Gas Turbine Combustion Vibration Control via Fuel Database Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional gas turbine control methods struggle to accurately adjust fuel and air flow rates to prevent combustion vibrations, which can occur due to changes in fuel composition and compressor performance, limiting operational stability and availability.
Innovation Solution
The method involves dividing detected data into multiple databases based on value levels to reduce non-linearity and discontinuity in modeling combustion vibration characteristics, allowing for more accurate approximation and adjustment of fuel and air flow rates to prevent vibrations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a single bias coefficient is used to adjust fuel flow rate or air flow rate in response to fuel calorific value changes, then the control method is simple, but the degree of freedom for control adjustment is limited and it is difficult to converge to optimally controlled values
Solution Approach 1:
The patent divides the single bias coefficient into multiple bias coefficients corresponding to different frequency bands. Each frequency band has its own bias coefficient that can be independently adjusted, allowing for more precise control of combustion vibrations at different frequencies while maintaining operational simplicity.
Solution Approach 2:
The patent introduces a frequency band dimension to the control system. Instead of using a single scalar bias coefficient, the system now operates in a multi-dimensional space where each frequency band represents a dimension, enabling more degrees of freedom in control adjustment.
2Reliability
If conventional control methods are used with predetermined flow rates, then the control system is simple and stable, but combustion vibrations may occur due to deviations in fuel contents and compressor performance
Solution Approach 1:
The patent implements a feedback control mechanism where combustion vibrations are detected, analyzed by frequency band, and used to adjust the corresponding bias coefficients. This closed-loop feedback system enables the control method to adapt to changing fuel compositions and compressor performance while maintaining combustion stability.
Solution Approach 2:
The patent dynamically changes the bias coefficients based on detected combustion vibrations and their frequency band analysis. By adjusting these parameters in response to actual operating conditions, the system adapts to varying fuel contents and compressor performance, preventing combustion vibrations.
3Reliability
If the fuel flow rate or air flow rate is frequently adjusted to prevent combustion vibrations, then combustion stability is maintained, but the operational availability of the gas turbine may be reduced
Solution Approach 1:
The patent applies partial adjustment by only modifying the bias coefficients when combustion vibrations are detected in specific frequency bands. Instead of continuously or excessively adjusting flow rates, the system makes targeted, minimal adjustments only when and where needed, thereby maintaining combustion stability while preserving operational availability.
Data Source
AI summary
To restrain combustion vibration more precisely by alleviating the relationship between the composition or heat quantity of the fuel and the combustion vibration characteristic and by grasping the combustion vibration characteristic more accurately. The combustion vibration generated by burner 32 of gas turbine 1 as well as the composition or heat quantity and other plant status amounts of fuel f supplied to the burner are detected. The combustion vibration characteristic is grasped based on the detected values. When the combustion flow rate or air flow rate supplied to the burner is increased/decreased in order to obtain operation condition under which no combustion vibration occurs, the detected values of the combustion vibration as well as the composition or heat quantity and other plant status amounts of the fuel are divided and stored in multiple databases corresponding to the value of the composition or heat quantity of the fuel. In this way, the accuracy of the treatment for approximating the combustion vibration characteristic performed for each data stored in each database is improved. When two databases redundantly store the data in the boundary region of the composition or heat quantity of the fuel, the discontinuity of the model illustrating the combustion vibration characteristic derived by means of the approximation treatment during data processing is reduced.