A Novel Signal Processing Method of Coriolis Mass Flowmeter Based on Svr and Hilbert Transform

The traditional Coriolis flowmeter signal processing method obtains the mass flow by calculating the time difference. However, in the actual measurement, noise and changes in the surrounding environment will affect the accuracy of the mass flow measurement. To improve the measurement accuracy of mass flow and reduce the impact of noise, a novel signal processing method for Coriolis mass flowmeter based on SVR and Hilbert transform is proposed in this paper. The Hilbert transform is performed on the vibration signal obtained by the sensor firstly, the phase difference extracted by the Hilbert transform and the mass flow of the standard meter are used as the sample feature and the label value respectively to build the support vector regression machine model. Apply the trained model to the test set to make predictions on the data. The test result shows that the mean square error of the prediction result obtained by this algorithm is 1.0003E-03, which is better than 3.937E-03 of the checked meter.