Machine learning methods for predicting the outcome of hypervelocity impact events
作者：
Highlights：
• Two machine learning techniques are applied to characterize hypervelocity impacts.
• Both techniques offered improved accuracy over conventional semi-analytical methods.
• Further improvements are restricted by the statistical quality of the training data.
• A novel approach is proposed for identifying ‘gaps’ in the training dataset.
• This approach is successfully demonstrated to improve the network predictions.
摘要
•Two machine learning techniques are applied to characterize hypervelocity impacts.•Both techniques offered improved accuracy over conventional semi-analytical methods.•Further improvements are restricted by the statistical quality of the training data.•A novel approach is proposed for identifying ‘gaps’ in the training dataset.•This approach is successfully demonstrated to improve the network predictions.
论文关键词：Hypervelocity impact,Artificial neural network,Support vector machine,Terminal ballistics
论文评审过程：Received 7 July 2015, Revised 21 September 2015, Accepted 23 September 2015, Available online 3 October 2015, Version of Record 16 October 2015.
论文官网地址：https://doi.org/10.1016/j.eswa.2015.09.038
