Study On The Structural Damage Identification Of Offshore Platform Based On Principal Component Analysis Under Environmental Variations

The offshore platforms with the characteristics of complex structure,large size,long-term service in the harsh marine environment,are highly prone to damage.So it has important theoretical and practical significance to implement the health monitoring and damage identification of the offshore platform.The damage identification method based on vibration can identify the damage according to the change of the structural characteristic parameters pre and post damage,however,the changes of characteristic parameter induced by the environmental variations mask the changes of characteristic parameter induced by damage.Therefore,principal component analysis(PCA)is employed to identify the structural damage of offshore platform under the influence of environmental variations in this paper.The main contents are as follows:(1)The influence of environmental variations(temperature,quality)on the structural feature parameters is summarized.The methods used for eliminating the influence of environmental factors on structural damage identification are reviewed,and their advantages and disadvantages are also pointed out.(2)Study on structural damage identification based on Auto-Regressive(AR)model and PCA under the influence of environmental factors.Firstly,the AR model is used to fit the measured acceleration of different nodes,and the model coefficients are extracted,then,principal component analysis is used to project the standardized AR model coefficients into the low dimensional space which is more easily to be influenced by the environmental factors,after that,which are re-mapped back to the original space,the Q statistic is constructed in the original space,which is considered to have eliminated the influence of environmental factors,and used as the damage index.Finally,the statistical control chart is employed to identify structural damage.Numerical simulation and model experiment results of offshore platform show that the proposed method can effectively eliminate the influence of environmental factors on damage identification on the premise of reasonable selection of principal components,but,the identification effect is poor for damage case with relatively small damage extent.(3)Study on structural damage identification based on AR model and kernel principle component analysis(KPCA)under the influence of environmental factors.Firstly,the AR model is used to fit the measured acceleration of different nodes,and the model coefficients are extracted,then,kernel function is used to project the standardized AR model coefficients into the high dimensional feature space.The PCA is used to analyze standardized AR model coefficients in the high dimensional feature space,and Q statistic and score chart is constructed.According to the score chart,the separability of KPCA is better than that of PCA.Finally,the Q statistic is used as the damage index which is considered to have eliminated the influence of environmental factors,and the statistical control chart is used to identify structural damage.Numerical simulation and model experiment results of offshore platform show that the technique of KPCA not only eliminates the influence of environmental factors on damage identification,but also identifies the damage of the structure with a small damage extent.