Internal valve leakage in a natural gas pipeline seriously impairs the safe operation on pipelines, and the recognition of leakages has therefore been a major concern of the industry. In this study, a novel leakage detection scheme based on kernel principal component analysis (kernel PCA) and the support vector machine (SVM) classifier for the recognition of the leakage level is constructed. Using this approach, the acoustic signal of the leakage is obtained as the feature source using an acoustic emission (AE) sensor. The kernel PCA is used to reduce the dimensionality of the features and extract the optimal features for the classification process, and the SVM is applied to perform the recognition of the leakage levels. The performance of the classification process based on kernel PCA and the classifier are evaluated in terms of the accuracy, Cohen's kappa number and training time. The experimental results demonstrate that the intelligent recognition model based on kernel PCA and SVM classifier is very effective for recognizing the leakage level of a valve in a natural gas pipeline.The inevitable events of gas pipeline valve leakage during the gas transportation process pose serious problems to the availability, reliability, and economy of the pipeline. Thus, the possibly used safety measures focused on early detection of both small and large leakages is not only a guarantee for safer operation but also a help for reducing the costs of the industry. To monitor the valve yield condition for substantial cost savings and safer working conditions, a great number of methods have been developed. However, the currently available leakage detection methods provide little capability for the quantitative recognition of leakage levels.

No abstract

No abstract

AEThis paper presents an acoustical signal analysis scheme model for intelligent recognition of the leak level of a gas pipeline valve. The scheme is based on wavelet packet energy theory and a support vector machine (SVM) model. In this approach, the acoustical signal of the leak is obtained using an acoustic emission (AE) sensor. The energy of each node at the fourth level of the wavelet packet decomposed signal is extracted as a leak feature for the SVM classification process. SVM is applied to perform recognition of the leak level and the performance of the classification process due to the kernel function for the SVM and classifier is evaluated in terms of its accuracy, Cohen's kappa and training time. The experimental results demonstrate that the intelligent recognition model based on the wavelet packet energy feature parameter and SVM classifier (with linear kernel function) is optimal for recognising the leak level of a gas pipeline valve.