Rotor Blade Tip Timing for Noisy Turbomachine Vibration Monitoring

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for inferring vibration frequency, amplitude, and phase of rotor blades in turbomachines face challenges due to high noise levels and limitations in data measurement, particularly with conventional Fourier transform and Auto-Regressive methods, which are not effective in all scenarios.

Innovation Solution

A Bayesian statistical model is employed to estimate vibration characteristics by measuring Time-of-Arrival (ToA) of rotor blade tips, using Bayesian linear regression and inference to determine probabilistic model coefficients, which provides a more robust approach to infer vibration behavior and damage accumulation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional Fourier transform or Auto-Regressive methods are used to infer vibration characteristics, then the measurement process is simple, but the measurement precision deteriorates due to high noise levels and limitations in data measurement

Engineering Contradiction:

Improvevibration characteristic estimation accuracyVSAvoidBayesian statistical model complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces a Bayesian statistical model as an intermediary layer between the raw BTT measurements and the vibration characteristic estimation. This model incorporates prior knowledge and handles measurement uncertainties, acting as a mediator that filters noise and provides more accurate estimates of vibration frequency, amplitude, and phase despite the complexity of the statistical framework

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the parameter representation from deterministic values to probabilistic distributions. By using Bayesian inference, the vibration characteristics are estimated as probability distributions rather than fixed values, allowing the system to quantify uncertainty and improve measurement precision in noisy environments through parameter transformation

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If more measurement data is collected to improve vibration analysis, then the measurement precision improves, but the loss of time increases due to extended measurement periods

Engineering Contradiction:

Improvevibration characteristic estimation accuracyVSAvoidmeasurement acquisition time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by incorporating prior knowledge into the Bayesian statistical model before actual measurements are taken. The prior distributions encode existing information about vibration characteristics, allowing the system to make accurate estimates with fewer measurements and reduced acquisition time while maintaining measurement precision

Inventive Principle:

Principle #10Preliminary action

3Reliability

If deterministic methods are used to solve vibration model coefficients, then the calculation process is straightforward, but the reliability deteriorates due to inability to handle noise and uncertainty

Engineering Contradiction:

Improverobustness against noise and uncertaintyVSAvoidBayesian inference implementation complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent converts the harmful effect of noise and uncertainty into a benefit by using Bayesian inference. Instead of treating noise as a problem to be eliminated, the methodology incorporates it into the probabilistic framework, where measurement uncertainties are explicitly modeled and transformed into reliable probability distributions for vibration characteristics, making the system more robust through the very uncertainties that previously harmed it

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

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 effectively handles noise and uncertainty, providing accurate estimates of vibration characteristics and damage prediction, leading to improved maintenance scheduling and reduced risk of catastrophic failures.

Implementation Method 1

measuring, by at least one proximity sensor, a Time-of-Arrival (ToA) of a proximate tip of a moving rotor blade

Methodology Applied

Scientific EffectProximity sensing:

Data Source

PatentUS10895496B2Method and system for monitoring rotor blades of a turbomachine using blade tip timing (BTT)

Publication Date: 2021.01.19 UNIV OF PRETORIA

AI summary

A method of estimating vibration characteristics of moving rotor blades in a turbomachine comprising a housing and rotor including a shaft with the rotor blades attached thereto includes measuring, by at least one proximity sensor, a Time-of-Arrival (ToA) of a proximate tip of a moving rotor blade. The method includes manipulating, by a control module, timing data to determine a corresponding blade tip deflection at each measured ToA, expressing blade tip vibration behaviour as a Bayesian statistical model consisting of intermediate model parameters, wherein model coefficients in the statistical model are probabilistic quantities as opposed to conventional, deterministic quantities, and expressing, by the control module, the determined blade tip deflections as observed values in the statistical model. The method includes manipulating the intermediate model parameters in the statistical model using Bayesian linear regression or inference to estimate the intermediate model parameters and consequently the vibration behaviour of the rotor blade.