid in understanding physical processes for development of mathematical models and for use code validation. These efforts support the development, operation, and maintenance of laboratory-scale cold-flow experiments for data generation. A range of small-scale, cold flow test units will be created and exercised to generate the data. This approach will help to provide cost-effective, well-characterized data in a timely fashion. The experimental units will cover the range of fixed, moving, bubbling, turbulent, and transport fluidized bed. Materials of construction will allow for visualization of solids and fluid phases and allow for measurements using a variety of flow diagnostic techniques such as Laser Doppler Velocimetry (LDV) and Particle Image Velocimetry (PIV). These methods, combined with the standard temperature, transient pressure, and gas composition measurements will provide a comprehensive data set for validation. These experiments will also provide development platforms for development and validation of novel measurement techniques. For example, accurate measurement of solids circulation rate for circulating fluid bed (CFB) application is critical data for model validation.  The MFS laboratory facilities are used for this work.