化工学报 ›› 2020, Vol. 71 ›› Issue (3): 1095-1102.doi: 10.11949/0438-1157.20190762

• 过程系统工程 • 上一篇    下一篇

基于时延挖掘模糊时间认知图的化工过程多变量时序预测方法

蔡涛(),杨博,李宏光()   

  1. 北京化工大学信息科学与技术学院,北京 100029
  • 收稿日期:2019-07-04 修回日期:2019-09-19 出版日期:2020-03-05 发布日期:2019-11-28
  • 通讯作者: 李宏光 E-mail:caitao2012_2@163.com;lihg@mail.buct.edu.cn
  • 作者简介:蔡涛(1991—),男,硕士研究生,caitao2012_2@163.com

Chemical process multivariate time series predictions based on time-delay-mining fuzzy time cognitive maps

Tao CAI(),Bo YANG,Hongguang LI()   

  1. College of Information Science & Technology, Beijing University of Chemical Technology, Beijing 100029, China
  • Received:2019-07-04 Revised:2019-09-19 Online:2020-03-05 Published:2019-11-28
  • Contact: Hongguang LI E-mail:caitao2012_2@163.com;lihg@mail.buct.edu.cn

摘要:

模糊认知图(fuzzy cognitive maps, FCM)作为一种复杂系统的建模工具,能够对系统的非线性和不确定性进行处理。由于工业过程变量间往往存在着时间延迟,传统的FCM模型难以处理这类多变量的时间序列数据,建立的预测模型往往不能反映系统内各变量真实的因果关系,从而导致预测结果的解释性差、准确度低等问题。为此,提出了一种时延挖掘模糊时间认知图(time-delay-mining fuzzy time cognitive maps, TM-FTCM),它使用互相关函数(cross-correlation function,CCF)从数据中挖掘时延信息,并通过在推理机制中添加自我影响因子和偏置及优化转换函数等参数,有效地解决了由于工业过程变量间的时延导致的预测模型不准确等问题。通过数值仿真实例及实际化工过程数据,验证了所提方法的有效性。

关键词: 模糊时间认知图, 预测, 互相关函数, 粒子群算法, 时滞系统

Abstract:

Fuzzy cognitive maps (FCM), as a modeling tool for complex systems, can handle the nonlinearity and uncertainty of the system. However, time-delay among industrial process variables is always ignored in traditional FCM models. The causal relationship between variables can t be calculated accurately. Therefore, the prediction result is inconvincible and unpredictable. The time-delay-mining fuzzy time cognitive maps (TM-FTCM) method is proposed to enhance the accuracy of the time-delay prediction model. The cross-correlation functions (CCF) helps to find the time-delay factors hiding in the big data, thus revealing the actual structure of the model. Furthermore, the optimization of self-impact factors, bias and transfer functions enhances the efficiency of the prediction process. The TM-FTCM method has been verified by numerical simulations and actual chemical plant process data to be efficient and practical.

Key words: fuzzy time cognitive maps, prediction, cross-correlation function, particle swarm optimization, time-delay system

中图分类号: 

  • TQ 028.8

图1

FCM的拓扑结构"

图2

FTCM模型"

图3

改进的FTCM的结构示意图"

表1

粒子群寻优的伪代码"

算法1使用PSO求解FTCM的参数

Input:activation:x1,x2,…,xn; targets: t1,t2,…,tn; weight matrix W; memory factor γ; cooperation index w0; ?(x) parameterτ; performance index(e.g. RMSE);

Output:optimized W, γ, w0, τ

Repeat

For k=1 to N do

Pass k-th activation xk;

Compute FCM response yk using Eq.(4);

According yk and tk to compute performance index;

Adjust W, γ, w0, τ to minimize the performance index using PSO;

End

Until performance index or algorithm’s iterations has been exceeded

表2

布谷鸟搜索算法的伪代码"

算法2布谷鸟搜索算法执行过程

Begin

Initial population: n host nests Xi(i =1,2,…,n)

Calculation fitness: fi (i =1,2,…,n);

While(not met stop condition)

using Levy flight to get new solution Xi,calculation new fitness fj;

select candidate solution Xi;

If (fi > fj):

replace candidate solution with new solution;

End

According to probability pa to abandon bad solution, using a preference random walk to generate a new solution instead of a discarded solution, retain the optimal solution

End

End

图4

实验环节流程"

图5

x1曲线"

图6

x2曲线"

图7

数值实例仿真结果"

图8

汽包工艺图"

表3

相关变量选取"

概念节点描述
C1给水温度
C2给水流量
C3排污流量
C4蒸汽流量
C5汽包液位

表4

时延挖掘"

相关变量最大相关系数时延/s
C1C50.1022807
C2C50.6929750
C3C50.537378
C4C50.5307375
C5C51

图9

FTCM(PSO)模型"

表5

不同因素的误差结果"

τRMSE×102
传统FCM改进FTCM无时延无自影响γ无偏置w0
135.931817.571133.086439.003233.3274
237.83518.838922.710342.451022.0586
342.94565.906213.421142.838112.2322
443.51646.30644.268843.58059.9159
543.00801.39273.150841.00926.5976
643.32432.99862.756043.35173.8549
743.88565.51323.043343.39334.6063
843.32934.857010.414745.34987.6283
943.795513.025213.938542.95755.2725

图10

实验结果"

1 陈龙, 刘全利, 王霖青, 等. 基于数据的流程工业生产过程指标预测方法综述[J]. 自动化学报, 2017, 43(6): 944-954.
Chen L, Liu Q L, Wang L Q, et al. Data-driven prediction on performance indicators in process industry: a survey[J]. Acta Automatica Sinica, 2017, 43(6): 944-954.
2 张浩, 刘振娟, 李宏光, 等. 基于关联变量时滞分析卷积神经网络的生产过程时间序列预测方法[J]. 化工学报, 2017,68(9): 3501-3510.
Zhang H, Liu Z J, Li H G, et al. Process time series prediction based on application of correlated process variables to CNN time delayed analyses[J]. CIESC Journal, 2017, 68(9): 3501-3510.
3 Kosko B. Fuzzy cognitive maps[J]. International Journal of Man-Machine Studies, 1986, 24(1): 65-75.
4 陈宁, 彭俊洁, 王磊, 等. 模糊灰色认知网络的建模方法及应用[J]. 自动化学报, 2018, 44(7): 1227-1236.
Chen N, Peng J J, Wang L, et al. Fuzzy grey cognitive networks modeling and its application[J]. Acta Automatica Sinica, 2018, 44(7): 1227-1236.
5 Mls K, Cimler R, Vaščák J, et al. Interactive evolutionary optimization of fuzzy cognitive maps[J]. Neurocomputing, 2017, 232: 58-68.
6 Homenda W, Jastrzebska A. Clustering techniques for fuzzy cognitive map design for time series modeling[J]. Neurocomputing, 2017, 232: 3-15.
7 Froelich W, Pedrycz W. Fuzzy cognitive maps in the modeling of granular time series[J]. Knowledge-Based Systems, 2017, 115: 110-122.
8 Lu W, Yang J, Liu X, et al. The modeling and prediction of time series based on synergy of high-order fuzzy cognitive map and fuzzy c-means clustering[J]. Knowledge-Based Systems, 2014, 70: 242-255.
9 Song H, Miao C, Roel W, et al. Implementation of fuzzy cognitive maps based on fuzzy neural network and application in prediction of time series[J]. IEEE Transactions on Fuzzy Systems, 2010, 18(2): 233-250.
10 Pedrycz W, Jastrzebska A, Homenda W. Design of fuzzy cognitive maps for modeling time series[J]. IEEE Transactions on Fuzzy Systems, 2016, 24(1): 120-130.
11 Papageorgiou E I, Poczęta K. A two-stage model for time series prediction based on fuzzy cognitive maps and neural networks[J]. Neurocomputing, 2017, 232: 113-121.
12 Song H J, Miao C Y, Shen Z Q, et al. Design of fuzzy cognitive maps using neural networks for predicting chaotic time series[J]. Neural Networks, 2010, 23(10): 1264-1275.
13 Papageorgiou E I, Froelich W. Application of evolutionary fuzzy cognitive maps for prediction of pulmonary infections[J]. IEEE Transactions on Information Technology in Biomedicine, 2012.16(1): 143-149.
14 Papageorgiou E I, Froelich W. Multi-step prediction of pulmonary infection with the use of evolutionary fuzzy cognitive maps[J]. Neurocomputing, 2012, 92: 28-35.
15 Froelich W, Papageorgiou E I, Samarinas M, et al. Application of evolutionary fuzzy cognitive maps to the long-term prediction of prostate cancer[J]. Applied Soft Computing, 2012, 12(12): 3810-3817.
16 Stach W, Kurgan L, Pedrycz W, et al. Genetic learning of fuzzy cognitive maps[J]. Fuzzy Sets and Systems, 2005, 153(3): 371-401.
17 Papageorgiou E I, Parsopoulos K E, Stylios C S, et al. Fuzzy cognitive maps learning using particle swarm optimization[J]. Journal of Intelligent Information Systems, 2005, 25(1): 95-121.
18 Salmeron J L, Ruiz-Celma A, Mena A. Learning FCMs with multi-local and balanced memetic algorithms for forecasting industrial drying processes[J]. Neurocomputing, 2017, 232: 52-57.
19 Papageorgiou E I. Learning algorithms for fuzzy cognitive maps—a review study[J]. IEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews), 2012, 42(2): 150-163.
20 Liu Z Q, Zhang J Y. Interrogating the structure of fuzzy cognitive maps[J]. Soft Computing, 2003, 7(3): 148-153.
21 Zhang J Y, Liu Z Q, Zhou S. Dynamic domination in fuzzy causal networks[J]. IEEE Transactions on Fuzzy Systems, 2006, 14(1): 42-57.
22 Hagiwara M. Extended fuzzy cognitive maps[C]//IEEE International Conference on Fuzzy Systems. IEEE, 1992: 795-801.
23 骆祥峰, 高隽. 概率模糊认知图[J]. 中国科学技术大学学报, 2003, 33(1): 26-33.
Luo X F, Gao J. Probabilistic fuzzy fognitive maps[J]. Journal of University of Science and Technology of China, 2003, 33(1): 26-33.
24 Park K S, Kim S H. Fuzzy cognitive maps considering time relationships[J]. International Journal of Human-Computer Studies, 1995, 42(2): 157-168.
25 Wei Z, Lu L, Yanchun Z. Using fuzzy cognitive time maps for modeling and evaluating trust dynamics in the virtual enterprises[J]. Expert Systems with Applications, 2008, 35(4): 1583-1592.
26 Homenda W, Jastrzebska A, Pedrycz W. Modeling time series with fuzzy cognitive maps[C]// IEEE International Conference on Fuzzy Systems. IEEE, 2014.
27 Yang F, Shah S L, Xiao D, et a1. Improved correlation analysis and visualization of industrial alarm data[J]. ISA Transactions,2012, 51(4): 499-506.
28 Yang B, Li H, Wen B. A dynamic time delay analysis approach for correlated process variables[J]. Chemical Engineering Research and Design, 2017, 122: 141-150.
29 Nasiriyan-Rad H, Amirkhani A, Naimi A, et al. Learning fuzzy cognitive map with PSO algorithm for grading celiac disease[C]// 2016 23rd Iranian Conference on Biomedical Engineering and 2016 1st International Iranian Conference on Biomedical Engineering (ICBME). IEEE, 2016.
30 Salmeron J L, Rahimi S A, Navali A M, et al. Medical diagnosis of Rheumatoid Arthritis using data driven PSO-FCM with scarce datasets[J]. Neurocomputing, 2017, 232: 104-112.
31 Salmeron J L, Froelich W. Dynamic optimization of fuzzy cognitive maps for time series forecasting[J]. Knowledge-Based Systems, 2016, 105: 29-37.
32 Yang X S, Deb S. Cuckoo search via Levy flights[C]//Nature&Biologically inspired computing. World Congress on IEEE,2009: 210-214.
[1] 王乐乐, 戴源德, 田思瑶, 林秦汉. R290在小管径水平微肋管内沸腾传热的实验研究[J]. 化工学报, 2020, 71(3): 1026-1034.
[2] 何德峰, 张永达, 李廉明, 仇翔. 循环流化床燃烧系统无终端约束字典序经济预测控制[J]. 化工学报, 2020, 71(3): 1210-1216.
[3] 赵文英, 李文文, 孙晓岩, 曹晓荣, 项曙光. 基于PR立方型状态方程普遍化温度函数的研究与评价[J]. 化工学报, 2020, 71(3): 1234-1245.
[4] 罗小燕, 戴聪聪, 程铁栋, 蔡改贫, 刘鑫, 刘吉顺. 基于改进EWT-多尺度熵和KELM的球磨机负荷识别方法[J]. 化工学报, 2020, 71(3): 1264-1277.
[5] 杜宇浩, 阎高伟, 李荣, 王芳. 基于局部线性嵌入的测地线流式核多工况软测量建模方法[J]. 化工学报, 2020, 71(3): 1278-1287.
[6] 关宏伟, 叶凌箭, 沈非凡, 顾德, 宋执环. 基于经济模型预测控制的金氰化浸出过程动态实时优化[J]. 化工学报, 2020, 71(3): 1122-1130.
[7] 田洲, 焦栋, 王金强, 刘柏平. 序列分布导向的CGC催化乙烯与1-辛烯共聚过程建模[J]. 化工学报, 2020, 71(2): 651-659.
[8] 邢瑞, 江南, 刘冰, 安亚雄, 汪亚燕, 张东辉. 基于MPC控制技术优化VPSA制氧工艺的模拟[J]. 化工学报, 2020, 71(2): 669-679.
[9] 唐振浩,张宝凯,曹生现,王恭,赵波. 基于多模型智能组合算法的锅炉炉膛温度建模[J]. 化工学报, 2019, 70(S2): 301-310.
[10] 支恩玮, 闫飞, 任密蜂, 阎高伟. 基于迁移变分自编码器-标签映射的湿式球磨机负荷参数软测量[J]. 化工学报, 2019, 70(S1): 150-157.
[11] 王羽鹏, 梁俊伟, 罗向龙, 李逸帆, 陈健勇, 陈颖. 基于神经网络的有机朗肯循环过程及循环性能计算方法[J]. 化工学报, 2019, 70(9): 3256-3266.
[12] 廉小亲, 王俐伟, 安飒, 魏伟, 刘载文. 基于SOM-RBF神经网络的COD软测量方法[J]. 化工学报, 2019, 70(9): 3465-3472.
[13] 柴伟, 郭龙航, 池彬彬. 污水处理厂出水水质变量区间预测建模[J]. 化工学报, 2019, 70(9): 3449-3457.
[14] 杨协和, 沈文锋, 张扬, 张海, 张建胜, 吕俊复. 甲醇-空气层流火焰速度的数值研究和预测模型[J]. 化工学报, 2019, 70(8): 3011-3020.
[15] 张黎, 王新宇, 李征, 谷俊峰, 阮诗伦, 申长雨. 熔融沉积3D打印材料粘接强度及模量预测研究[J]. 化工学报, 2019, 70(7): 2727-2736.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 赵众, 马楠楠, 潘立登, 徐宁, 孙康. Direct Adaptive Control Based on Gradient Estimation[J]. CIESC Journal, 2008, 16(5): 752 -761 .
[2] . 我国聚氨酯涂料“十五”发展方向和计划 [J]. CIESC Journal, 2001, 52(11): 956 .
[3] 王建友,王世昌. HIGH-PURITY WATER PRODUCTION BY RO/EDI SYSTEM [J]. CIESC Journal, 2001, 52(1): 15 -16 .
[4] 张曼徵,郭慎独,汪师俊. Ⅳ.工业钒催化剂的内部表面利用率 [J]. CIESC Journal, 1959, 10(2): 102 -107 .
[5] 陈国权,冯昌贻,梁东白,韦士平. 低压氢气下粗苯的脱硫精制 第二报 [J]. CIESC Journal, 1965, 16(1): 13 -24 .
[6] . JOURNAL OF CHEMICAL INDUSTRY AND ENGINEERING(CHINA) [J]. CIESC Journal, 1982, 33(4): 390 -393 .
[7] 吴慧雄,卢洪,李成岳. 二氧化硫强制动态氧化过程的模型化(Ⅲ)催化剂装量升级的模试考察 [J]. CIESC Journal, 1998, 49(5): 566 -573 .
[8] 褚良银,陈文梅,李晓钟,吴持恭. 水力旋流器湍流结构控制与能耗降减 [J]. CIESC Journal, 1998, 49(6): 760 -763 .
[9] 姜元国,陈日志,邢卫红. 对硝基苯酚催化加氢研究进展 [J]. CIESC Journal, 2011, 30(2): 309 .
[10] 雍玉梅1,杨超1,尹小龙2,林军3. 不可压热流体中气体传质扩散过程的数值模拟[J]. 化工学报, 2012, 63(1): 25 -35 .