船舶结构海冰载荷的实船测量及反演方法研究

Field measurement and an identification method of sea ice load on ship structures

极地船舶通航中,海冰载荷是影响船舶安全航行的重要环境载荷,可造成船舶结构的严重损毁和疲劳破坏。由于海冰条件的差异和船体结构的复杂性,海冰荷载的实船测量是确定海冰载荷分布的关键途径。然而,船舶水线位置处往往因水密隔舱等密闭结构而不易安装传感器,这需要采用远场海冰载荷识别方法以克服传感器位置的限制。智能算法中的支持向量机方法具有小样本学习和强泛化性等算法特征,可适用于载荷识别过程中非线性映射关系的反演。为此,采用基于支持向量机方法的参数识别模型对船体结构上的海冰载荷进行远场反演分析。对船舶的海冰载荷反演结果在时域上很好地反映了船-冰碰撞特征,且该反演的海冰载荷与其他现场测量的冰载荷数值和离散性也比较接近,进而也验证了该海冰载荷识别方法的可靠性。基于支持向量机方法远场载荷识别方法是对传统船舶结构海冰载荷识别方法的扩展,有效地增加了海冰载荷的监测范围。

As a major environmental load during the navigation in a polar area,sea ice load can cause structural damage and fatigue damage to a ship structure.Due to the difference of sea ice conditions and the complexity of the ship structure,filed measurement is a key approach to determine the distribution of sea ice load on the ship hull.However,the installation of sensors around the waterline is not easy to operate due to waterproofed cabins.In this situation,a far-field ice load identification method is required to overcome the limitation of location of sensors.As an intelligent algorithm,the Support Vector Machine method has the small sample learning and the strong generalization characteristics in ice load identification which can be applied to determine the non-linear relationship between the sea ice load and the responses of ship hull.In this work,a far-field load identification model based on the Support Vector Machine method was established to determine ice load on the ship hull.The identified sea ice load of the ship hull in the present work can reflect the colliding characteristics between the ship hull and the ice in the time domain.Furthermore,the magnitude and the dispersion of identified ice load in this study are close to the results of other field tests which could verify the reliability of the identification model.The far-field sea ice load identification model based on the Support Vector Machine method is the extension of the traditional load identification model and it can widen the measurement field to an extent.