二元部分互溶液体闪点的支持向量机(SVM)预测与实验研究  

Prediction of Flash Points of Binary Partially Miscible Liquid Based on Support Vector Machine and Experiment Research

作　　者：陈莹[1] 蒋军成[1] 潘勇[1] 王睿[1] 

机构地区：[1]南京工业大学城市建设与安全工程学院,南京210009

出　　处：《工业安全与环保》2011年第3期1-2,共2页Industrial Safety and Environmental Protection

基　　金：国家自然科学基金项目(20976081)资助;高等学校博士学科点专项科研基金项目(200802910007)资助;江苏省自然科学基金项目(BK2009360)资助

摘　　要：建立了一个基于支持向量机的理论模型,用于预测二元部分互溶混合液体的闪点。根据所研究混合液体的物理性质,选择了纯物质的黏度、表面张力、配比、燃烧下限等物理参数来表征闪点,以这些参数作为输入参数,二元部分互溶液体的闪点作为输出值,应用支持向量机方法对两者之间的内在定量关系进行模拟。结果表明,闪点预测值与实验值符合良好。为工程上提出了一种预测二元部分互溶液体闪点的有效方法,可应用于评估混合溶液的火灾爆炸危害性及本质安全设计。A model based on support vector machine is established to predict the flash points of binary partially miscible liquid.Based on the physical properties of binary liquid,some physical parameters are used to represent flash point, such as viscosity,surface tension,mole fraction and lower flammable limit.Also,these physical parameters are used as input arguments,flash points of binary partially miscible liquid as output one and support vector machine is used to simulate the quantitative relationship between input arguments and output one.The result shows that the predicted flash points are in good agreement with the experimental value.The method proposed can be used to predict the flash points of binary partially miscible liquid for chemical engineering,which can be used to assess the dangers of fire and explosion of mixed-liquid and a safer design.

关 键 词：支持向量机 闪点 二元部分互溶液体 物理参数 

分 类 号：X932[环境科学与工程—安全科学]