Spectral Oscillation Amplitude Estimation for Turbulent Flow Instability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately predict the amplitude of oscillatory instabilities in industrial systems like gas turbines and wind turbines, especially under high-pressure conditions, without requiring excitation of large amplitude oscillations, which are difficult to achieve.
Innovation Solution
A system comprising a sensor, analog-to-digital converter, amplitude estimator, processing unit, and controller to estimate the amplitude of limit cycle oscillations by analyzing spectral measures of time series signals, and take control actions to prevent instability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods using flame describing function are used to estimate amplitude of oscillatory instabilities, then amplitude prediction can be achieved, but the method requires excitation of large amplitude oscillations which is difficult to achieve in high-pressure industrial systems
Solution Approach 1:
The patent introduces spectral density as an intermediary parameter that can be measured without requiring large amplitude oscillations. Instead of directly measuring the difficult-to-obtain flame describing function, the system uses spectral density of pressure fluctuations as a mediator to infer instability amplitude, making the measurement process feasible in high-pressure industrial conditions
Solution Approach 2:
The patent replaces the mechanical excitation approach (needing physical oscillation excitation) with a spectral analysis approach. By substituting the mechanical measurement method with frequency domain spectral density analysis, the system can predict instability amplitudes without requiring physical excitation of large oscillations
2Measurement precision
If flame describing function methods are used, then amplitude estimation is possible, but the accuracy is sensitive to acoustic boundary conditions which are difficult to determine in industrial systems
Solution Approach 1:
The patent extracts the essential instability information from the complex acoustic field by focusing solely on spectral density measurements. This extraction approach removes the dependency on knowing detailed acoustic boundary conditions, isolating the key parameter (spectral density) that directly relates to instability amplitude without requiring boundary condition data
3Reliability
If early detection system is implemented, then stability can be maintained, but additional monitoring and control equipment is required
Solution Approach 1:
The patent makes the monitoring system multi-functional by using the same spectral density analysis framework for both detection and prediction. The processing unit performs multiple functions (detecting onset, predicting amplitude, assessing risk) using a single measurement approach, reducing the need for separate specialized equipment for each function
Solution Approach 2:
The system uses the existing operational data and natural pressure fluctuations in the industrial system to perform stability assessment. Instead of requiring external excitation devices or additional complex instrumentation, the system leverages the system's own operational characteristics and ambient measurements to self-diagnose instability conditions
Data Source
Figure 1
Figure 2~3
AI summary
Embodiments herein provide a system (100) to estimate the amplitude of oscillations in a turbulent flow system (102) that exhibits oscillatory instabilities. The system (100) comprises of a sensor (102A) mounted on the turbulent flow system (102) to detect an oscillatory variable in the system obtaining a signal, a signal conditioner (104) that conditions the signal from the sensor, an amplitude estimator (110) that estimates the amplitude of the limit cycle oscillations, and also predict the proximity of the system to the oscillatory instability, a processor (108) connected to the amplitude estimator (110) to compare the predicted oscillation amplitude with a threshold value. The amplitude is estimated by estimating the spectral measure of the time series signal obtained from the system.