Advanced DoE techniques with innovative modeling algorithms
Today it is hard to imagine powertrain development without Design of Experiments (DoE). The main challenge, though, is to make this powerful technique easily applicable for all calibration engineers which are typically not familiar with the mathematical background of this approach. At the same time, powertrain development poses particular challenges requiring for specific adaptions of the generic DoE methodology.
A new software tool featuring DoE based application processes have been developed at FEV and is available since Q1/2015. xCAL makes the powerful DoE technique easily applicable and addresses the particular challenges of automotive development that require tailored adaptations of the generic DoE methodology. Special emphasis was placed on developing highly reliable and rapid modeling algorithms which are unique to the market. The today’s state-of-the-art global modeling techniques based on Gaussian processes have been adapted to address also the specific characteristics of engine and powertrain modeling. This approach, combined with intuitive visualization and user guidance, enables the engineer to quickly investigate and optimize for example the engine’s behavior.
>> Innovative DoE methods in combination with unique visualization options and intuitive user guidance. <<
Minimized development time, optimized results
Special attention has been placed on guiding the user through a dedicated, yet very flexible, workflow. The clear structure supported by a comfortable handling and rich visualization options facilitates the application of DoE based development processes even for non-experts. xCAL calculates optimized test plans based on statistical considerations, minimizing test bed usage or simulation effort while, at the same time, enabling high model quality. Based upon measured or simulated data, xCAL creates models that capture even complex system behavior and form a perfect basis for the subsequent optimization. The excellent results enable to exploit the full potential of modern powertrains allowing to meet the rising demands of the automotive industry.
- SIGNIFICANT Reduction in testing/simulation effort
- Reliable prediction of complex behavior from local and global data-driven models
- Best-in-class modeling algorithms
- Powerful numerical optimization routines
- Unique visualization options for quick and intuitive analysis
- Optimized user guidance through TOPEXPERT’s workflow concept
- xCAL-DoE-20202.5 MB