AI & ML
Nonlinear Fault Tolerant Control of Wind Turbine Pitch Systems: A Comparative Study
arXiv cs.SY0 views
arXiv:2611.09674v1 Announce Type: new Abstract: Electro-hydraulic pitch systems account for about one fifth of wind-turbine failures, motivating the development of resilient control strategies for fault mitigation. The study considers a large set of robust nonlinear control designs of different feedback principles and architectures, and assesses their fault-tolerant capabilities in comparison to each other and against the industrial standard, i.e. the Proportional-Integral (PI) controller. The implementation and tuning of each scheme is based on a high-fidelity simulator to ensure nominal performance, whereas each controller is deployed unchanged on an experimental apparatus that includes a full-scale electro-hydraulic pitch system. The comparative evaluation of the considered fault-tolerant solutions is carried out through several tests that cover nominal operation and twenty-five representative fault scenarios, including sensor, process, and actuator faults. The analysis highlights the trade-off between performance preservation under faults and increased actuator chattering. The experimental results show that introducing a dynamic input-reconstruction module into the design of the nonlinear controllers significantly reduces actuator chatter without compromising their fault-tolerant properties.
Source:https://arxiv.org/abs/2611.09674PreviousA Novel Algorithm to Simulate Conductance-based Neuronal NetworksNextAdaptive PD-like Fault Tolerant Controller for a Class of Nonlinear Systems
Related posts
Trump Backs Nvidia CEO Jensen Huang Amid Chinese Spy Affair Investigation
0Slashdot
JoesProxy – The free network workbench for your AI agent
0Hacker News New
A teen used her school Chromebook and Google AI Mode to plan her suicide
0Hacker News New
The New York Times Doesn't Understand AI's Role in Suppressing Wages
0Hacker News New
Show HN: First-year vs. renewal prices for 532 domain extensions, updated weekly
0Hacker News Show
Show HN: Live counts of AI assistants reading a diaspora business directory
0Hacker News Show