基于信息熵的BP网络在热工系统建模中的应用

doi:10.16182/j.issn1004731x.joss.201701030

系统仿真学报 ›› 2017, Vol. 29 ›› Issue (1): 226-233.doi: 10.16182/j.issn1004731x.joss.201701030

• 仿真应用工程 • 上一篇

基于信息熵的BP网络在热工系统建模中的应用

孙海蓉¹, 王蕊^1,2, 耿军亚^1,2

1.华北电力大学控制与计算机工程学院,保定 071003;
2.华北电力大学河北省发电过程仿真与优化控制工程技术研究中心,保定 071003

收稿日期:2016-04-13 修回日期:2016-06-02 出版日期:2017-01-08 发布日期:2020-06-01
作者简介:孙海蓉(1972-),女,北京,博士,副教授,研究方向为智能控制;王蕊(1987-),女,河北保定,硕士生,研究方向为数据挖掘。
基金资助:
中央高校基本科研业务费专项资金(2016MS143)

Thermal System Modeling Based on Entropy and BP Neural Network

Sun Hairong¹, Wang Rui^1,2, Geng Junya^1,2

1. School of Control and Computer Engineering, North China Electric Power University, Baoding 071003, China;
2. Hebei Engineering Research Center of Simulation & Optimized Control for Power Generation, North China Electric Power University, Baoding 071003, China

Received:2016-04-13 Revised:2016-06-02 Online:2017-01-08 Published:2020-06-01

摘要/Abstract

摘要： 针对火电厂热工对象实时建模困难、模型精度不高、以及神经网络因输入量增多导致收敛速度大幅降低的问题,将近似决策熵属性约简用于BP(Back Propagation)网络的建模中,提出了一种基于信息熵的BP网络建模方法。该方法采用k-means聚类算法对现场数据进行预处理及有效性评价,用近似决策熵对系统输入进行属性约简,用BP网络训练建立非线性模型。通过将该方法应用于主汽温和NOx排放浓度建模表明,该方法模型精度高,而且有效降低了BP网络输入层的维数,简化了网络结构,提高了训练速度,对热工系统实时建模具有重要实用价值。

关键词: 粗糙集, 属性约简, 近似决策熵, 神经网络, 非线性建模

Abstract: The real-time modeling is difficult on thermal system and the model precision is not high. The convergence rate of Neural Network (NN) decreases dramatically when there are too many inputs. The BP NN modeling method based on information entropy was proposed in which the attribute reduction based on the model of approximation decision entropy was used. Field data were preprocessed by the k-means clustering algorithm and the validity was then evaluated. The approximation decision entropy was used in attribute reduction of the inputs. And the nonlinear model was built by the training of BP NN. Eventually, the proposed method was applied to the modeling of main steam temperature and the modeling of NOx emission concentration and the result shows that, the method simplified the structure of network and increased the training speed with high-precision of the model. The method is of great significance on real-time modeling and has higher practical value.

Key words: rough set, attribute reduction, approximation decision entropy, NN, nonlinear modeling

中图分类号:

TP391

孙海蓉, 王蕊, 耿军亚. 基于信息熵的BP网络在热工系统建模中的应用[J]. 系统仿真学报, 2017, 29(1): 226-233.

Sun Hairong, Wang Rui, Geng Junya. Thermal System Modeling Based on Entropy and BP Neural Network[J]. Journal of System Simulation, 2017, 29(1): 226-233.

参考文献

[1] 王贺. 基于粗糙集知识约减的电站优化运行研究 [D]. 保定: 华北电力大学, 2014.
(Wang He.The research of rough set theory in power plant operation optimization [D]. Baoding, China: North China Electric Power University, 2014.)
[2] 李建强, 姜翻, 汪安明, 等. 粗糙集在控制燃煤电厂NO_x排放中的应用[J]. 电力科学与工程, 2014, 30(12): 19-23.
(Li Jianqiang, Jiang Fan, Wang Anming, et al.Application of rough set in controlling NO_x Emissions in coal-fired power plant[J]. Electric Power Science and Engineering, 2014, 30(12): 19-23.)
[3] 赵姝, 吕靖, 张燕平, 等. 不完整数据集的信息熵集成分类算法[J]. 模式识别与人工智能, 2014, 27(3): 193-198.
(Zhao Shu, Lv Jing, Zhang Yanping, et al.Information entropy ensemble classification algorithm for incomplete data[J]. PR & AI, 2014, 27(3): 193-198.)
[4] 高爽, 冬雷, 高阳, 等. 基于粗糙集理论的中长期风速预测[J]. 中国电机工程学报, 2012, 32(1): 32-37.
(Gao Shuang, Dong Lei, Gao Yang, et al.Mid-long term wind speed prediction based on rough set theory[J]. Proceeding of the CSEE, 2012, 32(1): 32-37.)
[5] 田虎森, 谢寿生, 王磊, 等. 基于粗糙集理论的飞行轨迹识别[J]. 火力与指挥控制, 2015, 40(5): 29-33.
(Tian Husen, Xie Shousheng, Wang Lei, et al.Flight track recognition based on rough set theory[J]. Fire Control & Command Control, 2015, 40(5): 29-33.)
[6] 胡玉杰. 基于粗糙集和神经网络的分类算法研究 [D].合肥: 安徽大学, 2015.
(Hu Yujie.The research of classification algorithm based on rough set and neural network [D]. Hefei, China: Anhui University, 2015.)
[7] 陈颖悦, 陈玉明. 基于信息熵与蚁群优化的属性约简算法[J]. 小型微型计算机系统, 2015, 36(3): 586-590.
(Chen Yingyue, Chen Yuming.Attribute reduction algorithm based on information entropy and colony optimization[J]. Journal of Chinese Computer Systems, 2015, 36(3): 586-590.)
[8] 江峰, 王莎莎, 杜军威, 等. 基于近似决策熵的属性约简[J]. 控制与决策, 2015, 30(1): 65-70.
(Jiang Feng, Wang Shasha, Du Junwei, et al.Attribute reduction based on approximation decision entropy[J]. Control and Decision, 2015, 30(1): 65-70.)
[9] 靳冰洁, 张步涵, 姚建国, 等. 基于信息熵的大型电力系统元件脆弱性评估[J]. 电力系统自动化, 2015, 39(5): 61-68.
(Jin Bingjie, Zhang Buhan, Yao Jianguo, et al.Large-scale power system components vulnerability assessment based on entropy[J]. Automation of Electric Power Systems, 2015, 39(5): 61-68.)
[10] 祖鸿娇, 米据生. 带权重条件嫡的属性约简算法[J]. 计算机科学与探索, 2016, 10(3): 445-450.
(Zu Hongqiao, Mi Jusheng.Attribute reduction algorithm based on conditional entropy with weights[J]. Journal of Frontiers of Computer Science and Technology, 2016, 10(3):445-450.)
[11] 周世兵. 聚类分析中的最佳聚类数确定方法研究及应用 [D]. 无锡: 江南大学, 2011.
(Zhou Shibing.Research and application on determining optimal number of clusters in cluster analysis [D]. Wuxi, China: Jiangnan University, 2011.)
[12] 侯晓宁, 孙海蓉. 基于现场数据和PSO算法的机组主汽温系统辨识[J]. 计算机仿真, 2014, 31(12): 133-136.
(Hou Xiaoning, Sun Hairong.System identification of main steam temperature control system for units based on online data and PSO algorithm[J]. Computer Simulation, 2014, 31(12): 133-136.)
[13] 侯晓宁. 1 000 MW超超临界机组主汽温建模及其精度研究 [D]. 保定: 华北电力大学, 2015.
(Hou Xiaoning.Modeling and accuracy research of 1000MW USC unit's main steam temperature [D]. Baoding, China: North China Electric Power University, 2015.)
[14] 陈其松. 智能优化支持向量机预测算法及应用研究 [D]. 贵阳: 贵州大学, 2009.
(Chen Qisong.Research on intelligent optimization support vector machine predictive algorithm and its application [D]. Guiyang, China: Guizhou University, 2009.)
[15] 刘吉臻, 吕游, 杨婷婷. 基于变量选择的锅炉NO_x排放的最小二乘支持向量机建模[J]. 中国电机工程学报, 2012, 32(20): 102-107, 146.
(Liu Jizhen, Lv You, Yang Tingting.Least squares support vector machine modeling on NO_x emission of boilers based on variable selection[J]. Proceeding of the CSEE, 2012, 32(20): 102-107, 146.)

基于信息熵的BP网络在热工系统建模中的应用

Thermal System Modeling Based on Entropy and BP Neural Network

PDF (PC)

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张森, 张孟炎, 邵敬平, 普杰信. 基于随机策略搜索的多机三维路径规划方法[J]. 系统仿真学报, 2022, 34(6): 1286-1295.
[2]	张立峰, 王会忍. 基于卷积神经网络及有限元仿真的电容层析成像图像重建[J]. 系统仿真学报, 2022, 34(4): 712-718.
[3]	康旭, 张晓峰. 基于生成对抗神经网络的雷达遥感数据增广方法[J]. 系统仿真学报, 2022, 34(4): 920-927.
[4]	周思锦, 陈棣成, 涂耿, 姜大志. 基于个性化和记忆机制的多模态情感计算模型[J]. 系统仿真学报, 2022, 34(4): 745-758.
[5]	孙晓安, 栾小丽, 刘飞. 基于智能优化灰色模型的电子固废预测[J]. 系统仿真学报, 2022, 34(3): 536-542.
[6]	魏娟, 游磊, 郭阳勇, 唐志海. 基于小波神经网络的多楼层疏散模型[J]. 系统仿真学报, 2022, 34(2): 269-277.
[7]	敖邦乾, 杨莎, 令狐金卿, 叶振环. 基于级联神经网络疲劳驾驶检测系统设计[J]. 系统仿真学报, 2022, 34(2): 323-333.
[8]	魏腾飞, 潘庭龙. 基于改进PSO优化LSTM网络的短期电力负荷预测[J]. 系统仿真学报, 2021, 33(8): 1866-1874.
[9]	赵瑛, 陆耀, 张健, 梁启弟, 龙炜. 基于深度神经网络的多视角人体动作识别[J]. 系统仿真学报, 2021, 33(5): 1019-1030.
[10]	刘晓琳, 吴竟祎. 飞机舵机电液加载系统控制补偿器设计与仿真[J]. 系统仿真学报, 2021, 33(5): 1031-1038.
[11]	于广宇, 董学平, 王祥民, 甘敏. 弹性网下基于LSTM的分解炉出口温度预测[J]. 系统仿真学报, 2021, 33(5): 1078-1085.
[12]	方群, 陈绪凯, 何昕, 祝玉军, 刘毅杨, 方阳阳, 李贺举. 基于SWIPT无线协作网络的中继选择算法研究[J]. 系统仿真学报, 2021, 33(4): 801-808.
[13]	仝卫国, 庞雪纯, 朱赓宏. 基于卷积神经网络的气液两相流流型识别方法[J]. 系统仿真学报, 2021, 33(4): 883-891.
[14]	冯新扬, 邵超. 跨卷积网络特征融合的SAR图像目标识别[J]. 系统仿真学报, 2021, 33(3): 554-561.
[15]	马立新, 朱勇杰, 季乐延. 神经网络优化的无感永磁同步电机控制系统[J]. 系统仿真学报, 2021, 33(3): 622-630.