Abstract
Monitoring the condition of rotating machinery, especially planetary gearboxes, is a challenging problem. In most of the available approaches, diagnostic procedures are related to advanced signal pre-processing/feature extraction methods or advanced data (features) analysis by using artificial intelligence. In this paper, the second approach is explored, so an application of decision trees for the classification of spectral-based 15D vectors of diagnostic data is proposed. The novelty of this paper is that by a combination of spectral analysis and the application of decision trees to a set of spectral features, we are able to take advantage of the multidimensionality of diagnostic data and classify/recognize the gearbox condition almost faultlessly even in non-stationary operating conditions. The diagnostics of time-varying systems are a complicated issue due to time-varying probability densities estimated for features. Using multidimensional data instead of an aggregated 1D feature, it is possible to improve the efficiency of diagnostics. It can be underlined that in comparison to previous work related to the same data, where the aggregated 1D variable was used, the efficiency of the proposed approach is around 99% (ca. 19% better). We tested several algorithms: classification and regression trees with the Gini index and entropy, as well as the random tree. We compare the obtained results with the K-nearest neighbors classification algorithm and meta-classifiers, namely: random forest and AdaBoost. As a result, we created the decision tree model with 99.74% classification accuracy on the test dataset.
Highlights
The planetary gearbox (PG) is a crucial element of many mechanical systems
Requirements related to the high power to be transmitted, high transmission ratio, and small dimensions/weight cause planetary gearboxes to be complex, and they are utilized in advanced systems, so there is a need to ensure their reliability during operation
We propose an efficient method of classifying planetary gearboxes’ diagnostic data based on decision trees and adaptive learning algorithms like classification and regression trees (CART)
Summary
The planetary gearbox (PG) is a crucial element of many mechanical systems. Requirements related to the high power to be transmitted, high transmission ratio, and small dimensions/weight cause planetary gearboxes to be complex, and they are utilized in advanced systems (helicopters, wind turbines, mining machines, etc.), so there is a need to ensure their reliability during operation.Damage detection in PGs is crucial, and one can find many interesting techniques developed in recent years.A general layout for condition monitoring is to pre-process the signal, extract features, and classify them, often with the use of advanced data analysis techniques. The planetary gearbox (PG) is a crucial element of many mechanical systems. Requirements related to the high power to be transmitted, high transmission ratio, and small dimensions/weight cause planetary gearboxes to be complex, and they are utilized in advanced systems (helicopters, wind turbines, mining machines, etc.), so there is a need to ensure their reliability during operation. A general layout for condition monitoring is to pre-process the signal, extract features, and classify them, often with the use of advanced data analysis techniques. Most of the diagnostic procedures can be divided into two groups: the first one focuses on an advanced signal pre-processing/feature. Sensors 2020, 20, 5979 extraction and uses a simple decision scheme, and the second one is related to a set of simple features and advanced data (features) analysis by using artificial intelligence.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
