Dynamic Performance Analysis Of Semi-submersible Platform Based On Prototype Monitoring Data

At present,the semi-submersible platform is widely used in the exploitation of deep-sea oil and gas,because of the complexity of the marine environment,the study of the platform movement performance is more and more attention by national and international experts.The research methods of semi-submersible platform motion characteristic have three kinds,numerical simulation method,model test method and prototype monitoring method.Three methods can verify each other to ensure the accuracy of the research.For the served many years semi-submersible platform,equipment adjustment,personnel turnover,maintenance work and other activities lead to platform parameters and design parameters have great deviation.Thus,platform response analysis and safety performance prediction with design parameters,cause the results of the study is not precise enough.So to seek a method of parameter correction becomes particularly important.The research of platform response prediction,personnel comfort analysis,safety performance evaluation under the extreme environment load,etc based on correction model can suggest a more accurate advice for the design of the platform and early warning.Based on the prototype monitoring data collected from the "Nanhai Tiaozhan" semi-submersible platform,this paper carries out platform parameter updating and analyzes the dynamic performance of the semi-submersible platform model under extreme load.Firstly,based on the measured datum about the marine environment and platform response,using Latin hypercube sampling method and Deep Belief Network built the neural network of platform parameter and the response.Then using this neural network got the updating parameters.With the response comparison about correction parameters,design parameters and measured,found that parameter correction model is more close to the actual platform motion.And using correction model assessed the design indicators of the platform under 100-year return period extreme environmental load.secondly,using the correction model put forward a platform response forecasting method.Choosing measured datum in different period verified this method.The result proved the feasibility of the prediction method.The relationship between the wave of information and personnel comfort is calculated and listed.According to the sea state classification,when the classification is moderate sea,platform staff will begin to feel a little uncomfortable.And when the classification is surge even very rough sea,the staff will feel quite uncomfortable.This is a reference for people if they need to evacuation from the platform.Finally,using correction model and measured datum,analyzed the mooring system damage under the extreme load condition.In addition,calculated the mooring system damage under the marine environment of normal working conditon.By the comparison of two situations,the results showed mooring system damage under the extreme load condition was larger than the normal conditon several times.When platform in extreme loads,mooring system safety performance must be considered.