摘要
为识别绿色能源产品太阳能光伏组件的复杂工序关键质量特性(critical quality characteristics,CTQ),提出基于状态空间算法(state space algorithm,SSA)构建质量关系模型.引入偏最小二乘算法(partial least-squares regression,PLSR),将其导入质量关系模型,进而建立复杂工序产品CTQ识别模型,最后应用变量投影重要性指标(variable importance in projection,VIP)验证提取的CTQ有效性和正确性.以太阳能光伏组件生产过程为例进行了实证分析.结果表明:新方法在克服各工序质量特性多重相关的情况下,不仅能有效地识别出光伏组件复杂工序输出的质量特性对终端产品质量的影响,还能提取出光伏组件复杂工序中关键质量特性;该方法的可行性得到了验证.
To identify the CTQ of solar photovoltaic modules complex process oriented to green energy products,the quality relational model was proposed based on state space algorithm. Partial least-squares regression( PLSR) was introduced and imported into quality relational model to establish CTQ identification model for complex process product. The variable importance in projection( VIP) was used to verify the validity and the correctness of critical quality characteristics. The productive process of solar photovoltaic modules was analyzed as an example. The results show that the new method not only can effectively identify the quality characteristic of solar photovoltaic module complex process output for the impact of terminal product,but also can extract the CTQ of solar photovoltaic module from discrete production process. The feasibility of the proposed method is verified.
作者
于影霞
盛敬峰
YU Yingxia SHENG Jingfeng(School of Mechanotronics and Vehic|e Engineering, East China Jiaotong University, Nanchang, Jiangxi 330013, China)
出处
《江苏大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2017年第2期197-203,210,共8页
Journal of Jiangsu University:Natural Science Edition
基金
国家自然科学基金资助项目(51265013)
关键词
绿色能源产品
太阳能光伏组件
偏最小二乘法
状态空间模型
关键质量特性
green energy product
solar photovoltaic modules
partial least-squares regression
state space model
critical quality characteristics