DocumentCode :
635138
Title :
Neural network based model predictive control performance monitoring-data-driven approach
Author :
Lu Wang ; Ning Li ; Shaoyuan Li ; Kang Li
Author_Institution :
Dept. of Autom., Shanghai Jiao Tong Univ., Shanghai, China
fYear :
2013
fDate :
23-26 June 2013
Firstpage :
1
Lastpage :
6
Abstract :
A data-driven neural network based approach for model predictive control performance diagnosis was proposed. Considering four common MPC degradation factors, namely noise variance change, model mismatch, control variables constraint saturation, and manipulated variables constraint saturation, MPC performance patterns were divided into four categories. Performance signatures are extracted from the process input and output variables directly, and classifier is constructed via neural network. The effectiveness of the proposed method was demonstrated on NIAT platform by a two tank liquid level process.
Keywords :
chemical industry; level control; neurocontrollers; predictive control; tanks (containers); MPC degradation factors; MPC performance patterns; NIAT platform; manipulated variables constraint saturation; model mismatch; model predictive control performance diagnosis; neural network; neural network based model predictive control performance monitoring-data-driven approach; noise variance change; performance signatures; two tank liquid level process; Benchmark testing; Monitoring; Neural networks; Noise; Predictive control; Valves; NIAT platform; data-driven; model predictive control; neural networks; performance diagnosis; performance monitoring;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control Conference (ASCC), 2013 9th Asian
Conference_Location :
Istanbul
Print_ISBN :
978-1-4673-5767-8
Type :
conf
DOI :
10.1109/ASCC.2013.6606358
Filename :
6606358
Link To Document :
بازگشت