Numerical study on mode curvature for damage detection of a drilling riser using transfer matrix technique

In order to prevent marine pipeline accidents caused by corrosion defect, crack or loose contact, the technique of non-destructive testing or examination (NDT or NDE) becomes great important in the flied of structural health monitoring (SHM) in ocean engineering. The objective of this study is to propose a numerical method to localize and size the structure damage of a drilling riser. One theoretical development is a simple and effective algorithm used for ocean risers’ mode extraction and damage detection based on the transfer matrix technique. The merits of the developed method are that damage of risers with different boundary conditions and variable cross-sections can be localized and sized. Additionally, the relationships between damage locations and modes are discussed by comparing mode shape difference, mode curvature and mode curvature difference. Numerical results show that mode curvature values of damaged elements have little influence on those of intact elements. Based on this fact, a damage index method is developed to evaluate damage severities and explore the influence of single and multiple damage severities on damage indexes. More importantly, the outcomes are verified to be accurate enough to indicate the feasibility and reliability of the numerical method by the standard FEM package.     

基于理论的气旋条件下双极化C波段合成孔径雷达波参数检测

Theoretical-based ocean wave retrieval algorithms are applied by inverting a synthetic aperture radar(SAR)intensity spectrum into a wave spectrum, that has been developed based on a SAR wave mapping mechanism. In our previous studies, it was shown that the wave retrieval algorithm, named the parameterized first-guess spectrum method(PFSM), works for C-band and X-band SAR at low to moderate sea states. In this work, we investigate the performance of the PFSM algorithm when it is applied for dual-polarization c-band sentinel-1(S-1) SAR acquired in extra wide-swath(EW) and interferometric wide-swath(IW) mode under cyclonic conditions.Strong winds are retrieved from six vertical-horizontal(VH) polarization S-1 SAR images using the c-band crosspolarization coupled-parameters ocean(C-3 PO) model and then wave parameters are obtained from the image at the vertical-vertical(VV) polarization channel. significant wave height(SWH) and mean wave period(MWP) are compared with simulations from the WAVEWATCH-III(WW3) model. The validation shows a 0.69 m root mean square error(RMSE) of SWH with a –0.01 m bias and a 0.62 s RMSE of MWP with a –0.17 s bias. Although the PFSM algorithm relies on a good quality SAR spectrum, this study confirms the applicability for wave retrieval from an S-1 SAR image. Moreover, it is found that the retrieved results have less accuracy on the right sector of cyclone eyes where swell directly affects strong wind-sea, while the PFSM algorithm works well on the left and rear sectors of cyclone eyes where the interaction of wind-sea and swell is relatively poor.     

Developing a robust SHM method for offshore jacket platform using model updating and fuzzy logic system

Structural monitoring is essential for ensuring the structural safety performance during the service life. The process is of paramount importance in case of the offshore jacket-type platforms due to the underwater structural parts subjected to the marine environmental conditions. This work is an experimental investigation on a laboratory model of a jacket platform with the objective of establishing a baseline finite element (FE) model for long-term structural health monitoring for this type of structures. A robust damage diagnosis system is also developed which is less sensitive to both the measurements and the modeling uncertainties. Experimental vibration tests are conducted on a physical platform model to obtain dynamic characteristics and then, the initial FE-model of the intact structure is developed to determine them numerically. Some differences between numerically and experimentally identified characteristics emerge due to various uncertainties in the FE-model and measured vibration data. To minimize these differences, initial FE-model is updated according to the experimental results. The updated FE-model is employed to predict the changes in the dynamic characteristics under variety of damage scenarios which are imposed by reducing the stiffness at the components of the model. Fuzzy logic system (FLS) and probabilistic analysis is developed for linguistic classification of damage and global damage diagnosis. Incorporation of the FLS fault isolation technique into FE-model updating method are proposed and evaluated for two different FLS methods to develop a vigorous damage diagnosis method. The efficiency of the technique is validated by different damage scenarios foreseen on the physical model. This technique is shown to be effective for diagnosing the presence of degradation and quantify it.     

激光单分子BTEX异常的人工神经网络评价技术

采用BP(Back-Propagation Network)神经网络方法,根据试验区BTEX指标实测数据,结合油气化探、地质、地球物理等资料建立BTEX异常综合评价指标体系及评分标准,完成BTEX异常的BP神经网络综合评价模型并对试验区进行含油气远景评价,研究结果表明该技术具有较好的应用前景。