摘要
针对目前起重机起升机构安全评价中评价指标多、评价过程烦琐的特点,本文利用SVM(支持向量机)算法理论并结合MATLAB编程工具建立起重机起升机构安全评价系统。该系统首先对多年现场检测并经过量化后的评价指标样本进行训练得到映射模型后,然后对测试样本进行预测。实验结果表明,SVM针对起升机构安全评价中小样本、非线性的特征具有高可靠性,从而为起重机安全评价提供了新思路。
In view of the characteristics of many evaluation indexes and cumbersome evaluation process in safety evaluation of crane hoisting mechanism at present,in this paper,the safety evaluation system of crane hoisting mechanism is established by using the theory of SVM(Support Vector Machine)algorithm and the programming tool of MATLAB.In this system,first,the evaluation indexes samples that field-tested for many years and quantified are trained to obtain the mapping model,then,the test samples are predicted.The results show that SVM has high reliability for the small sample and non-linear characteristics of hoisting mechanism safety evaluation,so it provides a new idea for safety evaluation of crane.
作者
安昆
孙刚
An Kun;Sun Gang(Shaanxi Special Equipment Inspection and Testing Institute,Xi'an 710048)
出处
《中国特种设备安全》
2019年第12期31-34,共4页
China Special Equipment Safety
关键词
起重机起升机构
安全评价
评价指标
SVM
映射模型
Crane hoisting mechanism
Safety evaluation
Evaluation indexes
SVM
Mapping model
