International Journal of Engineering and Technology

U-Tube Manometer is used to measure pressure. It is calibrated using the variation in capacitance. In U-tube manometer, the relation between the level of mercury and the capacitance developed across the copper plates of the manometer is found to be highly non-linear. Due to its non-predictive nature and non-linear relationship, artificial intelligence techniques are used to calibrate the system. The artificial intelligence technique used here is Adaptive Neuro-Fuzzy Inference System (ANFIS). The system is implemented using the microcontroller ARM Cortex-M3. In U-Tube Manometer, the capacitance values obtained from capacitive level sensor are converted into frequency through an astable multivibrator. The counted frequencies are estimated using the Kalman Filter. The Kalman filter is employed to suppress the abnormalities from minute capacitance change of measurements for promoting efficiency in frequency estimation and amplitude estimation of the distorted signal. After frequency estimation, training is done using the ANFIS. The implemented hardware can be used to efficiently calibrate the U-Tube manometer. This system shows that the ANFIS can be implemented in hardware and can be used to calibrate the non-linear systems effectively.