此页面上的内容需要较新版本的 Adobe Flash Player。

获取 Adobe Flash Player

Model identification of continuous stirred tank reactor based on QKLMS algorithm

LI Jun1,2,3,  LI Xiang-yue1

 

(1. School of Automation & Electrical Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China2. Gansu Provincial Engineering Technology Center for Informatization of Logistics & Transport Equipment, Lanzhou 730070, China3. Gansu Provincial Industry Technology Center of Logistics & Transport Equipment, Lanzhou 730070, China

 

 

AbstractThe continuous stirred tank reactor (CSTR) is one of the typical chemical processes. Aiming at its strong nonlinear characteristics, a quantized kernel least mean square(QKLMS) algorithm is proposed. The QKLMS algorithm is based on a simple online vector quantization technology instead of sparsification, which can compress the input or feature space and suppress the growth of the radial basis function (RBF) structure in the kernel learning algorithm. To verify the effectiveness of the algorithm, it is applied to the model identification of CSTR process to construct a nonlinear mapping relationship between coolant flow rate and product concentration. In additiion, the proposed algorithm is further compared with least squares support vector machine (LS-SVM), echo state network (ESN), extreme learning machine with kernels (KELM), etc. The experimental results show that the proposed algorithm has higher identification accuracy and better online learning ability under the same conditions.

 

Key wordskernel learning algorithmquantized kernel least mean square (QKLMS)continuous stirred tank reactor (CSTR)system identification

 

CLD numberTP183             doi10.3969/j.issn.1674-8042.2020.04.010

 

References

 

1Li T, Wei F K, Wang Y C, et al. Recovery of long-term storage of nitrosated flocculent sludge activity. Environmental Science, 2018, 39(9): 4310-4316.

2Torres-Mancera P, Ancheyta J, Martínez J. Deactivation of a hydrotreating catalyst in a bench-scale continuous stirred tank reactor at different operating conditions. Fuel, 2018, 234: 326-334.

3Zhao L L, Feng J, Zhao L X, et al. Research progress on optimization of full-mix anaerobic fermentation reactor. China Biogas, 2018, 36(4): 33-38.

4Kothare M V. Process control: modeling, design and simulation. IEEE Control Systems Magazine, 2007, 26(6): 106-108.

5Ding X Q, Yang X L, Yang H, Nonlinear predictive controller design for CSTR process. Control Engineering of China, 2009, 16(2): 145-147.

6Li D J. Adaptive neural network control of continuous stirred tank reactor. CIESC Journal, 2013, 64(12): 4674-4680.

7Beyhan S, Alci M. Mixed structured RBF network for direct inverse control of nonlinear systems. In: Proceedings of 2009 Fifth IEEE International Conference on Soft Computing, Computing with Words and Perceptions in System Analysis, Decision and Control, Famagusta, IEEE, 2009: 1-4.

8Chen F, Jing Z L, Yao X D. Fast parameter learning algorithm for fuzzy neural networks. Control Theory and Applications, 2002, 19(4): 583-587.

9Kamalabady A S, Salahshoor K. A new on-line identification approach for affine modeling of nonlinear processes using an adaptive neural network. In: Proceedings of 16th Mediterranean Conference on Control and Automation, Ajaccio, France, IEEE, 2008: 735-740.

10Li X H, Li J. Identification of continuous stirred tank reactor based on echo state network. Information and control, 2014, 43(2): 223-228.

11Li J, Shi Q. Model Identification of Continuous Stirring Reactor Based on KELM. Control Engineering, 2017, (10): 171-177.

12Rosipal R J, Trejo L J. Kernel partial least squares regression in reproducing kernel hilbert space. Journal of Machine Learning Research, 2002, 2: 97-123.

13Li H L, Qian Z H, Tian H L. Research on indoor localization algorithmbased on kernel principal component analysis. Journal on Communications, 2017, 38(1): 158-167.

14Engel Y, Mannor S, Meir R. The kernel recursive least-squares algorithm. IEEE Transactions on Signal Processing, 2004, 52(8): 2275-2285.

15Li J, Wang Q L. Short-term traffic flow online forecasting based on kernel adaptive filter. Journal of Measurement Science & Instrumentation, 2018, 9(4): 326-334.

16Chen B, Zhao S, Zhu P, et al. Quantized kernel least mean square algorithm. IEEE Transactions on Neural Networks and Learning Systems, 2012, 23(1): 22-32.

17Lightbody G, Irwin G W. Nonlinear control structures based on embedded neural system models. IEEE transactions on Neural Networks, 1997, 8(3): 553-567.

18Li H S, Zhong Z Y, Zhang Y L. Predictive Control Based on Least Squares Support Vector Machine. Computing Technology and Automation, 2009, 28(1): 32-36.

 

 

基于量化核最小均方算法的连续搅拌反应釜模型辨识

 

 1,2,3, 李香月1

 

(1. 兰州交通大学 自动化与电气工程学院, 甘肃 兰州 7300702. 甘肃省物流及运输装备信息化工程技术研究中心, 甘肃 兰州 7300703. 甘肃省物流与运输装备行业技术中心, 甘肃 兰州 730070

  

 :  连续搅拌反应釜(Continuous stirred tank reactor, CSTR)是典型的化工过程之一, 本文针对其强非线性特性, 提出了一种量化核最小均方(Quantized kernel least mean square, QKLMS)算法。 该算法基于一种简单在线矢量量化技术替代稀疏化准则, 可以对输入空间进行压缩, 从而抑制核学习算法中径向基函数(Radial basis function, RBF)结构的增长。 为验证该算法的有效性, 将其应用于CSTR过程的模型辨识中, 构建冷却剂流量与生成物浓度之间的非线性映射关系。 此外, 将所提算法与最小二乘支持向量机(Least square support vector machine, LS-SVM)、 回声状态网络(Echo state network, ESN)以及核极限学习机(Extreme learning machine with kernels, KELM)等算法进行比较。 实验结果表明, 在同等条件下, 本文所提算法具有更高的辨识精度和更好的在线学习能力。

 

关键词:  核学习算法; 量化核最小均方; 连续搅拌反应釜; 系统辨识

 

引用格式:  LI Jun, LI Xiang-yue. Model identification of continuous stirred tank reactor based on QKLMS algorithm. Journal of Measurement Science and Instrumentation, 2020, 114): 382-387. doi10.3969j.issn.1674-8042.2020.04.010

 

[full text view]