ECU optimization support tool: ORANGE Navigator & Optimizer
ORANGE-related Q&A
General
- What are the features of ORANGE?
- How many variables can be fitted?
- ORANGE is expensive, so I only want to purchase some of its features.
Experimental Design
- Is there a guideline for what percentage of virtual illuminance corresponds to what model evaluation (R2 value)?
- Do you have any recommendations for experimental design methods?
- Can a global DOE and global model be created?
- In D-optimal planning, if higher-order terms are used, the D-efficiency value (%) decreases no matter how much the planning score is increased. Why is this?
Planned operation
- When integrating with the FAMS system, what is the maximum number of variables that can be measured?
- How much can measurement time be reduced by using experimental design?
- Furthermore, how much can measurement time be reduced by using sequential experimental design?
Collected data
- While a correlation between design variables in a scatter matrix can lead to multicollinearity and is problematic, what happens when a correlation exists between the design variables and the response variable?
- Can the appropriate variables (design variables) be changed for each group (rotational speed, torque, etc.)?
Model
- Do you have any recommendations for modeling methods?
- The R² value can sometimes be negative, but since it's the square of the correlation coefficient, shouldn't it be impossible for it to be negative?
optimization
- Do you have any recommendations for optimization methods?
- Is multi-objective optimization possible even without constraints?