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

doi:10.11949/0438-1157.20190762

摘要/Abstract

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

蔡涛, 杨博, 李宏光. 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[J]. 化工学报, 2020, 71(3): 1095-1102.

Tao CAI, Bo YANG, Hongguang LI. Chemical process multivariate time series predictions based on time-delay-mining fuzzy time cognitive maps[J]. CIESC Journal, 2020, 71(3): 1095-1102.

图/表 15

参考文献 32

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．

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

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

相关变量	最大相关系数	时延/s
C₁→C₅	0.1022	807
C₂→C₅	0.6929	750
C₃→C₅	0.5373	78
C₄→C₅	0.5307	375
C₅→C₅	—	1

相关变量	最大相关系数	时延/s
C₁→C₅	0.1022	807
C₂→C₅	0.6929	750
C₃→C₅	0.5373	78
C₄→C₅	0.5307	375
C₅→C₅	—	1

τ	RMSE×10²
τ	传统FCM	改进FTCM	无时延	无自影响γ	无偏置w₀
1	35.9318	17.5711	33.0864	39.0032	33.3274
2	37.8351	8.8389	22.7103	42.4510	22.0586
3	42.9456	5.9062	13.4211	42.8381	12.2322
4	43.5164	6.3064	4.2688	43.5805	9.9159
5	43.0080	1.3927	3.1508	41.0092	6.5976
6	43.3243	2.9986	2.7560	43.3517	3.8549
7	43.8856	5.5132	3.0433	43.3933	4.6063
8	43.3293	4.8570	10.4147	45.3498	7.6283
9	43.7955	13.0252	13.9385	42.9575	5.2725