New method to design large-scale high-recirculation airlift reactors

Abstract

Highlights: Fast and efficient methods are presented for designing High-Recirculation Airlift-Reactors.  Two-phase flow of water and air is modelled.  Improvements are made by refining steady-state models of fluid dynamics.  Solutions are produced by varying superficial gas velocity.  Solutions are produced by simultaneously varying down-comer and riser diameters.ABSTRACT: High-Recirculation Airlift-Reactors (HRARs) are efficient. They use air to push a mixture around the reactor and to provide oxygen for biological micro-organisms. HRARs have been designed using programs created in the 1970s by ICI. They are limited in functionality and accuracy and need significant expert interpretation. This paper describes the creation of new structured methods that are faster and more efficient and replace the ICI methods. Models used for the calculations and the way in which they were implemented are described. The paper then describes improvements to design by analysing and improving the steady-state models of fluid dynamics within a HRAR. The models also delivered information about reactor design, especially which parameters to modify to reach a steady state result. Two-phase flow of water and air is modelled and applied to HRARs. Tests evaluated the models and some results are included. They show that varying superficial gas velocity or simultaneously varying down comer and riser diameters can create a steady-state solution. Mathematical models are used to forecast steady-state situations in the HRAR for specific gas or liquid flow rates and for various constructions. Experimental relationships forecast mass transfers between gas and liquid phases and they predict flow.

Combining Multiple Criteria Decision Making with Vector Manipulation to Decide on the Direction for a Powered Wheelchair

“…Future (and some current) research is investigating ways for the system to learn and how to mix AI tools [34,35] to use different AI tools to best effect. Specifically, Fuzzy networks [36][37][38][39], Rule Based Systems [40][41][42], Blackboard Systems [43][44][45], Expert Systems [46][47][48] and artificial neural networks [49,50].…”

Initial Results from Using Preference Ranking Organization Methods for Enrichment of Evaluations to Help Steer a Powered Wheelchair

“…Current and emerging research is examining ways to use different AI tools [30][31][32][33][34][35][36] to best effect. Specifically, Fuzzy networks [37,38], Rule Based Systems [33,34], Blackboard Systems [39][40][41], Expert Systems [42,43] and artificial neural networks [44][45].…”

Improving Human-Machine Interaction for a Powered Wheelchair Driver by Using Variable-Switches and Sensors that Reduce Wheelchair-Veer

“…Problems occur when the wheelchair is driven along sloping ground because casters can swivel in the direction of the slope and gravity can then make the wheelchair turn ("veer"). Several types of system were investigated that could assist a wheelchair user: -modelling of the environment [36][37][38][39], -predicting shape of the terrain in front of the sensor [40,41], -Artificial Intelligence [42][43][44][45][46][47][48][49][50][51][52], -Machine Intelligence [53][54][55][56][57][58][59], -and Path Planning [9,14,25,59,60].…”