通航船舶吃水实时检测系统数据处理方法研究

Research on data processing method of real-time detection system for dynamic ship draft

根据通航船舶吃水检测要求,研制了基于超声波传感器阵列的船舶动态吃水实时检测系统,获取了船底吃水的点云数据。分析点云数据突变点的分布特征和现场干扰因素,将异常数据分为2类。并提出将点云数据中值滤波算法应用于消除超声波传感器探测船底时"第1类异常数据"的干扰;针对"第2类异常数据"出现在船尾的特点,提出了最小二乘直线回归预测法消除船只螺旋浆尾流干扰。并采用现场实测数据,对算法进行了验证。现场实测表明,检测系统满足测量要求,采用的算法实用、有效。

A real-time detection system of dynamic ship draft based on ultrasonic sensor array is researched and developed according to the detection requirements of navigation ship draft. The point cloud data of ship draft is obtained using this system. The abnormal data in the point cloud are grouped into two categories on the basis of the distribution features and field interference source. The median filtering algorithm for the point cloud data is adopted to eliminate the interference of the first class abnormal data when ultrasonic sensors are used to detect the bottom of the ship. Because the second class abnormal data only appears in the ship stem, the least-square linear regression and prediction algorithm is put forward to eliminate the interference caused by the propeller wake. The two algorithms are verified with the field data. Field tests show that the proposed detection system meets the measurement requirements and the two algorithms are practical and effective.