Modeling of State-Dependent Switching System Based on Data-Driven
-
摘要:
切换系统是由一系列连续或离散的子系统和切换机制组合而成的一类复杂系统,状态依赖型切换系统因其复杂性而尚未被深入研究. 因此,通过系统的输入输出轨迹来对状态依赖的切换系统进行数据驱动建模,利用数据挖掘技术寻找数据之间的有用信息,建立输入与输出之间更形象的表达形式;在此基础上提出一种结构框架,根据辨识轨迹的切换时刻将数据分段,借助神经网络建立子系统的模型以及切换规则,深度挖掘状态依赖切换系统的信息,得到切换系统中子系统及子系统间的信息. 实验结果表明:相比传统的机理建模,本文提出的数据驱动方法将建模的复杂度降低了17.3%.
Abstract:A switched system is a class of complex systems that integrate a series of continuous or discrete subsystems and switching mechanisms. State-dependent switching systems have not been studied in depth due to complexity. Therefore, the modeling of state-dependent switching systems is explored through the input-output trajectories of the systems. The data mining technique is used to find useful information between data and establish a more specific and explicit representation between inputs and outputs. On this basis, a framework is proposed to segment the data according to the switching time of the identified trajectory, build the subsystem model by neural network to fit its switching rules, deeply mine the information of the state-dependent switching system, and obtain the information between the subsystems and subsystems in the switching system. The experimental results show that compared with the modeling of traditional mechanism, the proposed data-driven method reduces the modeling complexity by 17.3%.
-
Key words:
- state-dependent switching systems /
- data-driven /
- data model /
- neural network
-
-
[1] 郑刚,谭民,宋永华. 混杂系统的研究进展[J]. 控制与决策,2004,19(1): 7-11,16.ZHENG Gang, TAN Min, SONG Yonghua. Research on hybrid systems: a survey[J]. Control and Decision, 2004, 19(1): 7-11,16. [2] 程代展,郭宇骞. 切换系统进展[J]. 控制理论与应用,2005,22(6): 954-960.CHENG Daizhan, GUO Yuqian. Advances on switched systems[J]. Control Theory & Applications, 2005, 22(6): 954-960. [3] SHOHAM S, YAHAV E, FINK S, et al. Static specification mining using automata-based abstractions[J]. IEEE Transactions on Software Engineering, 2008, 34(5): 651-666. doi: 10.1109/TSE.2008.63 [4] GUDIÑO-MENDOZA B, LÓPEZ-MELLADO E. A modeling methodology for designing agents networks using timed hybrid Petri nets[J]. Simulation, 2017, 93(4): 323-333. doi: 10.1177/0037549716687835 [5] WANG R R, ZHOU J, JIANG H, et al. A general transfer learning-based Gaussian mixture model for clustering[J]. International Journal of Fuzzy Systems, 2021, 23(3): 776-793. doi: 10.1007/s40815-020-01016-3 [6] LIN H, ANTSAKLIS P J. Stability and stabilizability of switched linear systems: a survey of recent results[J]. IEEE Transactions on Automatic Control, 2009, 54(2): 308-322. doi: 10.1109/TAC.2008.2012009 [7] ZHANG Q, WANG Q J, LI G L. Switched system identification based on the constrained multi-objective optimization problem with application to the servo turntable[J]. International Journal of Control, Automation and Systems, 2016, 14(5): 1153-1159. doi: 10.1007/s12555-015-0057-4 [8] GARULLI A, PAOLETTI S, VICINO A. A survey on switched and piecewise affine system identification[J]. IFAC Proceedings Volumes, 2012, 45(16): 344-355. doi: 10.3182/20120711-3-BE-2027.00332 [9] HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine: a new learning scheme of feedforward neural networks[C]//IEEE International Joint Conference on Neural Networks. Budapest: IEEE, 2004: 985-990. [10] JI Y F, ZHANG S, YIN Y X, et al. Application of the improved the ELM algorithm for prediction of blast furnace gas utilization rate[J]. IFAC-PapersOnLine, 2018, 51(21): 59-64. doi: 10.1016/j.ifacol.2018.09.393 [11] YIN Y H, LI H F. Multi-view CSPMPR-ELM feature learning and classifying for RGB-D object recognition[J]. Cluster Computing, 2019, 22(4): 8181-8191. [12] 张斌. 切换系统的切换律及其输入u(t)整体最佳的设计方法研究[D]. 哈尔滨: 哈尔滨工程大学,2018. [13] TCHIOTSOP D, SAHA TCHINDA B, TCHINDA B, et al. Edge detection of intestinal parasites in stool microscopic images using multi-scale wavelet transform[J]. Signal, Image and Video Processing, 2015, 9(1): 121-134. [14] REIN S, REISSLEIN M. Scalable line-based wavelet image coding in wireless sensor networks[J]. Journal of Visual Communication and Image Representation, 2016, 40: 418-431. doi: 10.1016/j.jvcir.2016.07.006