Multi-Objective Optimization Employing Genetic Algorithm for the Torque Converter with Dual-Blade Stator

Paper #:
  • 2015-01-1119

Published:
  • 2015-04-14
DOI:
  • 10.4271/2015-01-1119
Citation:
Wu, G. and Wang, L., "Multi-Objective Optimization Employing Genetic Algorithm for the Torque Converter with Dual-Blade Stator," SAE Technical Paper 2015-01-1119, 2015, doi:10.4271/2015-01-1119.
Affiliated:
Pages:
10
Abstract:
The traditional automotive torque converter (TC) is equipped with a single-blade stator, at the suction side of which there is an apparent boundary layer separation at stalling condition because of its large impending angle. The separation flow behind the suction side of stator blade is found to create large area of low-energy flow which blocks effective flow passage area, produces more energy losses, decreases impeller torque capacity and transmission efficiency. It is found effective to suppress the boundary layer separation by separating the original single-blade stator into a primary and a secondary part. The gap between them guides high-energy flow at the pressurized side of the primary blade to the suction side of the secondary one, which helps to make boundary layer flow stable. As a result, the impeller torque capacity and torque ratio at low-speed ratio increase tremendously at the cost of little drop of maximum efficiency. The original matching property between the torque converter and the engine is changed with the increase of impeller torque capacity of the dual-blade stator TC. Furthermore, the Genetic Algorithm(GA) is utilized to optimize matching relation of blade angles among each runner based on the presented modified one-dimensional flow model, and the Pareto frontier with stalling torque ratio and maximum efficiency of the dual-blade stator TC considered is obtained at the premise of unchanged original torque capacity. The new Pareto frontier is improved significantly compared to that of the single-blade one.
Access
Now
SAE MOBILUS Subscriber? You may already have access.
Buy
Select
Price
List
Download
$27.00
Mail
$27.00
Members save up to 40% off list price.
Share
HTML for Linking to Page
Page URL

Related Items

Training / Education
2007-03-01
Technical Paper / Journal Article
2011-04-12
Training / Education
2017-12-04
Training / Education
2008-08-26
Standard
2012-12-07
Training / Education
2017-06-15
Article
2016-09-06