大型离心式通风机性能预测方法
|
摘要
针对现有离心式通风机性能预测方法不能充分利用离心式通风机历史运行数据、建模周期长等问题,提出了基于最小二乘支持向量机(LSSVM)与拉丁超立方采样(LHS)的大型离心式通风机性能预测方法。选取出口压力作为衡量通风机性能的指标,利用LSSVM建立离心式通风机性能预测模型;通过LHS方法采集离心式通风机的入口温度、入口压力、入口流量和转速,将采集的数据进行归一化处理后用于LSSVM模型的训练;通过测试数据验证所建立模型的有效性。仿真结果表明,基于LSSVM与LHS的大型离心式通风机性能预测方法能够充分利用已有通风机数据信息快速准确地预测通风机性能。
In view of problems that existing performance prediction methods for centrifugal ventilator cannot fully utilize historical operation data of centrifugal ventilator and have long modeling period,performance prediction method for large-centrifugal ventilator based on LSSVM and LHS was proposed.Outlet pressure is selected as index to measure performance of ventilator,and performance prediction model of centrifugal ventilator is established by using LSSVM.Inlet temperature,inlet pressure,inlet flow rate and rotational speed of the centrifugal ventilator are collected by LHS method,and the collected data are normalized for training of LSSVM model.Validity of the established model is verified by testing data.The simulation results show that the performance prediction method for large-centrifugal ventilator based on LSSVM and LHS can make full use of existing ventilator data information to quickly and accurately predict performance of ventilator.
In view of problems that existing performance prediction methods for centrifugal ventilator cannot fully utilize historical operation data of centrifugal ventilator and have long modeling period,performance prediction method for large-centrifugal ventilator based on LSSVM and LHS was proposed.Outlet pressure is selected as index to measure performance of ventilator,and performance prediction model of centrifugal ventilator is established by using LSSVM.Inlet temperature,inlet pressure,inlet flow rate and rotational speed of the centrifugal ventilator are collected by LHS method,and the collected data are normalized for training of LSSVM model.Validity of the established model is verified by testing data.The simulation results show that the performance prediction method for large-centrifugal ventilator based on LSSVM and LHS can make full use of existing ventilator data information to quickly and accurately predict performance of ventilator.
引文
[1]牛青.矿井通风机性能测试方法与系统的研究[D].西安:西安科技大学,2014.
[2]万雷.单级离心风机气动性能模拟与实验验证[D].哈尔滨:哈尔滨工程大学,2013.
[3]李超,伍继浩.离心通风机整机损失数学模型的研究及优化[J].风机技术,2000(4):15-17.
[4]姜华,宫武旗,张炜,等.带分流叶片离心叶轮非定常流场的实验研究[J].西安交通大学学报,2009,43(9):14-18.JIANG Hua,GONG Wuqi,ZHANG Wei,et al.Experimental investigation of the unsteady discharge flow in a centrifugal impeller with splitter blades[J].Journal of Xi'an Jiaotong University,2009,43(9):14-18.
[5]聂波,张俊林,满建楠.进口弯管对前向离心通风机性能影响的研究[J].流体机械,2018,46(1):6-9.NIE Bo,ZHANG Junlin,MAN Jiannan.Effect on the performance of the forward centrifugal fan with inlet bent pipe[J].Fluid Machinery,2018,46(1):6-9.
[6] ONMA P,CHANTRASMI T.Comparison of two methods to determine fan performance curves using computational fluid dynamics[C]//IOP Conference Series:Materials Science and Engineering,2018.
[7] MADHWESH N,KARANTH K V,SHARMA N Y.Effect of innovative circular shroud fences on a centrifugalfanforaugmentedperformance-A numerical analysis[J].Journal of Mechanical Science and Technology,2018,32(1):185-197.
[8]黄忠文,王培,韩海燕.离心通风机叶轮的有限元建模与应力分析[J].流体机械,2015,43(10):27-30.HUANG Zhongwen,WANG Pei,HAN Haiyan.Finite element modeling and stress analysis of centrifugal fan impeller[J].Fluid Machinery,2015,43(10):27-30.
[9]邵国建,陈帮,谢云川.进气箱对离心风机性能影响的研究[J].风机技术,2018,60(1):39-43.SHAO Guojian,CHEN Bang,XIE Yunchuan.Effect of inlet box on the performance of centrifugal fan[J].Chinese Journal of Turbomachinery,2018,60(1):39-43.
[10]张玮.基于神经网络的数据统计建模[D].杭州:浙江工业大学,2009.
[11] VAPNIK V N.An overview of statistical learning theory[J].IEEE transactions on neural networks,1999,10(5):988-999.
[12] CHERKASSKY V,MA Y.Practical selection of SVM parameters and noise estimation for SVM regression[J].Neural Networks,2004,17(1):113-126.
[13] SUYKENS J A K,VANDEWALLE J.Least squares support vector machine classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[14]周欣然.基于最小二乘支持向量机的在线建模与控制方法研究[D].长沙:湖南大学,2012.
[15]陈爱军.最小二乘支持向量机及其在工业过程建模中的应用[D].杭州:浙江大学,2006.
[16] LU Chao,CHEN Jie,HONG Rongjing,et al.Degradation trend estimation of slewing bearing based on LSSVM model[J].Mechanical Systems and Signal Processing,2016,76-77:353-366.
[17] WANG H,HU D.Comparison of SVM and LS-SVM for regression[C]//International Conference on Neural Networks and Brain,2005.
[18]张檑,李宏光.基于KKT条件选择被控变量的自优化控制方法[J].北京化工大学学报(自然科学版),2013,40(增刊1):67-71.ZHANG Lei,LI Hongguang.Self-optimizing control approach with Karush Kuhn Tucker(KKT)based controlled variable alternatives[J].Journal of Beijing University of Chemical Technology(Natural Science Edition),2013,40(S1):67-71.
[19]郑丽媛,孙朋,张素君.煤矿瓦斯突出预测的PSOLSSVM模型[J].仪表技术与传感器,2014(6):138-140.ZHENG Liyuan, SUNPeng, ZHENGSujun.Prediction on gas emission value based on PSOLSSVM[J].Instrument Technique and Sensor,2014(6):138-140.
[20] MCKAY M D,BECKMAN R J,CONOVER W J.A comparison of three methods for selecting values of input variables in the analysis of output from a computer code[J].Technometrics,2000,21(2):239-245.
[21]吴振君,王水林,葛修润.LHS方法在边坡可靠度分析中的应用[J].岩土力学,2010,31(4):1047-1054.WU Zhenjun,WANG Shuilin,GE Xiurun.Application of Latin hypercube sampling technique to slope reliability analysis[J].Rock and Soil Mechanics,2010,31(4):1047-1054.
[22] BOX G E P,HUNTER J S,HUNTER W G.Statistics for experimenters:design,innovation,and discovery[M].2nd edition.New York:WileyInterscience,2005.
[23]任彬,宋益勇,张中仕,等.多翼离心通风机全流场数值模拟与性能预测[J].风机技术,2011(2):21-23.REN Bin,SONG Yiyong,ZHANG Zhongshi,et al.Numerical simulation and performance forecast of whole flow field in a multi-blade centrifugal fan[J].Compressor,Blower&Fan Technology,2011(2):21-23.
[2]万雷.单级离心风机气动性能模拟与实验验证[D].哈尔滨:哈尔滨工程大学,2013.
[3]李超,伍继浩.离心通风机整机损失数学模型的研究及优化[J].风机技术,2000(4):15-17.
[4]姜华,宫武旗,张炜,等.带分流叶片离心叶轮非定常流场的实验研究[J].西安交通大学学报,2009,43(9):14-18.JIANG Hua,GONG Wuqi,ZHANG Wei,et al.Experimental investigation of the unsteady discharge flow in a centrifugal impeller with splitter blades[J].Journal of Xi'an Jiaotong University,2009,43(9):14-18.
[5]聂波,张俊林,满建楠.进口弯管对前向离心通风机性能影响的研究[J].流体机械,2018,46(1):6-9.NIE Bo,ZHANG Junlin,MAN Jiannan.Effect on the performance of the forward centrifugal fan with inlet bent pipe[J].Fluid Machinery,2018,46(1):6-9.
[6] ONMA P,CHANTRASMI T.Comparison of two methods to determine fan performance curves using computational fluid dynamics[C]//IOP Conference Series:Materials Science and Engineering,2018.
[7] MADHWESH N,KARANTH K V,SHARMA N Y.Effect of innovative circular shroud fences on a centrifugalfanforaugmentedperformance-A numerical analysis[J].Journal of Mechanical Science and Technology,2018,32(1):185-197.
[8]黄忠文,王培,韩海燕.离心通风机叶轮的有限元建模与应力分析[J].流体机械,2015,43(10):27-30.HUANG Zhongwen,WANG Pei,HAN Haiyan.Finite element modeling and stress analysis of centrifugal fan impeller[J].Fluid Machinery,2015,43(10):27-30.
[9]邵国建,陈帮,谢云川.进气箱对离心风机性能影响的研究[J].风机技术,2018,60(1):39-43.SHAO Guojian,CHEN Bang,XIE Yunchuan.Effect of inlet box on the performance of centrifugal fan[J].Chinese Journal of Turbomachinery,2018,60(1):39-43.
[10]张玮.基于神经网络的数据统计建模[D].杭州:浙江工业大学,2009.
[11] VAPNIK V N.An overview of statistical learning theory[J].IEEE transactions on neural networks,1999,10(5):988-999.
[12] CHERKASSKY V,MA Y.Practical selection of SVM parameters and noise estimation for SVM regression[J].Neural Networks,2004,17(1):113-126.
[13] SUYKENS J A K,VANDEWALLE J.Least squares support vector machine classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[14]周欣然.基于最小二乘支持向量机的在线建模与控制方法研究[D].长沙:湖南大学,2012.
[15]陈爱军.最小二乘支持向量机及其在工业过程建模中的应用[D].杭州:浙江大学,2006.
[16] LU Chao,CHEN Jie,HONG Rongjing,et al.Degradation trend estimation of slewing bearing based on LSSVM model[J].Mechanical Systems and Signal Processing,2016,76-77:353-366.
[17] WANG H,HU D.Comparison of SVM and LS-SVM for regression[C]//International Conference on Neural Networks and Brain,2005.
[18]张檑,李宏光.基于KKT条件选择被控变量的自优化控制方法[J].北京化工大学学报(自然科学版),2013,40(增刊1):67-71.ZHANG Lei,LI Hongguang.Self-optimizing control approach with Karush Kuhn Tucker(KKT)based controlled variable alternatives[J].Journal of Beijing University of Chemical Technology(Natural Science Edition),2013,40(S1):67-71.
[19]郑丽媛,孙朋,张素君.煤矿瓦斯突出预测的PSOLSSVM模型[J].仪表技术与传感器,2014(6):138-140.ZHENG Liyuan, SUNPeng, ZHENGSujun.Prediction on gas emission value based on PSOLSSVM[J].Instrument Technique and Sensor,2014(6):138-140.
[20] MCKAY M D,BECKMAN R J,CONOVER W J.A comparison of three methods for selecting values of input variables in the analysis of output from a computer code[J].Technometrics,2000,21(2):239-245.
[21]吴振君,王水林,葛修润.LHS方法在边坡可靠度分析中的应用[J].岩土力学,2010,31(4):1047-1054.WU Zhenjun,WANG Shuilin,GE Xiurun.Application of Latin hypercube sampling technique to slope reliability analysis[J].Rock and Soil Mechanics,2010,31(4):1047-1054.
[22] BOX G E P,HUNTER J S,HUNTER W G.Statistics for experimenters:design,innovation,and discovery[M].2nd edition.New York:WileyInterscience,2005.
[23]任彬,宋益勇,张中仕,等.多翼离心通风机全流场数值模拟与性能预测[J].风机技术,2011(2):21-23.REN Bin,SONG Yiyong,ZHANG Zhongshi,et al.Numerical simulation and performance forecast of whole flow field in a multi-blade centrifugal fan[J].Compressor,Blower&Fan Technology,2011(2):21-23.