Pipeline Leak Localization Based on FBG Hoop Strain Sensors Combined with BP Neural Network

Jia, Zhang; Ren, Liang; Li, Hong‐Nan; Sun, Wei

doi:10.3390/app8020146

Cited by 50 publications

(30 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Benefitting from the recent rapid development of structural health monitoring technology [51,52], many researches demonstrated that piezoceramic materials [53,54] have actuating and sensing ability to detect structural damages. Future research will involve conducting experiments on the damage detection of buried long pipelines with advanced sensing technologies, such as fiber optic sensors [55,56] and piezoceramic transducers [57][58][59], also contributing to the pipeline monitoring. In addition, based on theoretical analysis and numerical simulations, simple design calculations can be proposed.…”

Section: Discussion Of Future Workmentioning

confidence: 99%

Experimental and Numerical Study on the Strain Behavior of Buried Pipelines Subjected to an Impact Load

et al. 2019

View full text Add to dashboard Cite

Long-distance oil and gas pipelines are inevitably impacted by rockfalls during geologic hazards such as mud-rock flow and landslides, which have a serious effect on the safe operation of pipelines. In view of this, an experimental and numerical study on the strain behavior of buried pipelines under the impact load of rockfall was developed. The impact load exerted on the soil, and the strains of buried pipeline caused by the impact load were theoretically derived. A scale model experiment was conducted using a self-designed soil-box to simulate the complex geological conditions of the buried pipeline. The simulation model of hammer-soil-pipeline was established to investigate the dynamic response of the buried pipeline. Based on the theoretical, experimental, and finite element analysis (FEA) results, the overall strain behavior of the buried pipeline was obtained and the effects of parameters on the strain developments of the pipelines were analyzed. Research results show that the theoretical calculation results of the impact load and the peak strain were in good agreement with the experimental and FEA results, which indicates that the mathematical formula and the finite element models are accurate for the prediction of pipeline response under the impact load. In addition, decreasing the diameter, as well as increasing the wall thickness of the pipeline and the buried depth above the pipeline, could improve the ability of the pipeline to resist the impact load. These results could provide a reference for seismic design of pipelines in engineering.

show abstract

Section: Discussion Of Future Workmentioning

confidence: 99%

Experimental and Numerical Study on the Strain Behavior of Buried Pipelines Subjected to an Impact Load

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Fiber Optic (FO) cable is one of the preferable choices for long distance because of its vulnerability to electromagnetic interference, high accuracies and multiplexibility. This advantages has inspired [27] to proposed leakage localization techniques by integrating Fiber Bragg Grating (FBG) and Background Propagation (BP) neural network classification algorithm. Method of Characteristics (MOC) is used to solve the leakage occurrences hydraulic transient process.…”

Section: Non-sensor Basedmentioning

confidence: 99%

Localization Techniques for Water Pipeline Leakages: A Review

Lah

Dziyauddin

Yusoff

2018

IJIE

View full text Add to dashboard Cite

IntroductionAn efficient water distribution system relies on a good water pipeline system. Water pipeline leakages are one of a few challenges to the water utility companies all over the world in providing the service. Several factors have been identified contributed to the water pipeline leakages such as ageing infrastructure, severe environmental conditions and third-party damages [1]. A recent report published by World Bank in 2016 indicates there is 45 billion cubic meters of water loss in that year. From a financial point of view this huge amount of losses is equivalent to a value of USD 3 billion whilst from a social point of view it can supply water to 90 million people in a year [2]. During pipeline leakage, a portion of the water delivered within the water distribution network wasted and this affects the Non-Revenue Water (NRW) efficiency to the water utility company. NRW is the lowest hanging fruits to improve water utility efficiency [3]. Minimizing the NRW is one of the challenges encountered by the water utility company because the uncontrolled NRW can cause vicious cycle [4]. Researchers in [5] studied the detail of pipeline leakages and emphasized that there are two main factors that affecting to the NRW efficiency; apparent losses and real losses. The apparent losses were due to meter defect, illegal tapping, water storage tank maintenance, fire brigade use and others. The real losses were due to a pipe burst, pipe leakages and water overflow due to faulty valves. A statistic from [4] in 2008 exhibits that real losses accounted for 61.5% of total water loss recorded. Hence there is an urgency to detect the pipeline leakage and localize it in the shortest time with leak location accuracy is the main benchmark.Pipeline leakage incidents sometimes go unnoticed until the symptoms are visible such as low water pressure or water supply interruptions. This happens when the water pipeline is located underground or far away from the community area. Even though the leakages is visible and noticed, it still takes time to the relevant authorities to locate the leak location. This issue has gained an interest in a lot of researchers and in the past few decades, numerous studies and research have been conducted to solve or at least minimize this issue. The research area in the field of localization can be partitioned into two categories; the external and internal system. The external system refers to the hardware-based such Abstract: Pipeline leakages in water distribution network (WDN) is one of the prominent issues that has gain an interest among researchers in the past few years. Time and accuracy play an important role in leak localization as it has huge impact to the human population and economic point of view. The complexity of WDN has prompt numerous techniques and methods been introduced focusing on the accuracy and efficacy. In general, localization techniques can be divided into two broad categories; external and internal systems. This paper reviews some of the techniques that has been explored and pr...

show abstract

“…The progress in neural networks has greatly stimulated research. Studies [24] have reported the automatic classification systems of defects, elimination of the adverse effects of environmental parameters in capacitive pressure sensors [25], and pattern recognition for pipeline leakage localization systems [26]. Neural networks trained by certain samples can process enormous amounts of data and recognize the patterns of test samples in a short period.…”

Section: Introductionmentioning

confidence: 99%

Identification of Crack Length and Angle at the Center Weld Seam of Offshore Platforms Using a Neural Network Approach

Qi¹,

et al. 2020

JMSE

View full text Add to dashboard Cite

The reconstruction algorithm for the probabilistic inspection of damage (RAPID) is aimed at localizing structural damage via the signal difference coefficient (SDC) between the signals of the present and reference conditions. However, tomography is only capable of presenting the approximate location and not the length and angle of defects. Therefore, a new quantitative evaluation method called the multiple back propagation neural network (Multi-BPNN) is proposed in this work. The Multi-BPNN employs SDC values as input variables and outputs the predicted length and angle, with each output node depending on an individual hidden layer. The cracks of different lengths and angles at the center weld seam of offshore platforms are simulated numerically. The SDC values of the simulations and experiments were normalized for each sample to eliminate external interference in the experiments. Then, the normalized simulation data were employed to train the proposed neural network. The results of the simulations and experimental verification indicated that the Multi-BPNN can effectively predict crack length and angle, and has better stability and generalization capacity than the multi-input to multi-output back propagation neural network.

show abstract

Pipeline Leak Localization Based on FBG Hoop Strain Sensors Combined with BP Neural Network

Cited by 50 publications

References 16 publications

Experimental and Numerical Study on the Strain Behavior of Buried Pipelines Subjected to an Impact Load

Experimental and Numerical Study on the Strain Behavior of Buried Pipelines Subjected to an Impact Load

Localization Techniques for Water Pipeline Leakages: A Review

Identification of Crack Length and Angle at the Center Weld Seam of Offshore Platforms Using a Neural Network Approach

Contact Info

Product

Resources

About