ANN Modelling of Cu Type Omega Vibration Based Mass Flow Sensor

摘要

Artificial neural network (ANN) based model has been developed for copper omega type mass flow sensor. The significant phase shift parameter is modeled for various input factors like sensor location, drive frequency, mass flow rate and length of tube. This model is found in good agreement with the experimental results after comparison. The effectiveness of the ANN model was tested with test data. The correlation coefficient(R) between the predicted phase shift and the experimental phase shift for training, validation and test data was found to be acceptable for prediction of phase shift by taking sensor location, mass flow rate, length of tube and frequency excitation as input parameters. A reliable and useful predictor for phase shift for future studies in Cu type mass flow sensor may be developed based on large number of variables used during training the model.