【作者】 刘子祥； 李胜忠； 赵峰；

【Author】 LIU Zi-xiang;LI Sheng-zhong;ZHAO Feng;China Ship Scientific Research Center;

【机构】 中国船舶科学研究中心；

【摘要】 本文提出了一种新型的水面舰船阻力预报方法。该方法基于大样本数据库,利用基于决策树(Decision Tree)的随机森林(Random Forest)和梯度提升机(Gradient Boosting Tree)算法及动态样本(Dynamic Samples)方法建立动态化的预报模型,可对高速排水型舰船的船体阻力进行预报。应用该方法,对随机抽选的多艘船型进行预报测试,测试结果表明该方法具有较高的精度,具有较好的工程实用性。更多还原

【Abstract】 In this paper,a new method for resistance prediction of naval ships was present.This new method is based on large sample database,with random forest,gradient boosting tree and dynamic samples.This method could establish dynamic prediction models and predict the hull resistance of conventional naval ships like destroyers and frigates.This method is tested with multiple ships randomly selected from the database,and the result shows that the method has good accuracy and practicability.更多还原