Random Study The Ultimate Strength Of Stiffened Plates With Pitting Corrosion Damage

During the service period of offshore structures,due to the influence of various corrosive environments such as electrochemistry and microbial erosion,the surface of stressed components is prone to irregular pitting corrosion,which in turn leads to the degradation of component strength.It is an important basis for structural design and maintenance to study the degradation law of ultimate strength of irregular pitting structure.In this thesis,by building a model of stiffened plate with random pitting damage,the effects of strip slenderness ratio,reinforcement slenderness ratio,height-thickness ratio of stiffened web,strip pitting damage and reinforcement pitting damage on the ultimate strength of the structure are studied.The relationship between random pitting intensity DOV and equivalent coefficient is studied by the method of thickness equivalence,and the functional expression of pitting intensity DOV and thickness equivalent coefficient is obtained.A BP neural network with the influence parameters of ultimate strength of stiffened plates as input vectors is constructed,and a neural network model with high prediction accuracy is obtained.The main work is as follows:(1)In this thesis,the finite element model of random pitting damage of stiffened plate,strip and stiffener is constructed,the construction method of random distribution of pit shape,size and location is determined,and the construction method of finite element model when random pitting damage of stiffened plate,strip and stiffener exists simultaneously is expounded.The application methods of three boundary conditions and the initial defects of the finite element model are determined,and the effectiveness of the finite element model in this thesis is verified by the existing literature.In addition,in the study of the influence of the aspect ratio of stiffened plate and strip,it is found that the aspect ratio of stiffened plate and strip has little influence on the ultimate strength of the structure,so the influence of the aspect ratio of strip can be ignored in the follow-up study.(2)The random pitting damage of stiffened flange plate,stiffened web plate and stiffened plate strip under three boundary conditions is equivalent to uniform corrosion.It is found that the ultimate strength degradation of stiffened flange plate is small,and the maximum difference of strength degradation under three boundary conditions and four pitting strengths is 5%,and the failure mode is consistent,so it is reasonable for stiffened flange plate to be equivalent to uniform corrosion.The ultimate strength degradation of stiffened webs and stiffened plates is quite different,which can reach 11.1% at most,and the failure modes are also different,so the random pitting equivalence of stiffened webs and stiffened plates is unreasonable.In the study of pitting damage on both sides of stiffened web,it is found that the difference of ultimate strength degradation between them is less than 1.7%,and the failure mode is the same.Because the eccentricity under unilateral pitting load is more representative,unilateral pitting of stiffened web is chosen as the research object.(3)When studying the influence of the slenderness ratio of stiffened plates and strips,the slenderness ratio of stiffeners,the height-thickness ratio of stiffened webs and the pitting corrosion strength DOV on the ultimate strength degradation of the structure,it is found that the increase of the slenderness ratio of strips will lead to the increase of the ultimate strength degradation of the structure,whether the pitting corrosion damage of stiffeners is considered or not.When the reinforcement slenderness ratio λ is 0.40,the ultimate strength of the structure deteriorates most seriously.When there is random pitting damage in the strip,the height-thickness ratio of the stiffened web is 32,and the ultimate strength of the structure is most seriously degraded.When there is pitting damage in both the strip and the reinforcement,the height-thickness ratio of the stiffened web is 58,and the ultimate strength of the structure is most seriously degraded.In addition,the pitting damage of reinforcement mainly leads to more serious ultimate strength degradation of the structure by changing the failure mode of the structure.(4)The artificial neural network ANN model is used to predict the ultimate strength of the structure for the pitting damage of the stiffened plate with strip and the pitting damage of the stiffened plate with strip and stiffener at the same time,and the parameters of the input layer and output layer of the artificial neural network ANN under the two kinds of pitting damage are determined,and the neural network model with reasonable number of neurons in the hidden layer is obtained by training,and the neural network model with high prediction accuracy is obtained by using the existing finite element results.By using the ANN model in this paper to predict the existing literature,it is found that the ANN model in this paper has good prediction accuracy,and the maximum error of predicting two existing literatures is less than 3.5%.