USE OF NEURAL NETWORKS IN INTELLIGENT MEASUREMENT TOOLS
Abstract
The paper presents algorithms for processing the measurement signal with the possibilities of adaptation, learning and decision making. A comparative analysis of the methods of intellectual processing of measurement data is carried out. A model of a measuring instrument for determining the structure of a neural network has been developed. The problem of error reduction due to measurement noise filtering with the use of neural networks is considered. The structure of the neural network has been developed for intelligent processing of the measurement signal and ensuring the implementation of the functions of reconfiguration, calibration, self-diagnosis and self-control. A neural network training algorithm based on error back propagation was used. The results of the implementation of the neural network algorithm in measuring instruments with different training patterns are presented. The paper also describes the calibration of linear and non-linear smart sensors. The results of the study show that the proposed algorithm improves the quality of measurement of technological parameters.