Application of DSP Digital Signal Processor in Coriolis Mass Flow Meter

feifei

Application of DSP Digital Signal Processor in Coriolis Mass Flow Meter [Copy link]

Micro Motion, a subsidiary of Emerson Process Management
　　
, recently launched the MVD multi-parameter digital transmitter, which features the use of DSP digital signal processing technology, demonstrating Micro Motion's strong strength in Coriolis mass flow measurement technology. 　　Micro Motion's MVD multi-parameter digital technology provides a modular structure to redefine sensors and transmitters and make flow meters work more flexibly. The core processor of the DSP digital signal processor is installed together with the sensor, converting the analog signal from the Coriolis sensor into a digital signal and generating an electronic signal proportional to the mass flow.
　　
　　The 1000 series and 2000 series transmitters can be connected to the core processor through an ordinary 4-wire cable, which will output the final measurement signal, provide display and some other functions. The transmitter can also be installed integrally with the core processor.
　　
　　What is a DSP digital signal processor?
　　
　　The DSP digital signal processor is a microprocessor that processes signals in real time. The microprocessor in your home computer works with data stored in memory. This is fine for things like checking or playing video games, but it can't handle some real-world stuff like audio signals, video signals, signals from medical sensors, or signals from Coriolis sensors. Here we need a very fast microprocessor to do whatever analysis we want to do with these signals.
　　
　　Just like the microprocessor in your home computer needs a monitor, a disk drive, a printer, software, and some connecting cables, the DSP digital signal processor also needs supporting software and hardware. In the DSP world, the first thing we need to do is convert the real-world signals into the DSP world signals. The devices used are called "analog-to-digital converters."
　　
　　We also need some software to operate on the "digitalized" signals. Let's take an example to see what we can do with software. During long-distance calls, we sometimes hear an echo of our own voice, which is annoying. The human ear is used to filtering out short echoes, but long echoes make communication very difficult. The phone company makes a copy of your voice and then adds it to its reverse at the right time to cancel the echo. It's not that the echo doesn't happen, it's just filtered out by very sophisticated DSP digital signal processing software. In a Coriolis flowmeter, we vibrate the measuring tube at a known frequency, so any frequencies outside this vibration frequency range are "noise" and need to be removed to accurately determine mass flow. For example, a 50Hz or 60Hz signal is likely to come from coupling with a nearby power line. How to actually "filter" these unwanted signals requires some more contextual information available at that moment. Figure 1 shows how noise appears on the original converter signal, and the final signal after it has been filtered.
　　

600)this.width=600" border=0>

　　Now that we have processed the signal, we need to convert it from the digital world back to the real world. The device that completes this task is a "digital-to-analog converter". We need some memory to store the DSP digital signal processing program, and we also need some control devices to implement the DSP digital signal processor.
　　
　　2. The benefits of DSP digital signal processing technology for Coriolis mass flowmeters
　　
　　Just like the benefits of home computers in processing data, DSP digital signal processing technology also brings the same benefits to processing real-world signals: DSP digital signal processors are much smaller than traditional analog processors, which is how we can encapsulate all technologies into the core processor and make the sensor intelligent; compared with traditional analog processors, DSP digital signal processors use less energy and fewer components, and improve reliability; DSP digital signal processors are at least one order of magnitude more accurate than similar analog processors, which means that even poor sensor signals can produce better final measurements; through software updates, the core processor can be applied to other sensor types. For Micro Motion products, this means faster market development; for users, this means fewer spare parts.
　　
　　3. Some mathematical knowledge related to DSP digital signal processors
　　
　　Signals that exist in nature are generally continuous and can be represented by continuously changing voltage signals. The signal of the Coriolis flowmeter is also a continuous signal. When we send the signal through an analog-to-digital converter, we have actually quantized the signal into discrete or digital samples. For example, suppose we send the converter voltage through a 12-bit ADC at a sampling rate of 1000 samples per second. Every millisecond, we quantize the signal into one of 212 = 4096 possible levels. Figure 2 shows a quantized signal.

600)this.width=600" border=0>

　Running the ADC for one second we can take 1000 samples of the converter voltage, we call this number of samples N. If desired we can calculate the average of the converter voltage by adding all the samples together and dividing by N. In a similar fashion we can calculate the standard deviation of the signal, the average represents the actual signal we want to measure and the standard deviation represents the noise signal. The square of the average divided by the square of the standard deviation is called the signal-to-noise ratio or SNR. The higher the signal-to-noise ratio, the higher the quality of the data being analyzed. These calculations can be used to calculate the value of the measured variable. Filtering and bandwidth reduction (technically called decade droop) can be used to improve the signal-to-noise ratio and the accuracy of mass flow.
　　
　　Fourier Analysis
　　
　　Fourier analysis is an analysis method named after the French mathematician and physicist Jean Baptiste Loseph Fourier.
　　
　　Fourier believed that any continuous periodic signal can be described by a sum of appropriately chosen sinusoidal waves. Taking a continuous periodic signal and converting it to a family of sinusoidal waves is defined as performing a Fourier transform. The Fourier transform is mathematically complex but we only need to understand it briefly. The core processor takes the quantized converter signal and performs a Fourier transform of the signal, as shown in Figure 3.
　　

600)this.width=600" border=0>
5. Digital filtering
　　
　　In the spectrum of the signal in Figure 3, there is only one signal data, and the rest are noise. The 100Hz signal represents the vibration frequency of the measuring tube. We also see signals at frequencies such as 200Hz, 300Hz, and 400Hz, which are called second, third, and fourth harmonics. We also see a small 60Hz signal from the power line coupling.
　　
　　These data are just a table in the DSP memory. What we want to do is to discard any information that is not needed in the actual measurement, that is, to ignore the information outside the 100Hz measuring tube frequency. This is called digital filtering.
　　
　　Note that in Figure 3 there is only one signal near the 100Hz measuring tube frequency. In older sensors, it is usually very difficult to determine what is data and what is noise near the signal. Micro Motion sensors have an exceptionally high signal purity near the measuring tube operating frequency, which is an important reason why Micro Motion mass flow meters have high accuracy.
　　
　　6 . Practical significance of DSP digital signal processing technology for Micro Motion mass flowmeters
　　
　　One of the main benefits of using DSP digital signal processing technology is the ability to filter real-time signals at an increased sampling rate compared to using time constants to dampen and stabilize signals, which makes the flowmeter respond much faster to step changes in flow. The response time of using MVD multi-parameter digital transmitters is 2 to 4 times faster than traditional transmitters using analog signal processing. The faster response time will improve the efficiency and accuracy of short batch control. In the engine test rig, we can better measure the engine's response to step changes in fuel injection. Using a compact calibration device can also improve the ability to calibrate Micro Motion flowmeters in the field. Figure 4 shows the response of MVD multi-parameter digital Coriolis transmitters, pressure transmitters, and ordinary Coriolis transmitters to step changes in flow.

600)this.width=600" border=0>

　　Another valuable example of DSP digital signal processing technology is gas measurement. Gas measurement is a more challenging application because high-speed gas passing through the flow meter will cause relatively serious noise. Through the Micro Motion Elite series sensor, the noise mixed with the flow signal has been minimized. Now DSP digital signal processing technology can better filter and further reduce the sensitivity of mass flow meters to noise. The results of measuring gas using MVD multi-parameter digital transmitters have been significantly improved in repeatability and accuracy, as shown in Figures 5 and 6.

600)this.width=600" border=0>

600)this.width=600" border=0>

　7. Future
　　DSP digital signal processing technology provides a "window to processing". Today, when browsing this window, we first focus on the signal near the vibration frequency of the measuring tube. In fact, the rest of the information is intentionally discarded. It is very likely that the information hidden in these "useless" data will pave the way to new diagnostic technologies. For example, spectrum analysis may lead us to make progress in the measurement of entrained air or slug flow fluids. The adhesion of fluids to the inner wall of the measuring tube is another fault that is expected to be detected by DSP digital signal processing technology. The change of the spectrum is also likely to be used to predict sensor failure.
　　
　　8. Summary
　　Today, DSP digital signal processing technology has demonstrated its value by giving mass flow meters a faster, more reliable, more efficient, more stable and more flexible solution. This also makes our sensors more flexible. For the future, Micro Motion is full of confidence that DSP digital signal processing technology will show great potential in promoting flow measurement.