【摘要】 为了确保海洋平台作业过程中落物坠落对甲板撞击后的结构安全,对坠落立管撞击海洋平台甲板的过程进行了非线性仿真分析。运用ANSYS/LS-DYNA建立有限元模型,对坠落立管不同撞击角度撞击海洋平台甲板进行模拟计算,得到甲板在不同工况下的损伤情况。运用1种自适应性变异的粒子群优化算法SAPSO与BP神经网络结合[1],对海洋平台进行落物碰撞损伤分析。研究结果表明:落物与垂直方向偏离5°～10°为坠落时最危险的工况。SAPSO-BP提高了BP神经网络的拟合能力,减小了拟合误差,提高了拟合精度,验证了SAPSO-BP网络模型的实用性和可靠性。综合考虑制定适用于海洋平台落物安全的工作程序和平台及设备的防护措施,为海洋平台作业中落物风险评估和海洋平台作业安全保障提供参考。更多还原

【Abstract】 In order to ensure the structure safety of offshore platform after the impact of dropped object on deck during the operation process,the non-linear simulation analysis on the process of the dropped riser impacting the deck of offshore platform was carried out. A finite element model was established by using ANSYS/LSDYNA,and the simulation calculation on the dropped riser impacting the deck of offshore platform under different impacting angles was conducted,then the damage of deck under different conditions was obtained. The impacting damage of the falling object on the offshore platform was analyzed by using an adaptive mutation particle swarm optimization algorithm SAPSO combined with BP neural network. The results showed that SAPSO-BP improved the fitting ability of BP neural network,reduced the fitting error and improved the fitting accuracy. The practicability and reliability of SAPSO-BP network model were verified. Comprehensive consideration shall be given to the development of the working procedures and the protection measures of platform and equipments being applicable to the dropped objects safety of offshore platform,and it will provide reference for the risk assessment of dropped objects and safety guarantee of offshore platform operation.更多还原