High-precision Electromagnetic Flowmeter with Empty Pipe Detection via Complex Programmable Logic Device-based Waveform Recognition

We are interested in designing, implementing, and validating a CPOD driven electromagnetic flow meter. Exploring how waveform recognition elevates measurement of precision, ensuring stable empty pipette detection. Our challenges are suppressing electromagnetic interference, minimize the sensor thermal noise, isolating CPOD switching artifacts, and separating value weak flow signals from ambient noise, and make the result more stable. We found that 50 power frequency interference generates a distinct waveform patterns on the electrodes. When the tube is empty or contains air bubbles, this waveform exhibits specific characteristics. By analyzing these unique patterns, we can determine whether the tube is empty or contains bubbles. To meet the requirement of wide flow range detection, a variable gain operation amplifier circuit is designed to achieve higher precision. A multi-stage bandwidth hardware filter enhances the signal to noise ratio, while a software filter further improves the system stability. We wish to enhance noise resilient waveform analytics, adopt a CPOD algorithm for multiphase and policy flows, and embedded self calibration low power sensors for real time industrial IOG diagnostics.

[Narrator] To begin, take the induced electromotive force from both sides of the sensor as the input signal. Filter the noise using bypass capacitors. Apply a 10X differential amplifier to amplify the input signal. Feed the amplified signal into a second order band pass filter, starting with a high pass filter to remove low frequency components, then channel the filtered output through a coupling capacitor into the low pass filter stage. Using an inverting amplifier, amplify the denoised signal, then apply a gain of negative one through the inverting amplifier to convert the negative polarity signal into positive polarity, preserving the amplitude. Direct the positive and negative half cycle signals to two separate channels of the analog switch. Simultaneously input both signals into the comparator. Process the output signals from the comparator using a complex programmable logic device to detect pipeline vacancy and determine fluid flow direction. After signal gating via the analog switch, feed the signal into a third stage amplifier. Process the amplified signal using an integrating low pass filter. Transmit the final filtered signal to the microcontroller unit for computational processing. Position the signal amplifier near the band pass filter. Connect the amplifier to the output of the band pass filter, followed by the secondary amplifier to receive the band pass output. Configure two comparators below the analog switch. Finally, input the rectified signal from the analog switch into a variable gain amplifier. Route the output through a low pass filter and into the analog to digital conversion channel of the processor. The flow rate measurements from three repeated experiments using the same device showed highly consistent results across the entire measurement range, confirming strong data reproducibility and intrinsic linearity. When comparing the four experimental devices to the standard instrument, all devices showed high measurement consistency at identical standard flow rates, as well as excellent linearity over the full range. After applying linearity correction, the measurement deviations of the four devices from the standard values were significantly reduced, enhancing the system's accuracy. At low flow velocities, the relative error was noticeably higher and gradually decreased with increasing velocity, reflecting the influence of signal to noise ratio on measurement accuracy.