<abstract xmlns="http://www.w3.org/1999/xhtml"><p> This paper presents possible use of results of exhaust gas composition testing of self - ignition engine for technical state assessment of its charge exchange system under assumption that there is strong correlation between considered structure parameters and output signals in the form of concentration of toxic compounds (ZT) as well as unambiguous character of their changes. Concentration of the analyzed ZT may be hence considered to be symptoms of engine technical state. At given values of the signals and their estimates it is also possible to determine values of residues which may indicate a type of failure. Available tool programs aimed at analysis of experimental data commonly make use of multiple regression model which allows to investigate effects and interaction between model input quantities and one output variable. Application of multi-equation models provides great freedom during analysis of measurement data as it makes it possible to simultaneously analyze effects and interaction of many output variables. It may be also implemented as a tool for preparation of experimental material for other advanced diagnostic tools such as neural networks which, in contrast to multi-equation models, make it possible to recognize a state at multistate classification and - in consequence - to do diagnostic inference. Here , these authors present merits of application of the above mentioned analytical tools on the example of tests conducted on an experimental engine test stand.</p></abstract>

This paper presents possible use of results of exhaust gas composition testing of self - ignition engine for technical state assessment of its charge exchange system under assumption that there is strong correlation between considered structure parameters and output signals in the form of concentration of toxic compounds (ZT) as well as unambiguous character of their changes. Concentration of the analyzed ZT may be hence considered to be symptoms of engine technical state. At given values of the signals and their estimates it is also possible to determine values of residues which may indicate a type of failure. Available tool programs aimed at analysis of experimental data commonly make use of multiple regression model which allows to investigate effects and interaction between model input quantities and one output variable. Application of multi-equation models provides great freedom during analysis of measurement data as it makes it possible to simultaneously analyze effects and interaction of many output variables. It may be also implemented as a tool for preparation of experimental material for other advanced diagnostic tools such as neural networks which, in contrast to multi-equation models, make it possible to recognize a state at multistate classification and - in consequence - to do diagnostic inference. Here , these authors present merits of application of the above mentioned analytical tools on the example of tests conducted on an experimental engine test stand.

Technical State Assessment of Charge Exchange System of Self-Ignition Engine, Based On the Exhaust Gas Composition Testing

Polish Maritime Research

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

This paper presents possible use of results of exhaust gas composition testing of self - ignition engine for technical state assessment of its charge exchange...

Technical State Assessment of Charge Exchange System of Self-Ignition Engine, Based On the Exhaust Gas Composition Testing

Article Category: Special Issue Title: Coastal, Offshore and Ocean Engineering

Online veröffentlicht: 23. Mai 2017

Seitenbereich: 203 - 212

DOI: https://doi.org/10.1515/pomr-2017-0040

Schlüsselwörter
diagnostic model, self-ignition engine, exhaust gas components, artificial neural networks

© 2017

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Technical State Assessment of Charge Exchange System of Self-Ignition Engine, Based On the Exhaust Gas Composition Testing

Article Category: Special Issue Title: Coastal, Offshore and Ocean Engineering

Online veröffentlicht: 23. Mai 2017

Seitenbereich: 203 - 212

DOI: https://doi.org/10.1515/pomr-2017-0040

Schlüsselwörterdiagnostic model, self-ignition engine, exhaust gas components, artificial neural networks

© 2017

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Schlüsselwörter
diagnostic model, self-ignition engine, exhaust gas components, artificial neural networks