Gearbox pump failure prognostics in offshore wind turbine by an integrated data-driven model

Abstract

Offshore wind turbines face substantial challenges in operation and maintenance due to the harsh marine environment and remote locations. Predictive maintenance, encompassing fault diagnostics and failure prognostics, is a promising maintenance strategy to address these challenges. To contribute to this strategy, an integrated data-driven model is developed for probabilistic failure prognostics at the component level. The remaining useful life of a gearbox pump in an offshore wind turbine is predicted accurately based on supervisory control and data acquisition data. In this approach, light gradient boosting machines are tuned to model normal temperatures. The gated recurrent unit outperforms other neural networks and is selected to process temperature residuals with a Bayesian neural network. Results show that the prediction at the 50% percentile precedes the true failure time by 3.83 h. Moreover, there is 97.5% confidence that the true failure time falls within around ± 5.3 h of the predicted time. Furthermore, the earliest alarm is issued at the 2.5% percentile, precisely 9.17 h prior to the true failure time. This study demonstrates the effectiveness of supervised learning and normal behavior modeling for probabilistic failure prognostics of offshore wind turbine components. • Integrates LightGBM for normal behavior modeling and GRU with BNN for RUL prediction. • Provides a metric for LightGBM to minimize features while maintaining performance. • Rolling window and PCA boost prediction accuracy by reducing noise and seasonality. • BNN works with GRU to realize probabilistic RUL prediction and improve accuracy.

Topics & Concepts