Overview
The study evaluates long-term hourly wind measurements from six Indian sites and combines monthly-hourly wind power density analysis with operational maintenance-window selection. It aims to connect seasonal wind characteristics to lower-opportunity-cost, risk-controlled maintenance planning.
Methods at a glance
- Month–hour wind power density (WPD) heatmaps identify seasonal and diurnal operating patterns.
- Six-hour sliding-window selection targets lower expected energy loss during planned maintenance.
- A probabilistic low-wind constraint and bootstrap confidence intervals characterize maintenance reliability and uncertainty.
- Ramp-rate and volatility diagnostics, plus diurnal wind-regime clustering, describe shorter-term variability.
- Peaks-over-threshold modeling with a generalized Pareto distribution estimates rare wind-energy extremes.
Interpretation
The analysis provides a reproducible maintenance-planning framework for monsoon-influenced wind regimes. It is based on wind-climatology observations rather than direct component-failure measurements, and should not be described as a validated turbine fault-diagnosis system.
Recommended citation
Yadav, H. K., Chakraborty, S., Gupta, M. N., Sarkar, J., & Sarkar, A. (2026). AI calibrated framework for adaptive maintenance of wind turbines. Ocean Engineering, 365, 127313. https://doi.org/10.1016/j.oceaneng.2026.127313