The invention discloses a ship control motion forecasting method and system based on local Gaussian process regression, and the method comprises the steps: carrying out a clustering analysis algorithm on pre-collected ship motion data, dividing the data into a plurality of clusters, and determining a clustering center corresponding to each cluster; calculating the distance between the to-be-predicted sample and each clustering center, and obtaining the nearest cluster; and calculating a forecast result of the to-be-predicted sample on the nearest cluster. On the basis of the local data thought of clustering analysis, a ship motion data set is automatically divided into a plurality of clusters according to the similarity, the forecasting result of a to-be-predicted sample is calculated on the cluster with the closest center distance, calculation on the whole data set is not needed, forecasting can be accelerated, and certain forecasting precision can be guaranteed. Compared with a traditional GPR method, the method is higher in calculation efficiency, and the precision loss is not obvious.
本发明公开了基于局部高斯过程回归的船舶操纵运动预报方法及系统,将预先采集的船舶运动数据执行聚类分析算法,划分为若干个聚类,并确定每个聚类所对应的聚类中心;计算待预测样本与各聚类中心的距离,获取距离最近的聚类;在该距离最近的聚类上计算待预测样本的预报结果。基于聚类分析的局部数据思想,将船舶运动数据集按照相似度自动划分为若干个聚类,待预测样本的预报结果在中心距离最近的聚类上进行计算,而不必在整个数据集上进行计算,这样不仅可以加速预报,还能保证一定的预报精度。相比传统GPR方法不仅具有更高的计算效率,而且精度的损失也不明显。
Ship control motion forecasting method and system based on local Gaussian process regression
基于局部高斯过程回归的船舶操纵运动预报方法及系统
2023-11-10
Patent
Electronic Resource
Chinese
Gaussian process regression for forecasting battery state of health
Online Contents | 2017
|