In oil pipeline leak detection and location, noise in the pressure signal collected at the end of the pipeline affects the accuracy of leak detection and the error of leakage location. To reduce the noise interference, an improved local mean decomposition signal analysis method is proposed. The production functions (PFs) that are related to the leak signal can be exacted, and it is necessary to know the characteristics of leak signals or noise in advance. According to the cross-correlation function, there is a significant peak between the measured signals, which are decomposed into a number of PFs. These reconstructed principal PF components are obtained, and a wavelet analysis is used to remove the noise in the reconstructed signal. On this basis, the signal features are extracted according to the time-domain feature and the waveform feature, which are input into the least squares twin support vector machine (LSTSVM), to recognize pipeline leaks. According to the reconstructed signal after wavelet denoising, the time-delay estimate of the negative pressure signal at the end of the pipeline is obtained by the cross-correlation function, and the leak location is ultimately calculated by combining the time delay with the leak signal propagation velocity. A flow model for pipeline leakage is proposed based on the Flowmaster software, where the collected data of the different working conditions are processed. The experimental results show that the proposed method can effectively identify different working conditions and accurately locate the leakage point.

When an oil pipeline has small leaks, the change of the sound velocity of an ultrasonic signal cannot be easily detected. Based on the method of characteristics (MOC) theory, the leakage occurrence mechanism of a pipeline and the propagation mechanism of the sound velocity of the ultrasonic signal along the pipeline were studied. Additionally, the computational models of generating the sound velocity of the ultrasonic signal induced by pipeline leakage and calculating the propagation attenuation of the ultrasonic wave along the pipeline were established. A small leak localization method for pipelines based on information fusion is proposed, which is a combination of the sound velocity of the ultrasonic and flow rate signals. First, the fusion signals are extracted from the signal inflection point using a modified cross-correlation to determine the time difference of the fusion signal changes at the ends of the pipeline; then, the leak position is calculated based on the information of the pipeline length and the negative pressure wave propagation velocity. Since the change of the information fusion signal induced by a small leak is larger than the change of the sound velocity of the ultrasonic signal and the change of pressure signal, the proposed method can be used to locate a small leak in a pipeline. By analyzing the acquired data of a small leak in an experimental pipeline, the results show that the proposed method can quickly and effectively locate small leaks.

Pipeline transportation systems play an increasingly essential role in the development of the economy. An accurate location of leakage is important to reduce economic losses and maintain the safe operation of pipelines. In this article, a pipeline leakage detection method based on infrasound and negative pressure waves (NPWs) is adopted. This method can locate the leakage only by collecting the infrasound and pressure signals at the first station of the pipeline, which effectively solves the disadvantages of low positioning accuracy of a signal and easily produces false alarms. In this article, the signal denoising method of cross-spectral analysis combined with ensemble empirical mode decomposition (EEMD) was adopted. First, the leakage signals of sound waves and NPWs are processed by EEMD to obtain a series of intrinsic mode functions (IMFs). The characteristic frequency band is obtained by cross-spectral analysis of the leakage signal. Within this frequency band, the dominant component of the leakage source signal is divided by correlation analysis, and the IMF component is reconstructed to complete the signal denoising. When compared to the single signal pipeline leak detection method, the positioning accuracy of this article showed a 1–3 m improvement on average.

This article considers a multiple leaks' localization method in fluid pipelines based on the change in the sound velocity of an ultrasonic signal. First, the sound velocity of an ultrasonic signal is denoised by wavelet packet analysis (WPA). Second, the mathematical relationship between the position of a leak and the change in the sound velocity of an ultrasonic signal at the ends of a pipeline is established based on the motion equation and the continuity equation in the transient flow equation of a pipeline. As a consequence, the localization of the leakage position of the proposed method depends only on the change in the sound velocity of the ultrasonic signal at the ends of the pipeline rather than the time difference in a negative pressure wave arriving at the ends of the pipeline. Finally, the change in the sound velocity of the ultrasonic signal is used and it is proven to be efficient to locate the actual leak number. The simulation results show that the proposed method can locate multiple leak positions in a pipeline.

A method based on the multilayer feature fusion multiscale GhostNet (MFMSGN) is proposed to improve the accuracy of magnetic flux leakage (MFL) image recognition of pipeline corrosion defects. First, a multiscale ghost module is constructed to make it more suitable for practical applications, which can obtain a large number of multiscale features through a small amount of calculation. Then, a parallel structure is used instead of the traditional method of improving accuracy by deepening the model depth. In this article, we introduce an adaptive spatial feature fusion (ASFF) method to fuse features of different resolutions to widen the width of the network and apply it to downstream classification tasks. Finally, a lightweight and efficient model is proposed. The results of the experiment showed that the accuracy of the eight-layer MFMSGN for MFL corrosion defect identification is better than ResNet50 and slightly inferior to ResNet101, but the computational effort is much smaller.

To address the issue of multipoint leakage detection in energy transportation systems, a multiscale convolutional neural network based on kurtosis and KL divergence (KKL-MSCNN) was proposed for multipoint leakage detection in energy transportation systems. Initially, the collected infrasound data undergoes complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN). Furthermore, a hierarchical processing technique for intrinsic mode functions (IMFs) is proposed to reconstruct IMFs at two different feature levels. Following this, a lightweight multiscale convolutional neural network (MSCNN) is constructed, comprising two channels. The reconstructed IMF features are then extracted through serial and parallel convolution at varying scales, facilitating the completion of pipeline leakage level classification. In comparison to conventional methods, the proposed approach achieves a significant 20.26% increase in accuracy for multipoint leakage detection.