Neural-network-based drag force model for polydisperse assemblies of irregular-shaped particles
Not provided.
Engineering↗
Engineering topics
Fan, Liang-Shih
Publications and source records attributed to Fan, Liang-Shih.
Unsupervised Learning Based Interaction Force Model for Nonspherical Particles in Incompressible Flows
This project provides a neural network-based interaction force model for gas-solid flows from low to intermediate Reynolds numbers and concentration, which can be linked to MFiX-DEM. We have constructed a database of the interaction force between the irregular-shaped particles using a spherical harmonic method and the fluid phase based on the particle-resolved direct numerical simulation (PR-DNS) with immersed boundary-based gas kinetic scheme. Unsupervised learning method, i.e., variational auto-encoder (VAE) has been applied to extract the primitive shape factors determining the drag force, lifting forces, and torque. The interaction force model has been trained and validated with a simple but effective multi-layer feed-forward neural network: multi-layer perceptron (MLP), which will be concatenated after the encoder of the previously trained VAE for geometry feature extraction for single, irregular particles. We have trained transpose convolutional neural networks with the PR-DNS data to predict the velocity and pressure gradient of the single particle systems and utilized them to calculate drag force of multi-particle systems. This model can provide high computational efficiency because it does not require collecting multiparticle system data from PR-DNS.
99 GENERAL AND MISCELLANEOUS↗
Biomass Gasification for Chemicals Production Using Chemical Looping Techniques (Final Report)
The Ohio State University (OSU) is investigating the Biomass to Syngas (BTS) chemical looping technology to produce syngas for chemical production applications from biomass under US Department of Energy (DOE) Award #DE-EE0007530. The BTS process aligns with the programmatic area of interest of “Conversion, via biological, thermal, catalytic or chemical means, of acceptable feedstocks into advanced biofuels and/or biobased products including intermediate and end-use products”. Compared to conventional biomass gasification processes, the BTS process eliminates the need for air separation units and tar reforming reactors, which leads to energy efficiency improvement and capital cost reduction. The overall objective is to ascertain the potential of biomass gasification based on the chemical looping technique through mitigation of the possible techno-economic challenges in the steps of scale up for commercialization. The scope of work consists of 1) designing, constructing and operating a 10 kWth commercially scalable sub-pilot BTS system and; 2) completing a comprehensive techno-economic analysis (TEA) of the BTS process using methanol production as an example. Over the course of the project, the project team completed the design, fabrication, and operation of a 10 kWth sub-pilot scale test unit for the BTS process. Corn cob and wood pellets were successfully tested in the unit for high purity syngas generation in extended test campaigns that totals over 200 hours. Syngas purity (H2 and CO) of >70% was achieved with a CH 4 concentration of <6%. The H 2 /CO ratio was greater than 1.8. A comprehensive techno-economic analysis was performed to compare the BTS process and a reference indirectly heated gasification process for methanol synthesis. The result, updated with experimental results for BTS process performance, shows a methanol required selling price (MSP) of $\$ $1.15/gal, compared to $1.28/gal for the reference case.
08 HYDROGEN↗