Coordinated control system modelling of ultra-supercritical unit based on a new T-S fuzzy structure

被引：33

作者：

Hou, Guolian ^{[1
]}

Du, Huan ^{[1
]}

Yang, Yu ^{[2
]}

Huang, Congzhi ^{[1
,3
]}

Zhang, Jianhua ^{[1
]}

机构：

[1] North China Elect Power Univ, Sch Control & Comp Engn, Beijing 102206, Peoples R China

[2] Nanjing NARI RELAYS Elect Co Ltd, Nanjing 211100, Jiangsu, Peoples R China

[3] Guilin Univ Elect Technol, Coll & Univ Key Lab Intelligent Integrated Automa, Guangxi 541004, Peoples R China

来源：

ISA TRANSACTIONS | 2018年 / 74卷

基金：

美国国家科学基金会;

关键词：

Ultra-supercritical unit; Coordinated control system; T-S fuzzy model; k-means plus; Stochastic gradient; SUPPORT VECTOR MACHINE; BOILER-TURBINE UNIT; PREDICTIVE CONTROL; POWER-PLANT; NEURAL-NETWORKS; ALGORITHM; IDENTIFICATION; RECOGNITION; QUALITY;

D O I：

10.1016/j.isatra.2018.01.022

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

The thermal power plant, especially the ultra-supercritical unit is featured with severe nonlinearity, strong multivariable coupling. In order to deal with these difficulties, it is of great importance to build an accurate and simple model of the coordinated control system (CCS) in the ultra-supercritical unit. In this paper, an improved T-S fuzzy model identification approach is proposed. First of all, the k-means++ algorithm is employed to identify the premise parameters so as to guarantee the number of fuzzy rules. Then, the local linearized models are determined by using the incremental historical data around the cluster centers, which are obtained via the stochastic gradient descent algorithm with momentum and variable learning rate. Finally, with the proposed method, the CCS model of a 1000 MW USC unit in Tai Zhou power plant is developed. The effectiveness of the proposed approach is validated by the given extensive simulation results, and it can be further employed to design the overall advanced controllers for the CCS in an USC unit. (C) 2018 ISA. Published by Elsevier Ltd. All rights reserved.

引用

页码：120 / 133

页数：14

共 28 条

[1]

Arthur D, 2007, PROCEEDINGS OF THE EIGHTEENTH ANNUAL ACM-SIAM SYMPOSIUM ON DISCRETE ALGORITHMS, P1027

[2] An efficient approximation to the K-means clustering for massive data [J].

Capo, Marco ;

Perez, Aritz ;

Lozano, Jose A. .

KNOWLEDGE-BASED SYSTEMS, 2017, 117 :56-69

[3] Parameter estimation for a dual-rate system with time delay [J].

Chen, Lei ;

Han, Lili ;

Huang, Biao ;

Liu, Fei .

ISA TRANSACTIONS, 2014, 53 (05) :1368-1376

[4] Recursive fuzzy c-means clustering for recursive fuzzy identification of time-varying processes [J].

Dovzan, Dejan ;

Skrjanc, Igor .

ISA TRANSACTIONS, 2011, 50 (02) :159-169

[5] Control chart pattern recognition using K-MICA clustering and neural networks [J].

Ebrahimzadeh, Ataollah ;

Addeh, Jalil ;

Rahmani, Zahra .

ISA TRANSACTIONS, 2012, 51 (01) :111-119

[6] A New Approach of Modeling an Ultra-Super-Critical Power Plant for Performance Improvement [J].

Hou, Guolian ;

Yang, Yu ;

Jiang, Zhuo ;

Li, Quan ;

Zhang, Jianhua .

ENERGIES, 2016, 9 (05)

[7]

Hou GL, 2011, C IND ELECT APPL, P1308, DOI 10.1109/ICIEA.2011.5975789

[8] A novel KFCM based fault diagnosis method for unknown faults in satellite reaction wheels [J].

Hu, Di ;

Sarosh, Ali ;

Dong, Yun-Feng .

ISA TRANSACTIONS, 2012, 51 (02) :309-316

[9] Effective fuzzy c-means clustering algorithms for data clustering problems [J].

Kannan, S. R. ;

Ramathilagam, S. ;

Chung, P. C. .

EXPERT SYSTEMS WITH APPLICATIONS, 2012, 39 (07) :6292-6300

[10] A fast iterative recursive least squares algorithm for Wiener model identification of highly nonlinear systems [J].

Kazemi, Mandi ;

Arefi, Mohammad Mehdi .

ISA TRANSACTIONS, 2017, 67 :382-388

← 1 2 3 →