Data-based modeling for stationary behavior

The design of experiments (DoE) element enables you to achieve maximum model accuracy with minimum measurement effort. The design of experiments (DoE) module (ExpeDes) makes it easy to plan the underlying measurements and to propose the position of the measuring points in a methodical and space-filling way. It allows you to adjust the range of parameter variation, the time and effort required for measurement, the order and compression of measuring points, and to incorporate test bench automation requirements. Graphical representations of the points in all dimensions facilitate straightforward validation of your DoE plan.

ASCMO-STATIC makes it easy for less experienced users to apply the advanced modelling methods required for accurate, robust prediction of system behavior. At the same time, it gives experts the freedom to adjust model parameters with a high degree of flexibility. In ASCMO-STATIC, modelling is based on Gaussian process models. In this way, the tool automatically determines the specific mathematical function that best represents the real system behavior.

Creating such a model is simple. After selecting the relevant inputs (influencing variables) and outputs (target values) of the system, you can immediately start training the model without further parameterization. The tool also provides other algorithms for special tasks such as classification and complexity reduction. All models can be exported in a variety of formats, including Simulink, C code, and FMU/FMI.

ASCMO-STATIC offers you a wide range of options for analyzing and visualizing the behavior of the modelled system. One of the standard views is the so-called intersection plot, which allows multidimensional dependencies to be displayed in a straightforward manner. It shows how variations in the inputs affect the output, while also indicating the reliability of the model prediction. In addition, ASCMO-STATIC offers you several other ways to visualize dependencies and influences, such as calibration maps or scatter plots.

ASCMO-STATIC allows you to optimize the input variables for controlling for example an engine by defining various criteria such as minimization/maximization, upper/lower limit, and target values for the output variables. These criteria can either be weighted or the full trade-off curve can be calculated from which you can then select the appropriate compromise. For engine applications, you can perform a global optimization over the entire operating range and obtain an optimum calibration of all input variables in a very short time. Results can be exported in a variety of formats, including DCM, CDFX, Excel, and CSV.

ASCMO-STATIC allows you to use the models as a virtual measuring instrument to artificially generate large amounts of measured data. You can define any grid size for the input variables in terms of step size, number of steps and free break points directly in the tool or by importing Excel lists. The model then calculates the corresponding output for all desired parameter combinations of input variables. Just like real measurements (for example from an engine test bench), this artificial data can be used, analyzed graphically, and exported as a CSV file for further processing in Excel.