摘要
真空玻璃传热过程是非线性复杂的系统。为了研究真空玻璃的保温性能,提出一种基于模糊信息粒化和LSSVM真空玻璃保温性能预测研究的智能检测方法。根据工业现场采集数据,考虑真空玻璃传热过程的选择透过性,将采集的多元样本数据进行模糊粒化处理,提取各窗口有效的分量信息,建立基于最小二乘支持向量机的真空玻璃保温性能的预测模型,实现对真空玻璃非热源一侧温度平均值和波动范围的联合预测。利用自适应模糊粒子群算法进行迭代,获取更优的模型参数,提高模型的性能。研究结果表明:预测结果在0℃~0.5℃,在一定波动范围内,能够有效预测真空玻璃的保温性能。
Vacuum glass heat transfer process is a nonlinear complex system. In order to study the thermal insulation performance of vacuum glass,this paper presents an intelligent detection method based on fuzzy information granulation and LSSVM vacuum glass thermal insulation performanceprediction. According to the data collected at the industrial field, considering the selective transmission of the vacuum glass heat transfer process,the collected multi-sample data is subjected to fuzzy granulation,and in each window,the active ingredient information is extracted,and then a prediction model of thermal insulation performance of vacuum glass based on the least squares support vector machine is established. The adaptive fuzzy particle swarm optimization algorithm is used to obtain the optimal model parameters,and the model is optimized to realize the joint prediction of the temperature mean and fluctuation range of the non-heat source side of the vacuum glass. The experimentalresults show that the prediction method can effectively predict the thermal insulation performance of vacuum glass in the range of 0 ~ 0. 5 degrees.
出处
《广西大学学报（自然科学版）》
CAS
北大核心
2017年第6期2230-2238,共9页
Journal of Guangxi University（Natural Science Edition）
基金
国家自然科学基金资助项目(61463011)
国家重点研发计划课题项目(2016YFC0700804)
南海海洋资源利用国家重点实验室(海南大学)开放课题项目(2016010)
关键词
真空玻璃
保温性能
模糊粒化
最小二乘支持向量机(LSSVM)
vacuum glass
fuzzy granulation
thermal insulation performance
Least Squares Sup-port Vector Machine （LSSVM）
