Refine Your Search

Search Results

Viewing 1 to 2 of 2
Technical Paper

Non-Constant Variance - Emission Modeling Methods for Offline Optimization and Calibration of Engine Management Systems

Calibrating the engine control unit to satisfy pollutant and performance objectives can be a challenging task. Due to the large number of variables and their interactive complexities, many firms apply design of experiment methods and modeling techniques to the acquired test data. This establishes a “black box” or “gray box” simulation model that predicts power and emissions as a function of the engine parameters. An offline optimization procedure on the fitted model(s) will identify the engine control strategy that best satisfies pollutant and performance objectives. A review of the literature reveals that the General Linear Modeling method and Neural Network modeling architectures are widely used in the development of “black box” or “gray box” simulation models. While Neural Network methods are “assumption free”, the General Linear Model method is limited to those problems in which the errors, ε, are normally distributed and have constant variance, σ2.
Technical Paper

Statistical Process Control and Design of Experiment Process Improvement Methods for the Powertrain Laboratory

The application of Statistical Process Control and Design of Experiment methods in the research laboratory can lead to significant gains in the Powertrain development process. Empirical methods such as Design of Experiments, Regression, and Neural Network techniques can be applied to help researchers gain better understanding of the cause and effect relationships of emission, alternative fuel source, performance, fuel economy, and engine management system - calibration studies. The use of these empirical modeling techniques along with model based Genetic Algorithm, Gradient, or Constraint based solution search methods will help identify the “process settings” that improve fuel economy, improve performance, and reduce pollutants. Since empirical methods are fundamentally based on the acquired test data, it is vitally important that the laboratory measurements are repeatable, consistent, and void of sources of variance that have a significant effect on the acquired test data.