Gas Turbine Flashback Detection via Combustion Dynamics Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Gas turbine engines lack a reliable method to detect impending flashback or flame holding events, which can cause significant damage and unscheduled shutdowns due to the inability to monitor combustion dynamics effectively under varying operating conditions.
Innovation Solution
A system and method that utilize combustion dynamics simulation and prediction tools to compare sensor signals from the combustor with predicted frequencies and amplitudes, providing an impending flashback indicator and enabling control measures to prevent damage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If gas turbine engines operate under varying load conditions, then performance requirements change, but reliable flashback detection becomes difficult
Solution Approach 1:
The system dynamically adapts the flashback detection criteria based on operating conditions. The controller receives signals from multiple sensors (ion current, pressure, temperature) and adjusts the detection thresholds and frequency analysis parameters according to the current load condition, allowing reliable detection across varying operational states
Solution Approach 2:
The system changes detection parameters based on operating conditions. By monitoring ion current frequency spectra and comparing against predicted flashback frequencies that vary with load, the system maintains detection reliability across different temperature, pressure, and fuel flow conditions
2Device complexity
If traditional sensor systems are used, then device complexity is low, but flashback detection capability is insufficient
Solution Approach 1:
The ion current sensor serves multiple functions: it monitors combustion stability, detects flashback conditions through frequency analysis, and provides operational data for control optimization. This multi-functionality enhances detection capability without proportionally increasing device complexity
Solution Approach 2:
The system implements feedback by continuously analyzing ion current frequency spectra and comparing them against predicted flashback frequencies. When a mismatch is detected, the controller receives feedback signals to adjust operating parameters or trigger alarms, enabling reliable flashback detection with existing sensor infrastructure
3Object-affected harmful factors
If flashback occurs, then damage to injectors and nozzles is prevented by control measures, but unscheduled shutdown may occur without detection
Solution Approach 1:
The system performs preliminary detection of flashback conditions by analyzing ion current frequency spectra before actual hardware damage occurs. By identifying frequency mismatches indicative of flashback onset, the controller can take preventive action (adjust fuel-air ratio, modify operating parameters) to avoid both damage and unnecessary shutdowns
Solution Approach 2:
The system applies preliminary anti-action by detecting early signs of flashback through frequency analysis and counteracting the harmful effect before it propagates. The controller adjusts operating conditions to prevent flame propagation upstream, thereby protecting hardware while maintaining continuous operation
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables early detection and prevention of flashback events, ensuring uninterrupted operation and reducing hardware damage by adjusting operating conditions or shutting down the turbine when impending flashback is detected.
Implementation Method 1
predicting frequencies of combustion dynamics in a combustor using operating conditions of a gas turbine, receiving a signal from a sensor that is indicative of combustion dynamics in the combustor
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for monitoring and controlling a gas turbine (10), comprises predicting (86) frequencies of combustion dynamics in a combustor (14) using operating conditions (40) of a gas turbine (10), receiving (88) a signal (38) from a sensor (36) that is indicative of combustion dynamics in the combustor (14), and detecting (94) a flashback if a frequency of the received signal does not correspond to the predicted frequencies.