Damage Detection Of Offshore Platform Structures Based On Neural Network And Wavelet Analysis

Large civil engineering structures (for example, offshore platforms) are prone to be damaged during their service life caused by factors such as corrosion, impact and sea environment loads, so the occurrence of damage during the life of an offshore structure is inevitable. Once the damage is accumulated to some extent that leads the structure to collapse, there will be great losses to the life and property. Therefore, it is essential to detect the structural damage.The damage detection method based on structural vibration response and systemic dynamic parameters is a growing interest problem in nondestructive evaluation. Although the method has been applied in engineering to some extent, detecting structural damage and identifying damaged elements in a large complex structure are still challenging tasks since the in situ measured data of large civil engineering structures such as offshore platforms are inaccurate (owing to noise disturbance) and often incomplete (for economy consideration). After reviewing the past works, 3 tasks of researches are developed in this dissertation as following:1. A method for multiple steps damage detection of offshore platform by artificial neural networks is proposed. Damage detection is divided into three steps. Firstly, the located direction of damaged member is determined by radial basis function neural network with input of the change rate of normalized modal frequency. Secondly, the profile and layer of the damaged member is also determined by probabilistic neural network with input of the normalized damage-signal index. Finally, the damage extent is determined by the back-propagation neural networks with input of the change rate of squared modal frequency. In order to check the validity of methods for multiple steps damage detection of offshore platform by artificial neural networks, numerical simulation, impact model test and shaking table model test are carried out. The results of numerical simulation and model tests show that only few low modal parameters are needed to execute the damage detection of offshore platform by the proposed method, which has the ability to resist the modal parameter errors and noise disturbance.2. Because that the structural vibration response signal will change after damage, the...