XTRACTIS FOR predictive maintenance

Prediction of the Degradation of a Naval Propulsion Unit

Benchmark vs. Random Forests, Boosted Trees & Neural Networks

How to successfully predict the functional degradation of a naval propulsion unit compressor, given the hyper-complexity of the phenomenon (strongly nonlinear behavior)?

Goals & benefits

Allow business experts and maintenance managers to understand the causal relationships between some turbine parameters and its future state of degradation.

Find the truly influential parameters for assessing the state of degradation and thus reduce measurement and maintenance costs.

Carry out maintenance actions specific to each turbine upstream in order to avoid critical damage, thanks to rapid and systematic diagnostics, while justifying each intervention.

XTRACTIS RESULTS

We get a Predictive Model that is:

Intelligible.

A Decision System composed of 428 unchained gradual rules (aggregated into 36 disjunctive fuzzy rules). Each rule uses some of the 12 variables that XTRACTIS identified as significant (out of 14 Potential Predictors that are turbine parameters).

Robust.

Excellent Real Performance on the Test dataset.

Efficient & Operational.

Running in real-time up to 70,000 predictions per second (i7 @2.5GHz with 8 physical cores), offline or online (API).

full use case

Use Case 08/2022 (v4.2)

Results by
XTRACTIS® GENERATE 12.2.43064 (08/2022)

DOCUMENT CONTENTS

  1. Problem Definition
  2. Xtractis Solution
  3. Top-Model Induction
  4. Explained Predictions for 2 cases
  5. Top-Models Benchmark