A New Risk Assessment Method For Jacket Platform And Nuclear Power Plant Based On Global Sensitivity Analysis And Global Uncertainty Analysis

Faced with the influence of global climate change and other uncertainty factors, thispaper proposed a new risk assessment method for Jacket platform and coastal nuclearplant design. This method is based on Global Sensitivity Analysis (GSA), GlobalUncertainty Analysis (GUA) and Multivariate Compound Extreme Value DistributionModel (MCEVD).Due to the complexity and randomness of the natural events, there are definitelyvarious kinds of uncertainty factors in risk assessment and design for platforms. Thetraditional GUA method usually requires the calculation of all these uncertaintyfactors, which would cost a lot of time and energy. The new method this paperproposed can quantify all the significances among random variables and theirparameters at first by GSA. Then, by comparing the significant level of everyuncertainty factor, all minor factors would be negligible. This would definitelysimplify the global uncertainty analysis greatly and also influence the accuracy ofassessment results.By global uncertainty analysis, all the main uncertainty factors are taken into account,including environmental loads, marine corrosion and sea level rise. In this paper, themultivariate compound extreme value distribution model (MCEVD) is applied topredict the extreme sea state of wave, current and wind. The maximum structuralstress and deformation of a Jacket platform are analyzed and compared with differentdesign standards. The calculation result sufficiently demonstrates the new riskassessment method’s rationality and security.At last, Global Uncertainty Analysis and Double Layer Nested Multi-objectiveProbability Model are applied to check the design standard for Lingao coastal nuclearplant. The risk assessment result is more severe than calculated values by Chinadesign code. This study shows the necessity of risk assessment of coastal defenseagainst typhoon attacks for NPP.