ETAS ASCMO enables the development of very accurate data-based models. Once created, these models can be used to optimize parameters of real systems such as engine ECUs and also as plant models in different simulation environments (e.g. Simulink or HiL-Systems). Both steady state and dynamic/transient behaviors can be captured.
The effort to create data (measured or simulated) for the real system can be reduced dramatically by using the DoE (Design of Experiments) approach. Traditionally, using DoE to determine the minimal number of necessary measured data points can be difficult – ETAS ASCMO makes it easy. As a next step, very accurate mathematical models can be generated automatically, using modern statistical learning processes (Gaussian processes). Even very complex system behaviors can be described without the need for detailed prior knowledge of the system or special mathematical expertise about the used algorithm. ETAS ASCMO offers a wide range of functionality for a number of use cases, i.e. visualization, model evaluation, as well as various powerful optimization algorithms.
So why use ETAS ASCMO? Because it lets even non-modeling experts generate the industry’s most accurate models significantly faster!
Application areas for ETAS ASCMO:
One of the main use cases is the creation of models describing the behavior of combustion engines on the basis of measurement data. By using DoE (Design of Experiments) techniques to identify the data required to train the model, the effort can be reduced considerably. With easy-to-use ETAS ASCMO this task can now be performed by anyone involved in the engine development process and is no longer restricted to modeling experts only.
ETAS ASCMO also handles desktop emissions testing, allowing the calibrator to evaluate multiple calibration solutions on the desktop in a matter of minutes instead of spending several test days on the emissions dyno.
For users of GT-Power models, supplementing these models with data-based ETAS ASCMO generated models significantly improves the predictive accuracy of emissions simulations, resulting in fewer test runs on the dyno or HiL system for validation.
Other use cases include optimization of hardware components, such as injection valves, sensors, or alternators, transient prediction of emissions and fuel consumption for driving cycles, and complex function parameterization.
Much more information on ETAS ASCMO is available on the ETAS website or feel free to contact us or comment below.