数据挖掘的高动态范围船舶横摇预测

Prediction of ship rolling in high dynamic range based on data mining

舰船在海面上行驶时会受到海浪和洋流等环境扰动产生首摇横摇等各种摇荡。传统的横摇预测方法是通过传播角度传感器进行测量预测,但是预测的结果时效性差,预测的结果往往滞后于船舶的运动,因此如何快速准确的预测舰船横摇运动让舰船设计人员和专家伤透了脑筋。就上述现象设计了1套在数据挖掘条件下基于卡尔曼滤波技术的高动态范围船舶横摇预测系统。在给定具体海况的情况下对给定船只进行了船舶横摇运动仿真预测,得到随机海浪仿真倾角曲线,输入到微机系统后在对船舶横摇运动角速度进行测量,同时对环境干扰采用白化处理。最后,运用卡尔曼滤波技术对船舶进行半物理仿真预测,得到船舶横摇预测结果。通过对预测结果进行分析,基于卡尔曼滤波技术的高动态范围船舶横摇预测系统可以准确的对船舶横摇进行预测,降低了预测的滞后性,具有科学性和可行性。

When a ship is traveling on the sea surface, it will be disturbed by waves, ocean currents and other environmental disturbances to produce various kinds of swings such as first roll and so on. The traditional rolling prediction method is to measure and predict through the propagation angle sensor, but the prediction result is not timely, and the prediction result often lags behind the ship′s motion. therefore, how to predict the ship′s rolling motion quickly and accurately is a headache for ship designers and experts. A high dynamic range ship roll prediction system based on Kalman filter technology under the condition of data mining is designed for the above phenomenon. Under the given specific sea condition, the ship rolling motion is simulated and predicted for a given ship, and a random wave simulation inclination curve is obtained. after being input into a microcomputer system, the ship rolling motion angular velocity is measured, and the environmental interference is whitened at the same time. Finally, the Kalman filter technology is used to carry out semi-physical simulation prediction on the ship, and the ship roll prediction results are obtained. By analyzing the prediction results, the high dynamic range ship roll prediction system based on Kalman filter technology can accurately predict the ship roll, reducing the lag of prediction, which is scientific and feasible.