| 基于AFSA-SVM的船体外板变形预测 |
| |
| 引用本文： | 董 梁,何祖军,齐 亮,葛成威.基于AFSA-SVM的船体外板变形预测[J].教育技术导刊,2009,19(10):129-132. |
| |
| 作者姓名： | 董 梁 何祖军 齐 亮 葛成威 |
| |
| 作者单位： | 1. 江苏科技大学 电子信息学院；2. 江苏科技大学 海洋装备研究院，江苏 镇江 212003 |
| |
| 基金项目： | 国家自然科学基金项目（51875270） |
| |
| 摘 要： | 船体外板复杂曲面自动化加工一直是船舶制造业研究热点和难点，由于加工过程中船板变形影响因素过多，导致船板加工变形预测一直不够准确快速。鉴于此，将两种复合变量用来表征加工过程中热源对应的众多加工参数，采用人工鱼群算法（AFSA）优化的支持向量机（SVM）预测船板变形。经实验验证，复合参数输入的AFSA-SVM模型预测船体外板水火线加热工艺变形线平均精确度为99.87%，角变形平均精确度为99.53%，且全局最优。将其与传统的PSO-SVM模型对比，不仅精确度有了提高，而且避免了局部极值导致的部分预测结果误差过大情况。
|
| 关 键 词： | 人工鱼群算法 支持向量机 船体外板 有限元仿真 |
| 收稿时间： | 2020-03-19 |
Deformation Prediction of Ship Plate Based on AFSA-SVM |
| |
| DONG Liang,HE Zu-jun,QI Liang,GE Chen-wei.Deformation Prediction of Ship Plate Based on AFSA-SVM[J].Introduction of Educational Technology,2009,19(10):129-132. |
| |
| Authors: | DONG Liang HE Zu-jun QI Liang GE Chen-wei |
| |
| Institution: | 1. Telecommunications College， Jiangsu University of Science and Technology； 2. Marine Equipment Research Institute， Jiangsu University of Science and Technology， Zhenjiang 212003，China |
| |
| Abstract: | The automatic machining for the complex surface of hull plate has always been a hot spot and difficult point in the shipbuilding industry. Due to the various factors affecting the deformation process for the ship plate in processing， the deformation prediction of the ship plate has not been accurate and fast enough. In view of this， the two composite variables are used to characterize many processing parameters corresponding to the heat source in the processing process， and the support vector machine （SVM） optimized by artificial fish swarm algorithm （AFSA） is used to predict the deformation of the ship plate. Experimental results show that the average accuracy of linear deformation and angular deformation for the AFSA-SVM model based on the composite parameters input are 99.87% and 99.53% respectively， and it is global optimization. By comparing it with the traditional PSO-SVM model， we find that not only the accuracy is improved， but also the large partial prediction error caused by local extremum is avoided. |
| |
| Keywords: | AFSA SVM ship’s hull plates FEA |
|
| 点击此处可从《教育技术导刊》浏览原始摘要信息 |
| 点击此处可从《教育技术导刊》下载免费的PDF全文 |
