PreprintArticleVersion 1Preserved in Portico This version is not peer-reviewed
Comparative Evaluation of the Models for Predicting the Rela-Tionship Between Electrical and Chemical Properties of Engine Lubricants: An Exact Monitoring Approach
Version 1
: Received: 13 March 2023 / Approved: 17 March 2023 / Online: 17 March 2023 (02:17:24 CET)
How to cite:
Pourramezan, M.; Rohani, A.; Abbaspour-Fard, M.H. Comparative Evaluation of the Models for Predicting the Rela-Tionship Between Electrical and Chemical Properties of Engine Lubricants: An Exact Monitoring Approach. Preprints2023, 2023030313. https://doi.org/10.20944/preprints202303.0313.v1
Pourramezan, M.; Rohani, A.; Abbaspour-Fard, M.H. Comparative Evaluation of the Models for Predicting the Rela-Tionship Between Electrical and Chemical Properties of Engine Lubricants: An Exact Monitoring Approach. Preprints 2023, 2023030313. https://doi.org/10.20944/preprints202303.0313.v1
Pourramezan, M.; Rohani, A.; Abbaspour-Fard, M.H. Comparative Evaluation of the Models for Predicting the Rela-Tionship Between Electrical and Chemical Properties of Engine Lubricants: An Exact Monitoring Approach. Preprints2023, 2023030313. https://doi.org/10.20944/preprints202303.0313.v1
APA Style
Pourramezan, M., Rohani, A., & Abbaspour-Fard, M.H. (2023). Comparative Evaluation of the Models for Predicting the Rela-Tionship Between Electrical and Chemical Properties of Engine Lubricants: An Exact Monitoring Approach. Preprints. https://doi.org/10.20944/preprints202303.0313.v1
Chicago/Turabian Style
Pourramezan, M., Abbas Rohani and Mohammad Hossein Abbaspour-Fard. 2023 "Comparative Evaluation of the Models for Predicting the Rela-Tionship Between Electrical and Chemical Properties of Engine Lubricants: An Exact Monitoring Approach" Preprints. https://doi.org/10.20944/preprints202303.0313.v1
Abstract
Condition monitoring is a primary principle for predictive maintenance. In this regard, several techniques have been employed, each of which has its strengths and weaknesses. In this work, the relationships between the electrical (ε', ε", tan δ) and chemical (Fe, Pb, Cu, Cr, Al, Si, Zn) properties of engine lubricants were investigated using soft computing models. The models' performance was evaluated considering some criteria, including RMSE and MAPE. The RBF model was determined as the best one to predict chemical properties (Fe, Pb, Cu, Cr, Al, Si, Zn) through electrical indexes (ε', ε", tan δ). The RBF’s parameters (e.g., hidden size and training algorithm) were then optimized for optimum performance. For instance, in the RBF model implemented to predict Al, the best hidden size and training algorithmic were 15 and 'trainlm', respectively. Finally, by sensitivity analysis it was observed that eliminating any of the inputs does not improve the performance of the RBF model performance. In general, the results of the current research showed that the electrical properties (e.g., ε', ε", tan δ) have the capability of estimating the contaminants in the lubricants which can improve the existing monitoring methods.
Keywords
Soft computing models; Monitoring engine lubricant condition; Dielectric; Chemical properties
Subject
Engineering, Automotive Engineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
