Research On Intrusion Risk Assessment Methodology And Protection Technology For Overseas Oil&Gas Fields

Chinese companies has become more and more active in overseas corporation since the "Belt and Road" initiative was put forward.In order to protect the personnel,assets,environment and reputation of overseas oil and gas field,the intrusion risk assessment methodology and protection technology are studied based on Analytic Hierarchy Process,Intrusion Risk Control Cost,Improved Wavelet Algorithm,Background Differencing,Convolutional Neural Network,Detection Accuracy Analysis,and Nonlinear Programming.The above mentioned studies are also experimentally applied in certain oil field in Middle East,achieving full coverage of intrusion detection,longitudinal-depth design of physical protection engineering,and systematic integration of emergency resources.The research mainly focusses on the following aspects.Intrusion risk factors influencing overseas oil and gas field are identified,and two tier indexes of intrusion risks are analyzed.The first-tire consists of 5 indexes,and the second-tier consists of 18 indexes.Analytic Hierarchy Process is applied on two tier indexes,and sensitivities are obtained.Five types of logical judgements are applied to develop the quantification methodology of intrusion risk level,whose results are verified through benchmarking with international security consultancy companies.In terms of the key indexes of intrusion risk,the principle of intrusion protection technology is studied.Firstly,the improved wavelet algorithm is studied and developed for the enhancing and denoising of infrared image.The low frequency coefficient of wavelet is enhanced through histogram equalization,and the high frequency coefficient of wavelet is denoised through threshold values.The definition and contrast ratio of infrared image are also improved.Secondly,the moving target detection algorithm based on background differencing is studied,achieving moving target positioning and improving detection ability.Thirdly,the target recognition methodology based on deep-learning is studied.The deep-learning is realized through Alexnet model in Matlab Convolutional Neural Network toolbox.Fourthly,the risk possibility and engineering cost of both side and front intrusion protection are studied.Intrusion risk probability,protection cost and risk control cost are obtained among different intrusion protection plans,realizing the optimization of intrusion protection.The intelligent intrusion detection system is analyzed and experimentally applied in certain oil field in Middle East.The system is systematically designed through network topology,consisting of infrared image collection and processing,intrusion target detection and recognition.Moreover,innovative deployment plan and accuracy analysis methodology of thermal infrared cameras is applied to avoid blind zone of intrusion detection.Through mix deployment of clockwise and anti-clockwise cameras,full coverage of security perimeter is achieved through only 14 infrared thermal cameras.Detection Rate for all test locations is 100%,and more than 90%of test locations are with high alarm accuracy(above 90%),and the best alarm distance is 175.72m.The infrared image collection and processing of intrusion target in the experimental study are of high accuracy,achieving the research target of intrusion detection.Intrusion delay and emergency response of certain oil field in Middle East are realized through physical protection and quick response technology.The physical protection engineering of both side and front intrusion are distinctively designed based on longitudinal-depth safeguarding.Nonlinear programming is conducted for the design of emergency center based on Ergonomics.Quick response of emergency forces is realized based on INC-MOD-MOS three-dimensional framework.The research targets of intrusion delay and rapid emergency response in the experimental study are achieved.