Based on numerical analysis and software development, a scouringidentification model was developed to identify the jack-up platformdamage according to the variation law of stress field in hot spotlocation of Jack-up’s pile legs and boots. Firstly, the whole finiteelement model including the jack-up platform and soil was constructedto accurately reflect the interaction between the pile leg and the soil. Onthe basis of investigating the relationship between the stress field andthe scour depth of Jack-up’s legs and boots, the scouring identificationmodel was built by applying neural network algorithm, and the nonlinearmapping relationship was established between the stress field andscour depth by training a large number of samples. Finally, through theMFC application program-Visual Studio, the visual software foridentifying the scour state of jack-up’s pile legs and boots wascompiled, which is convenient for engineering application. From thecomparative analysis between the results of numerical simulation andscouring identification model, it is concluded that the scouridentification model proposed in this paper has high accuracy and canprovide important reference for damage assessment of jack-up platform.
展开▼