PSO trained hybrid intelligent classifier using wavelet and statistical features for pipeline leak classification

Abstract EN

One of the main problems in the oil industrial field is the leakage in transporting pipelines due to its effect on human society , environment, and money loss.

Therefore, the bottleneck for most researches in this subject is to minimize false alarm rate (FAR) for the adopted leak detection method.

Although some recent methods succeed in classifying the existence or absence of the leak as a binary classification problem.

But this paper proposed a novel leak detection technique which predicts the leak location and estimates its size within certain pre-defined ranges.

In order to simulate the environmental conditions for real-time operating oil pipeline, accurate simulator known as OLGA program creates the oil physical parameters.

Various methods for features extraction are considered such as statistical and wavelet techniques which are implemented to get the features from the fluid simulated waveforms.

These features are organized and fed to an ANN classifier trained by PSO algorithm to determine the leak class out of 10 suggested classes.

The proposed leak detection technique is used to simulate 18 kilometers belonging to Iraqi crude oil pipelines company operated in Baghdad.

The achieved results of the true positive rate (TPR) for the proposed applied method for the leak detection and classification of different leak classes in terms of their positions and magnitudes were about 97%.