摘要
为提高船体舭部圆弧板的数据拟合精度、工程测量效率和准确性,在采集三维位置数据的基础上,对数据进行多项式曲面拟合,使用基于最小方差约束的粒子群算法对拟合轴线进行优化。在仅对数据进行多项式曲面拟合后,数据模拟对比发现数据点误差多集中于10%,最大误差可达30%,无法满足实际工程使用。采用基于最小方差约束的粒子群算法对圆弧轴线进行优化,可将最大误差缩减至3%。研究表明:基于最小方差约束的粒子群算法可显著提高数据分析的精度,计算得到的圆弧拟合曲线可满足实际需要。
In order to improve the data fitting accuracy of hull bilge arc plate,the engineering measurement efficiency and accuracy,on the basis of collecting data in three-dimension coordinate,polynomial surface fitting is carried out for the data,and particle swarm optimization based on minimum variance constraint is used to optimize the fitting axis.After fitting the data with polynomial surface only,it is found that the error of data points is mostly concentrated in 10%and the maximum error can reach 30%through data simulation and comparison,which can not meet the requirements of practical engineering.The maximum error is reduced to 3%after the optimization of the arc axis by introducing particle swarm optimization based on the minimum variance constraint.The results show that the particle swarm optimization based on the least variance constraint can significantly improve the data analysis accuracy,and the arc fitting curve obtained can meet the practical needs.
作者
乔元英
李昂振
王大伟
孙琳琳
何珍
QIAO Yuanying;LI Angzhen;WANG Dawei;SUN Linlin;HE Zhen(Dalian Shipbuilding Industry Co.,Ltd.,Dalian 116021,Liaoning,China)
出处
《船舶标准化工程师》
2022年第1期29-34,共6页
Ship Standardization Engineer
关键词
粒子群优化
圆弧板
曲率检测
particle swarm optimization
arc plate
curvature detection
