| 基于多特征融合与机器学习的井架钢结构智能诊断 |
| |
| 引用本文： | 黄岩,韩东颖,朱国庆,时培明.基于多特征融合与机器学习的井架钢结构智能诊断[J].科学技术与工程,2023,23(25):10800-10807. |
| |
| 作者姓名： | 黄岩 韩东颖 朱国庆 时培明 |
| |
| 作者单位： | 燕山大学车辆与能源学院;燕山大学电气工程学院 |
| |
| 基金项目： | 国家自然科学基金面上项目(51875500)；河北省自然科学基金面上项目(E2020203147)；河北省科学技术研究与发展计划中央引导地方专项 (216Z4301G)。 |
| |
| 摘 要： | 为了解决复杂的井架钢结构损伤识别问题，创新的将时域多参数信息融合与机器学习结合起来，根据井架钢结构的加速度响应信号，提出了一种基于主成分分析和支持向量机的井架钢结构损伤识别方法。首先利用加速度传感器提取在冲击载荷下井架钢结构加速度响应信号，进而获得其多个时域特征：脉冲因子、裕度因子和峭度；然后依据主成分分析法将以上特征进行融合，在保证尽可能多的信息被保留的情况下形成一个新的综合性特征；此时将该特征数据输入支持向量机模型进行损伤识别。利用以上理论进行仿真模拟计算和识别，同时利用ZJ70型井架钢结构实验室模型进行试验分析。结果表明：利用该方法识别井架钢结构单一或多处损伤准确率较高，并且简单易行耗时较少。
|
| 关 键 词： | 井架钢结构 损伤识别 主成分分析法 支持向量机 |
| 收稿时间： | 2022/10/1 0:00:00 |
| 修稿时间： | 2023/6/14 0:00:00 |
Intelligent Diagnosis of Petroleum Derrick Based on Principal Component Analysis and Support Vector Machine Analysis |
| |
| Huang Yan,Han Dongying,Zhu Guoqing,Shi Peiming.Intelligent Diagnosis of Petroleum Derrick Based on Principal Component Analysis and Support Vector Machine Analysis[J].Science Technology and Engineering,2023,23(25):10800-10807. |
| |
| Authors: | Huang Yan Han Dongying Zhu Guoqing Shi Peiming |
| |
| Abstract: | In order to solve the complex damage identification problem of derrick steel structure, an innovative combination of time-domain multi-parameter information fusion and machine learning is proposed. According to the acceleration response signal of derrick steel structure, a derrick steel based Structural damage identification methods. Firstly, the acceleration response signal of the derrick steel structure under impact load is extracted by the acceleration sensor, and then its multiple time-domain features are obtained: impulse factor, margin factor and kurtosis; A new comprehensive feature is formed under the condition of retaining as much information as possible; at this time, the feature data is input into the support vector machine model for damage identification. The above theory is used for simulation calculation and identification, and the laboratory model of ZJ70 derrick steel structure is used for experimental analysis. The results show that the method is used to identify single or multiple damages of the derrick steel structure with high accuracy, and it is simple and easy to implement and takes less time. |
| |
| Keywords: | derrick steel structure damage identification principal component analysis support vector machine |
|
| 点击此处可从《科学技术与工程》浏览原始摘要信息 |
|
点击此处可从《科学技术与工程》下载全文 |
