高级搜索

基于PSO-LSSVM算法的工业废气净化装置电源参数预测模型

,  , 

downloadPDF

上一篇

下一篇

荀倩, 王培良, 蔡志端. 基于PSO-LSSVM算法的工业废气净化装置电源参数预测模型[J]. 环境工程学报, 2016, 10(4): 1863-1868. doi: 10.12030/j.cjee.20160445
引用本文: 荀倩, 王培良, 蔡志端. 基于PSO-LSSVM算法的工业废气净化装置电源参数预测模型[J]. 环境工程学报, 2016, 10(4): 1863-1868. doi: 10.12030/j.cjee.20160445
shu
Xun Qian, Wang Peiliang, Cai Zhiduan. Power parameter prediction model for industrial waste gas purification device based on PSO-LSSVM algorithm[J]. Chinese Journal of Environmental Engineering, 2016, 10(4): 1863-1868. doi: 10.12030/j.cjee.20160445
Citation: Xun Qian, Wang Peiliang, Cai Zhiduan. Power parameter prediction model for industrial waste gas purification device based on PSO-LSSVM algorithm[J]. Chinese Journal of Environmental Engineering, 2016, 10(4): 1863-1868. doi: 10.12030/j.cjee.20160445
shu

基于PSO-LSSVM算法的工业废气净化装置电源参数预测模型

  • 1.

    湖州师范学院工学院, 湖州 313000

  • 基金项目:

    国家自然科学基金资助项目(61573137)

    浙江省公益性技术应用研究计划项目(2014C31091)

  • 中图分类号: X701.7

Power parameter prediction model for industrial waste gas purification device based on PSO-LSSVM algorithm

  • 1.

    School of Engineering, Huzhou University, Huzhou 313000, China

  • Fund Project:

  • 摘要: 为提高工业废气去除率与净化效率,针对传统净化装置中高频高压电源的输出电压幅值、频率参数不能随废气种类、流量和浓度进行在线调整而造成电能利用率降低的问题,提出一种基于PSO-LSSVM多元回归预测算法的工业废气净化装置电源参数预测模型。根据电源参数及其影响因素,将采集到的历史数据样本分为建模数据样本和实验数据样本,对废气净化装置的有关参数优化协调设置。为克服最小二乘支持向量机(LSSVM)对人为经验选择学习参数的依赖问题,采用粒子群优化算法(PSO)确定惩罚因子C和核函数参数σ2。结果表明,基于PSO-LSSVM的电源参数预测模型具有较高的精确度,可以真实反映电源参数随废气形式的变化。
    Abstract: To improve the removal efficiency of industrial waste gas, a novel power supply parameters prediction model based on the particle swarm optimization-least squares support vector machine (PSO-LSSVM) multivariate regression algorithm has been proposed. This new model aims for a low power utilization ratio resulting from the unadjusted online power parameters with the type, flow, and concentration of the waste gas in the traditional purification device. According to power parameters and influence factors, the collected historical data sample was divided into a modeling data sample and an experimental data sample to optimize and coordinate the relevant parameters of the purification device. In order to overcome the dependence problem of the LSSVM on selecting human learning parameters, PSO was used to determine the penalty factor C and kernel function parameter σ2. The experimental results showed that a power parameters prediction model based on PSO-LSSVM has a high accuracy rate, which can lead to a true reflection of power parameters varying with the form of waste gas.
  • 加载中
  • [1] 马竞涛,周则飞,吴祖成,等.低温等离子体处理恶臭 废气技术的工业应用研究.油技术与工程,2007,37(4):50-54 Ma Jingtao,Zhou Zefei,Wu Zucheng,et al.Study on commercial application of low-temperature plasma in treating fouling waste gas.Petroleum Refinery Engineering,2007,37(4):50-54(in Chinese)
    [2] 单志龙,刘兰辉,张迎胜,等.一种使用灰度预测模型的强自适应性移动节点定位算法.电子与信息学报,2014,36(6):1492-1497 Shan Zhilong,Liu Zhilan,Zhang Yingsheng,et al.A strong self-adaptivity localization algorithm based on gray prediction model for mobile nodes.Journal of Electronics & Information Technology,2014,36(6):1492-1497(in Chinese)
    [3] 段其昌,饶志波,黄大伟,等.基于EMD和PSO-SVM的电力系统中期负荷预测.控制工程,2012,19(5):913-917 Duan Qichang,Rao Zhibo,Huang Dawei,et al.Power system middle-term load forecasting based on EMD and SVM with PSO.Control Engineering of China,2012,19(5):913-917(in Chinese)
    [4] 王江荣.基于遗传算法模糊多元线性回归分析的瓦斯涌出量预测模型.工矿自动化,2013,39(12):34-38 Wang Jiangrong.Gas emission prediction model based on genetic algorithm and fuzzy multivariate linear regression analysis.Industry and Mine Automation,2013,39(12):34-3(in Chinese)
    [5] 王少军,刘琦,彭宇,等.CLS-SVM:一种时间序列预测的局部建模方法.仪器仪表学报,2011,32(8):1824-1829 Wang Shao,Liu Qi,Peng Yu,et al.CLS-SVM:A local modeling method for time series forecasting.Chinese Journal of Scientific Instrument,2011,32(8):1824-1829(in Chinese)
    [6] 梁栋,谢巧云,黄文江,等.最小二乘支持向量机用于时间序列叶面积指数预测.红外与激光工程,2014,43(1):243-248 Liang Dong,Xie Qiaoyun,Huang Wenjiang,et al.Using least squares support vector machines to estimate time series leaf area index.Infrared and Laser Engineering,2014,43(1):243-248(in Chinese)
    [7] 汤鸣.西安市工业废气排放总量的灰色预测模型.西安航空技术高等专科学校学报,2009,27(5):46-48 Tang Ming.Gray forecast module of total emission of industrial off-gas of Xi'an.Journal of Xi'an Aerotechnical College,2009,27(5):46-48(in Chinese)
    [8] 祝婕,袁新杰,马俊英,等.基于LMDI模型的乌鲁木齐工业废气排放影响因素研究.中国环境管理,2015,7(5):77-82 Zhu Jie,Yuan Xinjie,Ma Junying,et al.Research on factors of industrial waste gas emissions in Urumqi based on LMDI methods.Chinese Journal of Environmental Management,2015,7(5):77-82(in Chinese)
    [9] 张雪伍,蒋家超.基于GM-BP建模的城市工业废气排放量预测.常熟理工学院学报(自然科学),2014,28(2):107-111 Zhang Xuewu,Jiang Jiachao.Prediction of urban industrial emissions based on GM-BP neural network.Journal of Changshu Institute Technology (Natural Sciences),2014,28(2):107-111(in Chinese)
    [10] 何党联,崔虎,白文华.天然气净化厂LO-CAT装置参数优化及调整.石油化工,2014,33(9):118-121 He Danglian,Cui Hu,Bai Wenhua.Optimization and adjustment of parameters of LO-CAT device in natural gas purification plant.Petrochemical Industry Application,2014,33(9):118-121(in Chinese)
    [11] He Kang,Xu Qingsong,Jia Minping.Modeling and predicting surface roughness in hard turning using a Bayesian inference-based HMM-SVM model.IEEE Transactions on Automation Science and Engineering,2015,12(3):1092-1103
    [12] 代兴兰.遗传算法与最小二乘支持向量机在年径流预测中的应用.水资源与水工程学报,2014,25(6):231-235 Dai Xinglan.Application of genetic algorithm and least squares support vector machine in prediction of annual runoff.Journal of Water Resources Water Engineering,2014,25(6):231-235(in Chinese)
    [13] Sun Yongjun,Liu Yiwei,Liu Hong.Temperature compensation for a six-axis force/torque sensor based on the particle swarm optimization least square support vector machine for space manipulator.IEEE Sensors Journal,2015,16(3):798-805
    [14] Soeiro T.B.,Muhlethaler J.,Linner J.Automated design of a high-power high-frequency LCC resonant converter for electrostatic precipitators.Industrial Electronics,2013,60(11):4805-4819
  • 加载中
WeChat 点击查看大图
计量
  • 文章访问数:  2352
  • HTML全文浏览数:  1981
  • PDF下载数:  415
  • 施引文献:  0
出版历程
  • 收稿日期:  2016-01-04
  • 刊出日期:  2016-04-15
荀倩, 王培良, 蔡志端. 基于PSO-LSSVM算法的工业废气净化装置电源参数预测模型[J]. 环境工程学报, 2016, 10(4): 1863-1868. doi: 10.12030/j.cjee.20160445
引用本文: 荀倩, 王培良, 蔡志端. 基于PSO-LSSVM算法的工业废气净化装置电源参数预测模型[J]. 环境工程学报, 2016, 10(4): 1863-1868. doi: 10.12030/j.cjee.20160445
shu
Xun Qian, Wang Peiliang, Cai Zhiduan. Power parameter prediction model for industrial waste gas purification device based on PSO-LSSVM algorithm[J]. Chinese Journal of Environmental Engineering, 2016, 10(4): 1863-1868. doi: 10.12030/j.cjee.20160445
Citation: Xun Qian, Wang Peiliang, Cai Zhiduan. Power parameter prediction model for industrial waste gas purification device based on PSO-LSSVM algorithm[J]. Chinese Journal of Environmental Engineering, 2016, 10(4): 1863-1868. doi: 10.12030/j.cjee.20160445
shu

基于PSO-LSSVM算法的工业废气净化装置电源参数预测模型

  • 1. 湖州师范学院工学院, 湖州 313000
  • 收稿日期:  2016-01-04
基金项目:

国家自然科学基金资助项目(61573137)

浙江省公益性技术应用研究计划项目(2014C31091)

摘要: 为提高工业废气去除率与净化效率,针对传统净化装置中高频高压电源的输出电压幅值、频率参数不能随废气种类、流量和浓度进行在线调整而造成电能利用率降低的问题,提出一种基于PSO-LSSVM多元回归预测算法的工业废气净化装置电源参数预测模型。根据电源参数及其影响因素,将采集到的历史数据样本分为建模数据样本和实验数据样本,对废气净化装置的有关参数优化协调设置。为克服最小二乘支持向量机(LSSVM)对人为经验选择学习参数的依赖问题,采用粒子群优化算法(PSO)确定惩罚因子C和核函数参数σ2。结果表明,基于PSO-LSSVM的电源参数预测模型具有较高的精确度,可以真实反映电源参数随废气形式的变化。

English Abstract

    全文HTML

参考文献 (14)

返回顶部

目录

/

返回文章
返回

Copyright © 2019 中国科学院生态环境研究中心