Model-Based Calibration Toolbox

 

Model-Based Calibration Toolbox

Model and calibrate complex powertrain systems

Calibration Applications

Model and calibrate powertrain components to optimize performance.

Magnet and windings shown for a permanent magnet e-motor.
Cutaway view of an electric vehicle battery pack.
Mazda’s SKYACTIV-D engine.

Calibration Workflow

Model-based calibration workflow diagram showing design of experiments leading into physical testing and high-fidelity simulation, followed by data modeling, calibration generation, and, finally, the results: an accurate component model and efficient calibration.
Design Editor window showing the Pairwise Projections, Design Table, 3D Constraints, and 4D Design Projection.

Design of Experiments

Characterize your system’s response with an efficient test plan that uses proven experimental designs, including space-filling designs, optimal designs, and classical designs.

Model Fitting app screenshot showing the response surface and diagnostic statistics for a Gaussian process model.

Analyze and Fit Data to Statistical Models

Accurately represent your data with statistical models, including Gaussian process models, radial basis functions, and user-defined nonlinear models by using the MBC Model Fitting app.

MBC Optimization app screenshot showing optimization results, including a results contour, objective contours, a results surface, and a constraint

Generate Optimal Calibrations

Maximize component efficiency across the entire operating range by generating optimal steady-state or transient calibrations for filling lookup tables using the MBC Optimization app.

Model-Based Calibration Toolbox FAQs

Model-Based Calibration Toolbox provides apps and design tools for modeling and calibrating complex nonlinear systems, including powertrain systems such as engines, electric machines, batteries, pumps, and fans, as well as non-automotive systems like jet engines, marine hydrofoils, and drilling equipment.

Model‑Based Calibration (MBC) supports two main workflows: lookup table optimization and lookup table estimation. Lookup table optimization uses Design of Experiments, statistical model fitting, and optimization tools to generate optimal lookup tables. Lookup table estimation uses measured data to estimate lookup table parameters in a Simulink model.

The toolbox supports Gaussian process models, radial basis functions, polynomials and user-defined nonlinear models using the MBC Model Fitting app.

The MBC Optimization app maximizes component efficiency across the entire operating range by generating optimal steady-state calibrations for filling lookup tables. It also can be used to find optimal lookup table values that minimize the deviation from measured data.

Yes, models created with the toolbox can be exported to Simulink for control design, sensitivity analysis, hardware-in-the-loop testing, and other simulation activities, while calibration lookup tables can be exported to ETAS INCA and ATI Vision.

The toolbox enables you to define optimal test plans that require fewer test points, automatically fit statistical models, and generate calibrations for high-degree-of-freedom systems that would require exhaustive testing using traditional methods.

While the toolbox is generic in nature and can be used for any process, there are several examples and dedicated workflows that support automotive powertrain applications including motor control calibration, battery characterization, and engine control calibration.

Yes, you can automate the model fitting and calibration process using the toolbox apps or MATLAB functions.

No, the toolbox is designed for people with domain expertise who aren’t necessarily statisticians or programmers. You can use the apps interactively to specify the kind of problem you’re trying to solve, then let the tools do the statistical analysis for you and provide recommendations.

Interested in Model-Based Calibration Toolbox?