TY - JOUR ID - TI - PCA Classification of vibration signals in WSN based oil pipeline monitoring system AU - Waleed F. Shareef AU - Nasheed F. Mossa PY - 2019 VL - 11 IS - 1 SP - Comp Page 60 EP - 71 JO - Journal of Al-Qadisiyah for Computer Science and Mathematics مجلة القادسية لعلوم الحاسوب والرياضيات SN - 20740204 25213504 AB - Using wireless sensor network technology in structure health monitoring applications results in generating large amount of data. To sift through this data and extract useful information an extensive data analysis should be applied. In this paper, a Wireless Sensor Network (WSNs) is proposed for the oil pipeline monitoring system with proposed method for event detection and classification. The method depends on the Principal Component Analysis (PCA). It applied to features extracted from vibration signals of the monitored pipeline. These vibration signals are collected while applying damage events (knocking and drilling) to the oil pipeline. PCA is applied to features extracted from both time domain and frequency domain. The results manifest that this method is able to detect the existence of damage and also to distinguish between the different levels of harmful events applied to the pipeline.

ER -