Research On Calculating Method Of Subsea Wellhead Fatigue Life Based On Monitoring Data

This dissertation is supported by the project "Integrity technology and it`s engineering application for deepwater riser-subsea wellhead system in full life",which is a subproject of "Marine Deepwater Oil and Gas Field Development Engineering Technology" that is a project of national major science and technology projects in the 13 th Five-Year Plan.In accordance with the requirement of oil and gas exploration and development in the South China Sea,a research on the fatigue life calculation method of subsea wellheads based on monitoring data has been carried out.Using a combination of theoretical research,computer simulation,numerical calculation and machine learning algorithms,great progress has been made in research on the calculation method of subsea wellhead bending moment loads,the research on the prediction of the bending moment load of subsea wellheads based on neural networks,the calculation and prediction of subsea wellhead fatigue life.The relevant results can provide certain reference for the calculation and prediction of the fatigue life of subsea wellheads during drilling in the South China Sea.The main progress and results as follows.(1)The structure of the deepwater drilling riser system and subsea wellhead system was studied,and the stress conditions was analyzed A subsea wellhead fatigue life calculation method based on monitoring data was proposed.And the analysis steps was studied.The specific steps was studied.(2)Based on the transfer process of subsea wellhead bending moment load,two calculation methods of subsea wellhead bending moment,quasi-static method and dynamic method,were proposed,and the formula derivation process was given.The deepwater drilling riser-subsea wellhead dynamic analysis mathematical model was studied.The modeling method of the subsea wellhead equivalent model was studied,and a simplified coupling model of the riser-subsea wellhead system was established.The drilling riser-subsea wellhead finite element dynamic analysis was studied.The dynamic response was extracted based on the subsea wellhead bending moment calculation method,the subsea wellhead bending moment load under different methods was calculated and the results was compared.The the comparative analysis of the calculation results of the subsea wellhead bending moment was studied under different sea conditions.The most appropriate calculation method was selected,and the monitoring data was determined based on its dynamic response.(3)The prediction of subsea wellhead bending moment load based on dynamic neural network and the parameter setting of NARX neural network was studied.Monitoring data is used as input features.The prediction models of subsea wellhead bending moment loads under single sea state and complex sea conditions was established.The important parameters in the process of neural network training was presented through experiments.The training effect evaluation under the two models and the prediction results was compared and analyzed was studied,and the prediction model was determined.(4)The calculation method of subsea wellhead fatigue life was studied.Based on the subsea wellhead bending moment load predicted by the neural network,and combined with the M-N curve,the fatigue life of the subsea wellhead fatigue hot spots was studied.The calculation results of fatigue life under different operating conditions on the prediction model was studied.The subsea wellhead fatigue dynamic load database was established,and calculation research on subsea wellhead fatigue life prediction examples was studied.