Smart calibration technique for auto-ranging of LVDT using support vector machine

Design of a calibration circuit for linear variable differential transformer (LVDT) used in the measurement of thickness. The objective of the proposed work is to design a calibration technique, which is adaptive to variation in the range of measurement. Sensitivity in measurement is one of the important parameters, which is always expected to be higher for an ideal instrument. Sensitivity of an instrument is fixed during the process of calibration for an instrument and it depends on the minimum and maximum values of measurement. Whenever there exists a condition, involving only a part of measurement range the sensitivity remains constant, in general the sensitivity should have been increased. For varying the sensitivity, there will be a need to recalibrate the instrument which is time consuming and tedious. In the proposed work, a Support Vector Machine (SVM)-based learning algorithm is used in place of a conventional calibration circuit, which will calibrate automatically based on the specified range.