Turbine Engine Rotor Bow Mitigation via Adaptive Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Gas turbine engines experience bowed rotor conditions due to asymmetric heat release after shutdown, leading to rotor distortion, increased unbalance, and potential engine damage or performance loss, with existing solutions either being inefficient or requiring lengthy motoring periods that disrupt operational schedules.
Innovation Solution
A method utilizing a machine learning algorithm to generate an adjusted ground operation schedule based on engine operation, environmental, and location parameters, which includes rotor speed, acceleration, and duration of rotation to mitigate bowed rotor conditions, allowing for more efficient and rapid alleviation of thermal gradients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the rotor assembly is motored for long periods to reduce thermal gradient and alleviate bowed rotor condition, then the rotor bowing is reduced and engine damage is prevented, but the engine start time is significantly increased and operational efficiency is reduced
Solution Approach 1:
The system performs preliminary identification of bowed rotor conditions using machine learning algorithms that analyze sensor data during engine operation and shutdown. By detecting the condition early and predicting its severity, the system can apply targeted mitigation actions only when necessary, rather than always performing lengthy motoring sequences. This preliminary detection and prediction capability allows for optimized start procedures that prevent both unnecessary delays and potential damage.
2Ease of manufacture
If preset routines are applied to alleviate bowed rotor during engine restart, then a standardized solution is provided, but the solution does not adapt to actual BRS conditions and cannot quickly identify when BRS has occurred
Solution Approach 1:
The system transitions from static preset routines to dynamic, adaptive control by continuously monitoring engine parameters and using machine learning models to predict bowed rotor conditions in real-time. The control logic automatically adjusts mitigation strategies based on predicted condition severity, operational history, and current sensor readings. This dynamic approach maintains the ease of standardized operation while achieving adaptability to actual BRS conditions through intelligent algorithms that learn from operational data.
Solution Approach 2:
The system implements closed-loop feedback by continuously monitoring sensor data during engine operation and shutdown, comparing actual conditions against predicted BRS scenarios using machine learning models. The feedback mechanism allows the system to identify when BRS has occurred, assess its severity, and adjust the mitigation routine accordingly. This feedback-driven approach enables standardized procedures to adapt to actual conditions, optimizing both reliability and operational efficiency.
3Productivity
If machine learning algorithms are used to generate adjusted ground operation schedules, then turnaround time is reduced and operational efficiency is improved, but system complexity increases
Solution Approach 1:
The system employs self-service machine learning algorithms that automatically learn from operational data and improve their predictions over time without requiring manual reconfiguration. The algorithms autonomously identify patterns in sensor data, predict BRS conditions, and optimize ground operation schedules based on learned insights. This self-learning capability reduces the need for complex manual programming and system configuration, achieving high productivity with manageable complexity through autonomous adaptation.
Data Source
AI summary
A method for operating a turbine engine is provided. The method includes receiving operating data comprising at least an engine operation parameter, an environmental parameter, a location parameter, and a time parameter; operating the turbine engine based on a baseline ground operation schedule; generating an adjusted ground operation schedule based on the operating data and the baseline ground operation schedule, wherein generating the adjusted ground operation schedule is based on a machine learning algorithm; and operating the engine based on the adjusted ground operation schedule.