Tribological Application of Artificial Neural Networks

Satellites and other spacecraft with long mission durations require reliable bearings for as much as 15 years of continuous operation.  However, it is not feasible to test a bearing for this length of time, so accelerated-life testing must be performed to qualify the bearing.  Accelerated-life testing is inherently problematic, because it is often difficult to increase test parameters such as load and rotational speed significantly without exceeding the specifications of the bearing.

We have been researching the application of artificial neural network (ANN) technology to analysis of tribological behavior.  The report "Neural Network Models of Simple Mechanical Systems Illustrating the Feasibility of Accelerated Life Testing", NASA TM-107108 by Steven P. Jones, Ralph Jansen, and Robert L. Fusaro explores the design and training of various networks for modeling wear in three tribological tests.  The ultimate goal is to ascertain whether ANN models can be used for accelerated-life testing.

These models were constructed from "Input Layer Dampened Recurrent Networks" with supervised backpropagation training.  A number of architectures were investigated systematically before arriving at this design.  The GRNN was actually very accurate at modeling the training set according to the R^2 coefficient, but it could not generalize well.

You can access the abstract and ordering information for NASA TM-107108. (The report is not available for download from the Glenn report server.)
 

Neural Nets Modeling Tribological Apparatus

NASA logo Rub Shoe Pin-On-Disk
Exposed Data
(Learned)		 rub shoe learning results pin-on-disk learning results
Unexposed Data
(Generalized)		 rub shoe guessing results pin-on-disk guessing results

 

R^2 Coefficients For Various Design Architectures For Various Models Developed

Architecutral Design Rub Shoe Model Pin-on-Disk Model Four-Ball Model
3 Layer Backpropagation 0.76 0.89 0.67
4 Layer Backpropagation 0.78 0.89 0.69
5 Layer Backpropagation N/A 0.88 0.66
Input Layer Dampened Recurrent Network 0.84 0.93 0.92
Hidden Layer Dampened Recurrent Network 0.82 0.91 0.82
Output Layer Dampened Recurrent Network 0.77 0.89 0.60
2 Hidden Layers with different activation functions 0.63 0.90 0.69
3 Hidden Layers with different activation functions 0.74 0.89 0.67
2 Hidden Layers with different activation function and jump connection 0.76 0.90 0.67
3 Layers with jump connections 0.77 0.84 0.61
4 Layers with jump connections 0.76 0.84 0.54
5 Layers with jump connections 0.77 0.84 0.54
General Regression 0.97 0.90 0.71

 

Contribution Strengths for Input Variables in Pin-On-Disk Model

Input Variable Contribution Strength
Load 5.4
Speed 7.1
Viscosity 5.3
Sliding Distance 6.0
Friction Coefficient 5.0
Temperature 4.3

From an ANN standpoint, the problem with the bearing data is that they are few.  In bearing tests it is impractical to collect huge amounts of data.  This makes it very difficult for the ANN to model such a sophisticated relationship - bearing systems are extremely sensitive to a number of inputs, including load, speed, sliding distance, lubricant, temperature, atmosphere, manufacture, etc.  Our sets have consisted of about 50 data, and that's quite a large sample, to give you some idea.  We take some of the above input parameters and calculate a single output - wear volume.

We had two ideas for accelerated-life testing.  The first was to perform simple extrapolation, but these nets seem to be quite bad at this (worse than other statistical methods tried).  The other idea was to focus on detecting anomalies in the experimental data which might indicate long-term failure.  For this we might use an unsupervised network (Kohonen, to be exact) to detect anomalous behavior.  This is our current area of exploration.

You also maybe interested in the comp.ai.neural-nets newsgroup and its FAQ (Frequently Asked Questions document).  Unfortunately both the newsgroup and FAQ are ill-suited to beginners.

Space Mechanisms Project Page | Mechanical Components Branch

Some of the documents in this page are provided in PDF format and require Adobe’s Acrobat Reader.
Download the Adobe plug-in from www.adobe.com.

Curator & responsible official: Fred.B.Oswald_AT_nasa.gov.
NASA Web Privacy Policy and Important Notices
Last modified 28 Nov 2006