Vibratory Meter Noise Reduction via Intermodulation Distortion Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Vibratory meters face noise interference from intermodulation distortion signals in multi-tone drive signals, which can lead to incorrect frequency response characterization of sensor assemblies, making it difficult to detect changes such as erosion, corrosion, or damage.
Innovation Solution
A method is introduced to generate a compensating signal, either inverse weighted or canceling, which is applied to the sensor assembly to prevent noise formation or attenuation, using a non-linear system model of the transducer, such as a polynomial model, to predict and reduce intermodulation distortion signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a multi-tone drive signal is used to characterize the frequency response of the sensor assembly, then the frequency response can be determined simultaneously, but intermodulation distortion signals are induced that create noise in the sensor signals
Solution Approach 1:
The patent models the intermodulation distortion using a non-linear system model (polynomial model) and generates compensating signals that are inverse weighted versions of the predicted distortion. By injecting these compensating signals into the multi-tone drive signal, the harmful intermodulation distortion is counteracted, converting the noise problem into a beneficial noise cancellation solution
Solution Approach 2:
The system performs preliminary modeling of the sensor assembly's non-linear behavior to predict the intermodulation distortion signals before they are generated. The compensating signals are pre-calculated and injected into the drive signal in advance, preventing the distortion noise from corrupting the frequency response measurements
2Object-generated harmful factors
If the sequential tone approach is used to avoid intermodulation distortion, then noise is reduced, but the frequency response characterization is corrupted by time-variance in the system
Solution Approach 1:
Instead of avoiding multi-tone signals, the patent embraces them and uses non-linear modeling to predict and cancel the resulting intermodulation distortion. This allows simultaneous frequency response characterization while maintaining accuracy, eliminating the need for sequential tone approaches that are vulnerable to time-variance
3Object-generated harmful factors
If a non-linear system model is used to predict intermodulation distortion, then noise can be reduced, but the device complexity increases
Solution Approach 1:
The patent uses a polynomial model with adjustable coefficients to represent the non-linear behavior of the sensor assembly. By fitting the model parameters to measured data and using this compact mathematical representation, the system achieves accurate distortion prediction without requiring complex computational resources or additional hardware
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A vibratory meter (5, 1600) configured to predict and reduce noise in the vibratory meter (5, 1600). The vibratory meter (5, 1600) includes a sensor assembly (10, 1610) and a meter electronics (20, 1620) in communication with the sensor assembly (10, 1610). The meter electronics (20, 1620) is configured to provide a drive signal to a sensor assembly (10, 1610), receive a sensor signal from the sensor assembly (10, 1610) having one or more components, and generate a signal to be applied to one of the sensor signal and the drive signal to compensate for the one or more components.