Solutions
Case Study:
Multistat helps compress Turbine Wheel design time
The Company
The Company is a unit of a Fortune 100 diversified technology and manufacturing corporation serving customers worldwide. It is in a Division which is recognized around the world as one of the leading manufacturers of engine boosting systems for passenger cars and commercial vehicles and is a global supplier to automotive manufacturers.
The Problem: Complex interaction among design elements
The company designs turbocharging products for the automotive industry. These products must deliver top performance and long life at a very competitive price. The basic principal behind turbocharging is fairly simple: air flowing into an engine is compressed in order to allow more air and fuel into the cylinder to increase the power output. However, a turbocharger is a very complex piece of machinery. Not only must the components within the turbocharger itself be precisely coordinated, but the turbocharger and the engine it services must also be exactly matched. If they're not, engine inefficiency and even damage can be the results.
To achieve these critical specifications, the product design process relies on a variety of technologies including optimization. Moreover, the optimization process must accommodate a large number of variables for each component to be truly useful in the turbocharger design.
The principle engineer for core technology was faced with this dilemma in the design of turbine and compressor wheels for the company's core products. Central to the turbocharger are the turbine and compressor wheels, which require mechanical optimization, to balance the turbocharger systems. Without top performing optimization software tools, the design process for these key components is dramatically impacted.
“We have to use an optimization product that can handle many variables - this particular design of a turbine had 45 variables,” he stated. “With conventional products, that many variables would have been a tremendous barrier - slowing the design process down or requiring design changes to accomplish our goals.”
The Solution:
Multistat Visual Optimizer reduced design iterations and cut development time in half »
Maximize
Efficiency
Multistat requires much fewer evaluations to determine each Pareto Optimal point - frequently as few as 3-5 evaluations per Pareto Optimal point.
Scalability
Multistat works effectively with large-scale tasks with 1000-2000 design variables.
Accuracy
Multistat can find Pareto Optimal points within a specified tolerance of the Pareto surface, and can also generate a series of Pareto Optimal points to improve a single criteria.