欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
通过对仿真Z形试验数据的分析，应用具有不同不敏感因子占的占一支持向量回归机（Support Vector Regression，SVR）辨识了船舶操纵运动二阶线性响应模型中的K，T等操纵性指数，并利用回归得到的响应模型进行了Z形试验的数值模拟。通过比较采用不同不敏感因子占所得首向角和转艏角速度的预报结果，表明可以通过调节不敏感因子占值来控制样本输入中文持向量的个数与ε-SVR的回归精度。
英文摘要：
By analyzing the data from the simulated zigzag test, the parameters of the second-order linear response model are identified by applying ε-Support Vector Regression with different insensitive factors. Then the second-order linear response models are solved by Runge-Kutta method to simulate the zigzag manoeuvres and to predict the heading angle and yaw rate. The comparison of the prediction results obtained with different insensitive factors shows that by adjusting the insensitive factors, the ratio of the support vector to the training samples can be controlled to improve the identification accuracy of the ε -SVR.
同期刊论文项目
同项目期刊论文
Black-Box Modeling of Ship Manoeuvring Motion Based on Feed-Forward Neural Network with Chebyshev Or
Parametric Identification and Sensitivity Analysis for Autonomous Underwater Vehicles in Diving Plan
期刊信息
位置：