基于偏最小二乘法的碳纤维导线拉断力预测  

Prediction of carbon fiber wire breaking force based on partial least square method

作　　者：徐志伟 XU Zhiwei(Yangzhou Haochen Electric Power Design Co.,Ltd.,Yangzhou 225000,China)

机构地区：[1]扬州浩辰电力设计有限公司,江苏扬州225000

出　　处：《黑龙江电力》2023年第1期82-85,共4页Heilongjiang Electric Power

摘　　要：为预测并调整碳纤维导线的性能,提出基于偏最小二乘回归的碳纤维导线拉断力预测模型。针对碳纤维导线材料构成与拉断力之间易发生多重共线性问题,采用主成分分析方法提取导线材料参数构成的综合变量,在此基础上建立以碳纤维束数、玻璃纤维束数、玻璃纤维带根数和铝线截面积为输入变量的拉断力预测模型。试验结果表明,碳纤维导线拉断力的预测相对误差不大于4%,说明预测值与实测值吻合度良好,可用于指导碳纤维导线研发和生产。In order to predict and adjust the performance of carbon fiber wire,prediction model of carbon fiber wire breaking force based on partial least squares regression is proposed.In order to solve the multicollinearity problem between the composition of carbon fiber wire material and the breaking force,the principal component analysis method is used to extract the comprehensive variables of the composition of wire material parameters.On this basis,the prediction model of the breaking force is established with the number of carbon fiber bundles,the number of glass fiber bundles,the number of glass fiber strips and the cross-sectional area of aluminum wire as the input variables.The test results show that the relative error of the prediction is less than 4%,indicating that the predicted value is in good agreement with the measured value,which can be used to guide the development and production of carbon fiber wires.

关 键 词：碳纤维导线 偏最小二乘法 主成分分析 拉断力 

分 类 号：TM751[电气工程—电力系统及自动化]