An artificial neural network supported regression model for wear rate
| Autor: | Ivan Argatov, Young Suck Chai |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Statistics::Theory
Artificial neural network Mechanical Engineering Computer Science::Neural and Evolutionary Computation Degrees of freedom (statistics) Regression analysis 02 engineering and technology Surfaces and Interfaces 021001 nanoscience & nanotechnology Regression Surfaces Coatings and Films Matrix (mathematics) Nonlinear system 020303 mechanical engineering & transports 0203 mechanical engineering Mechanics of Materials Applied mathematics 0210 nano-technology Nonlinear regression Contact pressure Mathematics |
| Zdroj: | Tribology International. 138:211-214 |
| ISSN: | 0301-679X |
| DOI: | 10.1016/j.triboint.2019.05.040 |
| Popis: | It is suggested to use an artificial neural network as an element of a nonlinear regression model. Based on the Archard–Kragelsky model for the linear (or thickness) wear rate, which assumes its nonlinear power-law dependence on contact pressure and sliding velocity, the ANN-supported regression model has been developed. We show that the back-propagation algorithm can be adopted to train the ANN along with tuning the regression parameters. The model's applicability is illustrated on an example of rice husk ash reinforced aluminum alloy matrix composites, which is available in the literature. It is shown that the ANN-supported regression model has a superior performance compared to a standard ANN model with, at the same time, a much lower number of degrees of freedom. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect