Artificial neural network and wavelet features extraction applications in nitrate and sulphate water contamination estimation

• Main Authors: Azmi, A., Khaman, K. K., Ibrahim, S., Khairi, M. T. M., Faramarzi, M., Rahim, R. A., Yunus, M. A. M.
• Format: Article
• Published: Penerbit UTHM 2017
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/77156/; https://www.scopus.com/inward/record.uri?eid=2-s2.0-85041593226&partnerID=40&md5=5d07e1a4ad07f68d9a8ea6f3809a2126

This work expounds the review of non-destructive evaluation using near-field sensors and its application in environmental monitoring. Star array configuration of planar electromagnetic sensor is explained in this work for nitrate and sulphate detection in water. The experimental results show that the star array planar electromagnetic sensor was able to detect nitrate and sulphate at different concentrations. Artificial Neural Networks (ANN) is used to classify different levels of nitrate and sulphate contaminations in water sources. The star array planar electromagnetic sensors were subjected to different water samples contaminated by nitrate and sulphate. Classification using Wavelet Transform (WT) was applied to extract the output signals features. These features were fed to ANN consequently, for the classification of different levels of nitrate and sulphate concentration in water. The model is capable of distinguishing the concentration level in the presence of other types of contamination with a root mean square error (RMSE) of 0.0132 or 98.68% accuracy.