The paper presents the possibility of using an analytical study of the engine exhaust ignition to evaluate the technical condition of the selected components. Software tools available for the analysis of experimental data commonly use multiple regression model that allows the study of the effects and iterations between model input quantities and one output variable. The use of multi-equation models gives a lot of freedom in the analysis of the measurement results, as it allows the simultaneous analysis of multiple effects and iterations of output variables. It can also be used as a tool for the preparation of experimental material for other advanced diagnostic tools, such as models using neural networks, which allow, in contrast to multi-equation models, the recognition of the state with the number of classes of more than two-class, thus enabling diagnostic reasoning. Assuming that there is a strong correlation and unambiguous nature of the changes in the concentrations of toxic compounds ana-lyzed in the course of the experiment can be seen as symptoms of the technical condition of the engine and with the known values of the output signals (including concentrations of toxic compounds) and their estimates, the values of residuals can be determined, which may indicate the type of damage. The authors show the advantages of using these analytical tools on the example of research conducted on the engine test bench.
Authors
Additional information
- Category
- Publikacja w czasopiśmie
- Type
- artykuły w czasopismach recenzowanych i innych wydawnictwach ciągłych
- Language
- angielski
- Publication year
- 2015
