江苏大学学报（自然科学版）2017，Vol.38Issue(1)：52-55,4.DOI:10.3969/j.issn.1671-7775.2017.01.010
多输出支持向量机醋糟气化建模
Modeling of multi-output support vector machine for vinegar residue gasification
摘要
Abstract
The support vector machine gasification model was established based on the vinegar residue gasification experiment.Gasification temperature,vinegar water content and mass ratio of biomass to water vapor were input into the model with the main gas group of H2 ,CH4 ,CO and CO2 as output to realize the output process from input.To test the data from vinegar residue gasification experiment,3 sets of data were selected as model test group with the remaining data as training sample.The vinegar residue gasification model was established by Matlab.The best penalty parameter C was 9.551 2 with optimal width parameter sigma of 92.723.After the completion of model establishment,three groups of data were picked randomly and verified,and the prediction of gas composition was compared with the experimental data.The results show that the average errors of H2,CO,CH4 and CO2 are respectively 2.87%,3.28%, 2.66% and 3.04%.关键词
醋糟气化/多输出支持向量机/模型/惩罚参数/宽度参数Key words
vinegar residue gasification/multi output support vector machine/model/penalty parameter/width parameter分类
能源科技引用本文复制引用
王立群,白文斌,许超杰,王谦..多输出支持向量机醋糟气化建模[J].江苏大学学报（自然科学版）,2017,38(1):52-55,4.基金项目
国家科技支撑计划项目（2015BAD21B00）；科技部农业成果转化项目2008GB2C100099）；江苏省科技支撑计划项目（BE200851）；江苏省机械清洁能源与应用重点实验室项目 （）
