基于CPSO-LSSVM的汽轮机热耗率软测量模型

王莉莉, 陈国彬, 李一龙, 刘超, 牛培峰

PDF(957 KB)
动力工程学报 ›› 2018, Vol. 38 ›› Issue (9) : 706-712.
汽轮机与燃气轮机

基于CPSO-LSSVM的汽轮机热耗率软测量模型

  • 王莉莉1, 陈国彬1, 李一龙2, 刘超3, 牛培峰3
作者信息 +

Soft Sensor Modelling of Steam Turbine Heat Rate Based on CPSO-LSSVM

  • WANG Lili1, CHEN Guobin1, LI Yilong2, LIU Chao3, NIU Peifeng3
Author information +
History +

摘要

为了准确建立汽轮机热耗率预测模型,提出了一种基于变空间Logistic混沌粒子群算法(CPSO)优化最小二乘支持向量机(LSSVM)的汽轮机热耗率软测量模型。采用变空间Logistic混沌搜索策略和粒子镜像越界处理策略来改善粒子群算法(PSO)的全局优化性能,提出了CPSO优化最小二乘支持向量机的超参数以改善模型预测精度,并以某600 MW汽轮机组为研究对象,利用该机组的运行数据建立CPSO-LSSVM的热耗率预测模型。结果表明:CPSO-LSSVM模型具有更高的预测精度和更强的泛化能力,能够准确有效地预测热电厂的汽轮机热耗率。

Abstract

To accurately predict the heat rate of steam turbines, a soft-sensing model was proposed based on least square support vector machine (LSSVM) modified by chaos particle swarm optimization (CPSO). The specific way is to use the variable space chaos searching strategy and the particle cross mirror image processing strategy to improve the global search ability of particle swarm optimization (PSO). Then a CPSO algorithm was developed to find the optimal parameters of LSSVM to improve the regression accuracy and generalization ability of the model. Finally, a CPSO-LSSVM model of heat rate was established for a 600 MW steam turbine based on its operation data. Simulation results show that the CPSO-LSSVM model has a higher accuracy in prediction and stronger capability in parameter optimization and generation, which may help to accurately and effectively predict the heat rate of steam turbines.

关键词

热耗率 / 粒子群算法 / 最小二乘支持向量机 / 混沌搜索 / 软测量模型

Key words

heat rate / particle swarm optimization / least square support vector machine / chaos searching / soft sensor modelling

引用本文

导出引用
王莉莉, 陈国彬, 李一龙, 刘超, 牛培峰. 基于CPSO-LSSVM的汽轮机热耗率软测量模型. 动力工程学报. 2018, 38(9): 706-712
WANG Lili, CHEN Guobin, LI Yilong, LIU Chao, NIU Peifeng. Soft Sensor Modelling of Steam Turbine Heat Rate Based on CPSO-LSSVM. Journal of Chinese Society of Power Engineering. 2018, 38(9): 706-712

参考文献

[1] 张春发, 王惠杰, 宋之平, 等. 火电厂单元机组最优运行初压的定量研究[J]. 中国电机工程学报, 2006, 26(4):36-40. ZHANG Chunfa, WANG Huijie, SONG Zhiping, et al. Quantitative research of optimal initial operation pressure for the coal-fired power unit plant[J]. Proceedings of the CSEE, 2006, 26(4):36-40.
[2] ADIBHATLA S, KAUSHIK S C. Energy and exergy analysis of a supercritical thermal power plant at various load conditions under constant and pure sliding pressure operation[J]. Applied Thermal Engineering, 2014, 73(1):51-65.
[3] 刘超, 牛培峰, 段晓龙, 等. 基于相关向量机的汽轮机最优运行初压的确定[J]. 化工学报, 2016, 67(9):3812-3816. LIU Chao, NIU Peifeng, DUAN Xiaolong, et al. Determination of optimal initial steam pressure of turbine based on relevance vector machine[J]. CIESC Journal, 2016, 67(9):3812-3816.
[4] 张文琴, 付忠广, 靳涛, 等. 基于偏最小二乘算法的热耗率回归分析[J]. 现代电力, 2009, 26(5):56-59. ZHANG Wenqin, FU Zhongguang, JIN Tao, et al. Heat rate regression analysis based on partial least squares algorithm[J]. Modern Electric Power, 2009, 26(5):56-59.
[5] 牛培峰, 陈科, 马云鹏, 等. 基于磷虾群算法的汽轮机热耗率建模应用[J]. 动力工程学报, 2016, 36(10):781-787. NIU Peifeng, CHEN Ke, MA Yunpeng, et al. Modelling of turbine heat rate based on krill herd algorithm and its application[J]. Journal of Chinese Society of Power Engineering, 2016, 36(10):781-787.
[6] 朱誉, 冯利法, 徐治皋. 基于BP神经网络的热经济性在线计算模型[J]. 热力发电, 2008, 37(12):17-19, 30. ZHU Yu, FENG Lifa, XU Zhigao. An on-line calculation model of thermal economic efficiency based BP neural network[J]. Thermal Power Generation, 2008, 37(12):17-19, 30.
[7] NIU Peifeng, ZHANG Weiping. Model of turbine optimal initial pressure under off-design operation based on SVR and GA[J]. Neurocomputing, 2012, 78(1):64-71.
[8] SUN Wei, SUN Jingyi. Daily PM2.5 concentration prediction based on principal component analysis and LSSVM optimized by cuckoo search algorithm[J]. Journal of Environmental Management, 2017, 188:144-152.
[9] JORDEHI A R. Enhanced leader PSO (ELPSO):a new PSO variant for solving global optimisation problems[J]. Applied Soft Computing, 2015, 26:401-417.
[10] HUANG Li, DING Shuai, YU Shouhao, et al. Chaos-enhanced cuckoo search optimization algorithms for global optimization[J]. Applied Mathematical Modelling, 2016, 40(5/6):3860-3875.
[11] ZHANG Yanjun, ZHAO Yu, FU Xinghu, et al. A feature extraction method of the particle swarm optimization algorithm based on adaptive inertia weight and chaos optimization for Brillouin scattering spectra[J]. Optics Communications, 2016, 376:56-66.
[12] 李方伟, 张新跃, 朱江, 等. 基于APDE-RBF神经网络的网络安全态势预测方法[J]. 系统工程与电子技术, 2016, 38(12):2869-2875. LI Fangwei, ZHANG Xinyue, ZHU Jiang, et al. Network security situation prediction based on APDE-RBF neural network[J]. Systems Engineering and Electronics, 2016, 38(12):2869-2875.
[13] 刘超, 牛培峰, 游霞. 反向建模方法在汽轮机热耗率建模中的应用[J]. 动力工程学报, 2014, 34(11):867-872, 902. LIU Chao, NIU Peifeng, YOU Xia. Application of reversed modeling method in prediction of steam turbine heat rate[J]. Journal of Chinese Society of Power Engineering, 2014, 34(11):867-872, 902.
[14] 牛培峰, 刘超, 李国强, 等. 基于双层聚类与GSA-LSSVM的汽轮机热耗率多模型预测[J]. 电机与控制学报, 2016, 20(3):90-95. NIU Peifeng, LIU Chao, LI Guoqiang, et al. Multi-model for turbine heat rate forecasting based on double layer clustering algorithm and GSA-LSSVM[J]. Electric Machines and Control, 2016, 20(3):90-95.
[15] LIU Chao, NIU Peifeng, LI Guoqiang. A hybrid heat rate forecasting model using optimized LSSVM based on improved GSA[J]. Neural Processing Letters, 2017, 45(1):299-318.

基金

国家自然科学基金资助项目(61403331,61573306);重庆市教委科学技术研究资助项目(KJ133103)
PDF(957 KB)

595

Accesses

0

Citation

Detail

段落导航
相关文章

/