CALIBRATION AND MEASURAND RECONSTRUCTION FOR LEVEL AND FLOW MEASURING SYSTEM USING ANN

Abstract

Calibration is the essential need in any type of measurement (e.g. Chemical, mechanical, electrical etc.) for getting the desired results within the specific range. It permits the reliable service in the scientific and engineering jobs. Both manufactures and consumers feel satisfaction with the perfection of capacity to do job properly by the device is used. Artificial neural network provides an easiest way to calibrate the measurement system. Basically ANN is a computational structure inspired by the study of biological neural processing, which have capability to organize synthetic neurons to solve the same kinds of difficult, complex problems in a similar manner as we think the human brain may. Here error back propagation training algorithm (EBPTA) is used for training and measurand reconstruction for level and flow measuring system. It makes possible to train the proposed ANN with training patterns foi difficult systems. For data generation, Anshuman Process Control Trainer-II is used. The training patterns act as an input to the ANN and these are verified with desired output values