APPLICATION OF NEURAL STRUCTURES IN WATER QUALITY MEASUREMENTS

O. Postolache, M. D. Pereira, P. Girão, M. Cretu, C. Fosalau

Abstract

The paper presents a theoretical and practical study related with the implementation of artificial intelligence for the enhancement of water quality measurement systems accuracy. The measurement parameters related to water quality (WQ), considered in the present work, are essentially the pH and the conductivity (C). The fusion of sensor information is performed using two different architectures of the Artificial Neural Networks. Those architectures are compared in terms of the implementation, complexity of associated digital signal processing and accuracy enhancement. An additional measure of temperature is used to offer details regarding the presence of micro-organisms in tested water and also to compensate the temperature influences on the pH and C measurements. The test environment is implemented by a virtual measurement system, which includes a hardware component based on a data acquisition board and a software component developed in LabVIEW graphical programming language.

Keywords

water quality measurement, neural networks, accuracy

Download

IMEKO-WC-2000-AI-P594.pdf

DOI

-

Back to the proceedings