Gas Turbine Burner Malfunction Detection via Pressure Spectra
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting malfunctions in gas turbine combustion chambers are inefficient, as they cannot accurately locate faulty burners due to fluid mixing and varying angular relationships between thermocouples and burners, leading to increased maintenance time and costs.
Innovation Solution
A method using dynamic pressure sensors and a processing unit to monitor pressure oscillations at each burner, determining initial and updated spectral parameters, and comparing them to reference thresholds to identify and locate malfunctions, thereby generating alarm messages and reducing maintenance intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If thermocouples are used to monitor combustion parameters, then combustion process quality can be evaluated, but the angular position relationship between thermocouples and burners is not fixed due to fluid mixing and varying gas turbine load, making accurate burner location difficult
Solution Approach 1:
The patent replaces the mechanical/physical thermocouple-based monitoring system with an acoustic measurement system. Dynamic pressure sensors measure pressure oscillations in the combustion chamber, and frequency spectrum analysis identifies burner locations based on acoustic signatures. This substitution eliminates the need to track varying angular positions of thermocouples relative to burners, as the acoustic field provides a fixed reference frame for location determination.
Solution Approach 2:
The patent introduces frequency spectrum analysis as an intermediary between the pressure measurements and burner location identification. The frequency spectrum serves as a mediator that transforms the complex, time-varying pressure signals into a simplified spectral representation where characteristic frequencies correspond to specific burners. This intermediary process decouples the location identification from the complex fluid dynamics and thermocouple positioning.
2Productivity
If multiple thermocouples are arranged uniformly spaced apart, then combustion parameters can be monitored, but groups of three-to-five burners must be inspected individually during maintenance stop, increasing maintenance time and costs
Solution Approach 1:
The patent implements continuous monitoring during normal operation using dynamic pressure sensors and frequency spectrum analysis. This preliminary detection of burner malfunctions allows maintenance personnel to identify and locate faulty burners before they fail completely, enabling scheduled maintenance during planned downtime rather than emergency interventions. The system provides advance warning and location information that facilitates preparation for maintenance operations.
Solution Approach 2:
The patent establishes a feedback loop where continuous pressure measurements are processed through frequency spectrum analysis to identify burner conditions in real-time. The system provides ongoing feedback about burner health status and location, allowing maintenance strategies to be adjusted dynamically. This feedback mechanism enables transition from reactive maintenance to proactive maintenance, reducing overall maintenance time and costs.
3Measurement precision
If dynamic pressure sensors are installed at each burner to monitor pressure oscillations, then precise burner location can be determined, but the system complexity increases
Solution Approach 1:
The patent makes the dynamic pressure sensors multi-functional by using them for both general combustion monitoring and specific burner location identification. The same pressure oscillation measurements that indicate combustion quality are also analyzed through frequency spectrum analysis to determine burner locations. This universal use of the pressure sensors eliminates the need for separate sensors dedicated solely to location identification, reducing overall system complexity.
Solution Approach 2:
The patent transforms the pressure measurement data through frequency spectrum analysis, changing the parameter representation from time-domain pressure signals to frequency-domain spectral components. This parameter transformation simplifies the identification process by converting complex, multi-frequency pressure oscillations into distinct spectral signatures that can be directly associated with specific burners. The frequency spectrum serves as a simplified intermediate representation that eases both monitoring and location identification.
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
This method allows for precise detection and localization of malfunctions, reducing maintenance time and costs by continuously monitoring thermo-acoustic phenomena and burner deterioration, enabling timely interventions.
Implementation Method 1
dynamic pressure signals indicating dynamic pressure at each burner are provided
Implementation Method 2
frequency spectra analysis of pressure oscillations
Implementation Method 3
Thermo-acoustic instability phenomena, frequently also known as 'humming'
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A method for detecting malfunctions in a combustion chamber of a gas turbine plant includes providing dynamic pressure signals (P1I, P2,...., PN), each indicative of a dynamic pressure at the outlet of a respective burner of a combustion chamber of the plant. Frequency spectra (S1(f), S2(f),..., SN(f), S1*(f), S2* (f),..., SN *(f)) of the dynamic pressure signals (P1, P2,..., PN) are calculated, the malfunctions of the burners are recognized according to these spectra. Each burner subject to malfunctioning is identified according to a bijective correlation between the burners and the respective dynamic pressure signals (P1, P2,..., PN).