Coriolis Mass Flowmeter Phase Difference Algorithm Research And Realization Based On FPGA

With high speed modern industry development, there is a huge demand for reliable and high efficient precision mass flow meter. Coriolis mass flow meter which can measure mass flow directly based on the Coriolis force effect arouses wide concern and turns into research focus. Frequency and phase difference are the key parameters to reflect its flow mass measurement. By using traditional way of analog circuit processing with comparator, it can’t implement high precision’s zero-crossing detection. Digital signal processing is capable to improve precision of phase difference, but its theoretical algorithm is too complex in general. It needs too many multipliers and has high request for the processor. This paper aims to design algorithm which has low cost of multipliers to implement high precision’s phase difference detection on FPGA(Field Programmable Gate Array).Firstly, The paper uses the way of radix extraction to analyze the discrete fourier transform, and exploers the calculating amount and multiplier amount on the condition of different radix, then synthesize the amount of calculating and multiplier to get the right radix. This radix is simulated by Matlab program, which includes address transform and butterfly processing unit. Then the program is inputted with different points’sample signal which is mixed with white noise to analyze the point number’s influence to the phase precision. Research result indicates that the cost of multiplier increases with higher radix, but the calculating amount just decreases a little, and the radix 2 uses the lowest cost of multiplier, then the result indicate that 2 is the right radix. The phase result get by different amount of sample data indicates that increasing the amount is positive for the precision’s development, and the relative error can be reduced to two decimals with 1024 points.Secondly, after determining the algorithm, this paper designs the software and hardware system based on FPGA. On the hardware side, it mainly works for signal’s preprocessing, including filtering, amplifying and sampling. Filter block is active butterworth lowpass filter, its order is 4 which is calculated by corresponding normal filter model, and the block filters out stop-band signal beyond 120Hz with rejection gain of -49.6dB; Amplifying block amplifys small signal 50 times in way of negative feedback, then converts into differential output; Analog-to-digital block implements peripheral circuit of ADS 1255, and samples the signal outputted by amplifying block with high precision. On the software side, besides the necessary SPI(Serial Peripheral Interface), UART(Universal Asynchronous Receiver/Transmitter) communication bus, it mainly implements 1024 points’ radix-2 extraction on FPGA. SPI block implements same time sequence with ADS 1255’s SPI, and realizes data’s sending and receiving; UART block responds sending result back to PC, and realizes sending function without flow control; as software’s main content, radix-2 extraction block includes several submodules, such as address transform, data memory, butterfly processing unit and so on, and it realizes 1024 point’s radix-2 extraction algorithm. After finishing the hardware and software, input multigroup signal and the measured result shows that the phase difference detection precision reach 0.01 degree, and the system reachs target of high precision’s measuring, works reliably and efficiently.The FPGA development platform used during the design is provide by Guangxi Academy of Sciences.