Data processing: measuring – calibrating – or testing – Measurement system – Measured signal processing
Reexamination Certificate
2002-06-24
2004-12-14
Hoff, Marc S. (Department: 2857)
Data processing: measuring, calibrating, or testing
Measurement system
Measured signal processing
C324S222000
Reexamination Certificate
active
06832179
ABSTRACT:
TECHNICAL FIELD
Certain implementations relate generally to evaluating a signal, including analyzing and filtering a signal, and more particularly to analyzing and filtering a vortex flow meter signal.
BACKGROUND
Flow meters may measure the rate of flow of a fluid in a pipe or other pathway. The fluid may be, for example, a gas or a liquid, and may be compressible or incompressible. One type of flow meter is a vortex flow meter which is based on the principle of vortex shedding. Vortex shedding refers to a natural process in which a fluid passing a bluff body causes a boundary layer of slowly moving fluid to be formed along the surface of the bluff body. A low pressure area is created behind the bluff body and causes the boundary layer to roll up, which generates vortices in succession on opposite sides of the bluff body. The vortices induce pressure variations that may be sensed by a pressure sensor. The vortex-shedding pressure variations have a frequency that is related to the flow rate. Accordingly, by measuring the frequency of the pressure variations, the flow rate may be determined.
SUMMARY
One technique for measuring the frequency of vortex-shedding pressure variations includes converting the pressure values into an electric signal and determining the time between zero crossings of the electric signal. The term “vortex signal” is used to refer to the electric signal or, more generally, and depending on the context, to refer to the pressure variations or some other signal derived from the pressure variations. Inverting the time between zero crossings yields the frequency of the vortex signal. However, zero crossings of the vortex signal may be difficult to determine, particularly for low flow rates that may produce pressure variations of a lower magnitude than pressure variations produced for high flow rates. Determining zero crossings may also be difficult in the presence of noise. Noise may be present from, for example, turbulence and “plant noise” such as, for example, pump vibrations and pipe-line vibrations. A lower magnitude for the vortex signal and/or the presence of noise may result in a lower signal-to-noise ratio (“SNR”) for the vortex signal at low flow rates. Low SNRs may make it difficult to lock on to the vortex signal and/or to track the vortex signal using a zero crossing technique.
Low flow rates, and the associated low SNRs, may occur, for example, in at least three scenarios. The first scenario may occur when a flow meter has not acquired a lock on the vortex signal because the flow rate is low. For example, at start-up, a flow meter may not know the frequency of the vortex signal and, accordingly, may not filter out any noise due to the possibility that the vortex signal could also be filtered out. The second scenario may occur when a flow meter has acquired a lock on the vortex signal and is able to filter out noise that is not too close to the vortex signal frequency, but still cannot track the vortex signal to a lower flow rate because of the low SNR. The third scenario may occur when a flow meter has acquired a lock and is tracking the vortex signal at a low flow rate, but intermittent noise causes the flow meter to lose the lock and/or to track the noise.
Each of these scenarios, and others, can be addressed by providing the flow meter with additional functionality that determines that a flow rate is low and filters some of the noise out of the vortex signal. The flow rate may be determined to be low, for example, by determining that the amplitude of the vortex signal is low. The flow rate can be determined from the amplitude because the flow rate is directly related to the amplitude. The filtering may be done using, for example, a band-pass filter (“BPF”) having one or more pass bands. Filtering the vortex signal to remove noise may increase the SNR of the vortex signal and focus the flow meter on a smaller range of (low) flow values, which may help the flow meter to lock on and/or track the vortex signal.
Using amplitude detection to determine that the flow rate is low may also be more robust to noise than determining zero crossings at low flow rates. Thus, the amplitude detection may be expected to determine that the flow rate is low even when the zero crossing detection cannot. The amplitude detection may also include filtering. For example, an amplitude detector may detect peaks in the vortex signal and these peaks may be filtered to reduce the effect of noise on the peak measurements. In this way, the determination that the flow rate is low may be less likely to be changed inadvertently and to interrupt the corresponding filtering.
According to a general aspect, a signal processor for use with a zero crossing module of a vortex flow meter includes a peak detector, a comparator, and a filter module. The peak detector produces an amplitude estimate. The comparator is coupled to the peak detector and receives the amplitude estimate and a threshold amplitude. The comparator compares the amplitude estimate and the threshold amplitude to produce a comparison result. The filter module is coupled to the comparator and receives the comparison result and a signal. The filter module is operable to selectively filter the signal based on the comparison result and to provide the selectively filtered signal to a zero crossing module.
A peak filter may be disposed between the peak detector and the comparator. The peak filter may filter the amplitude estimates produced by the peak detector to produce a filtered amplitude estimate. The comparison result may indicate whether the amplitude estimate is less than the threshold amplitude, and the filter module may filter the signal using a first pass band if the comparison result indicates that the amplitude estimate is less than the threshold amplitude. The filter module may filter the signal using a second pass band if the first pass band is not used, or regardless of whether the first pass band is used. The second pass band may include a variable pass band that depends on an estimated vortex frequency of the signal. The filter module may include a first filter and a second filter. The first filter may be coupled to the second filter, and may selectively filter the signal using the first pass band. The second filter may selectively filter the signal using the second pass band.
According to another general aspect, a vortex flow meter includes a peak detector, a comparator, a filter module, and a frequency estimation module. The peak detector is operable to produce an amplitude estimate. The comparator is coupled to the peak detector, and receives and compares the amplitude estimate and a threshold amplitude to produce a comparison result. The filter module is coupled to the comparator and includes at least one filter. The filter module receives the comparison result and a signal, selectively filters the signal based on the comparison result, and provides the selectively filtered signal as an output. The frequency estimation module is coupled to the filter module and includes a zero-crossing detector and a frequency estimator. The frequency estimation module receives the selectively-filtered signal, detects zero crossings in the selectively-filtered signal, and estimates a vortex frequency of the selectively-filtered signal based on the detected zero crossings.
A peak filter may be disposed between the peak detector and the comparator. The peak filter may filter the amplitude estimates produced by the peak detector to produce a filtered amplitude estimate. The comparison result may indicate whether the amplitude estimate is less than the threshold amplitude, and the filter module may filter the signal using a first pass band if the comparison result indicates that the amplitude estimate is less than the threshold amplitude. The filter module may filter the signal using a second pass band if the first pass band is not used, or regardless of whether the first pass band is used.
According to another general aspect, processing a vortex signal in a vortex flow meter includes comparing an amplitude of a vortex signa
Clarke David W.
Ghaoud Tarek
Fish & Richardson P.C.
Hoff Marc S.
Invensys Systems Inc.
Raymond Edward
LandOfFree
Evaluating a vortex flow-meter signal does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Evaluating a vortex flow-meter signal, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Evaluating a vortex flow-meter signal will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3325899