| A pipeline system is a safe way of transporting oil and gas from one place to another,although sometimes suffer from leaks due to external and internal factors and fail to deliver the required products to the customers.The internal and external factors are crack,corrosion,sabotage,underground movements,earthquakes,human activities,pipe breakdown,etc.Lack of real-time small and larger gas leakage pipeline detection systems leads to financial,environmental,and health problems due to explosion when leakage happens.A gas pipeline leak detection system is necessary to be implemented along with the pipeline network for online real-time pipelines monitoring to ensure that the pipelines are delivering the products from one station to another at optimum conditions.Many pipeline leak detection systems have been implemented using sensors and algorithms.This is because,when gas or liquid energy is rapidly released from the pipe through a leak point,it generates acoustic wave signals which cause the changes of different physical characteristics like pressure,temperature,sound,and amplitude of acoustic waves alongside the pipeline.These physical characteristics can be detected and localizing the leakage.However,larger leakage can be easily detected while small leakage may not be detected for some time,especially when a pipeline is buried in remote places or under ice coverage.The acoustic wave signal of gas pipeline leakage has an advantage in short delay time and carries physical characteristics of the leaking gas.For a long-distance gas pipeline,sensing of physical characteristic signals has important meaning to the research of acoustic wave detection technology for the fact that the variation of acoustic wave signal can be detected by sensors installed on the pipeline.In this research,a literature review,numerical simulations,and experimental studies have been conducted for real-time natural gas leakage detection and localization based on acoustic waves.Surface acoustic wave sensors are used for detecting the nonlinear and non-stationary wave signals of the leaking gas.Since the detected acoustic wave signals are nonlinear and non-stationary,the Hilbert–Huang transform(HHT)algorithm is used for signal processing.Also,the leak-acoustics calculation equation and the attenuation model are obtained,which are applied for leak detection and location.Therefore,the innovation of the combination of the surface acoustic wave sensor and the Hilbert-Huang Transform has been successful for detecting,analyzing,and localizing small and larger gas leakages along the pipe.The main work and conclusions can be clearly explained as follows:(1)The essential gas leakage detection system requirements,principles,terminologies,and risks due to failure of the natural gas pipeline system are discussed while proposing a surface acoustic wave sensor with a combination of the Hilbert-Huang transform(HHT)as the best method for a real-time leak detection system for nonlinear and non-stationary acoustic wave signals for the gas pipeline network.The acoustic emission principle of the pipeline leakage and the physical characteristic of the pressure,temperature,sound,and amplitude of acoustic wave sources before and after the leak are first researched and analyzed.(2)The research was done by surveying the Tanzania Petroleum Development Corporation(TPDC)Natural Gas pipeline of 542 km long and the design for remote leakage detection and localization using surface acoustic wave sensor was successfully established.The simulation model was designed through OLGA software.The simulation results show that due to variation of pressure,temperature,and leakage gas flow rate when the leakage happens on the pipeline,the surface acoustic wave sensor can be implemented to detect those parameters.Therefore,since the TPDC natural gas pipeline has no remote leak detection system,it is suggested that this system should be implemented so that to enhance the safety of the pipeline network.(3)Theoretical formula and numerical simulation model to revealing the physical characteristics of leaking acoustic wave signals are established.The PROTEUS software was used to design a remote real-time leak detection system.The codes are written separately through Mikro C language and then loaded to Microcontroller and global system for mobile communication in PROTEUS software.The simulation was done and the surface acoustic sensor linked with the remote terminal unit(RTU)was used to detect the acoustic wave signal.The detected acoustic signals were analyzed with the Hilbert–Huang Transform(HHT)algorithm for leak location with the help of a cross-correlation algorithm.Therefore,the RTU with the help of the global system for mobile communication(GSM)broadcasts the detected signal to supervisory control and data acquisition(SCADA)as well as the mobile unit.(4)The leakage acoustic wave propagation formula under the gas flow condition in the gas pipeline,a theoretical model of the leakage acoustic wave amplitude attenuation,and an experimental fitting model is established.A leak location method based on the acoustic wave amplitude attenuation model is proposed.With original signal and HHT features,the location method based on the acoustic wave amplitude is used to calculate the leakage location accuracy,and compared with the traditional location method.A small gas pipe section of 40 m length,20 mm internal diameter,and two valves as the artificial leaks have experimented on the laboratory scale.The leakage detection and location method based on an acoustic wave with an improved HHT method and the propagation model were suggested and designed.When the two key points are established,the experiential equipment is constructed on a laboratory scale and the experiments are conducted.The TD is calculated,the propagation model and the leakage localization methods are also verified.The results of the experiments are discussed and analyzed indicate:(i)through regulating the valves diameters different acoustic wave physical characteristics are achieved and detected with sensors.(ii)The results have shown that all the changes of physical characteristics are detected,the error obtained by the improved method is between-0.30% and 0.58% and distributed uniformly near the zero lines while the error obtained by the HHT method is located between-4.26% and 13.66%.The average value and standard deviation of the improved method were-0.150% and 0.879% respectively,while those for the HHT method were 1.160% and 6.893%respectively.These results verify that the improved HHT method is superior to the HHT method for time difference(TD)calculations.Therefore,it is concluded that the new methods are efficient,can protect and monitor natural gas pipelines in real-time.The simulation and experimental test results demonstrate that this approach can improve the detection ability of small leakage with a low false alarm rate. |
PDF Full Text Request