ABSTRACT
To deal with the adverse effects of multi-input and multi-disturbance on thermal plant model identification, a thermal closed-loop process identification method integrating machine learning is proposed. Firstly, the optimal method of identification data based on feature classification machine learning is used to select the historical operation data suitable for identification. The method establishes the feature construction rules of historical operation data, reduces the dimension of data and can better represent the dynamic information. The random forest is used to establish the identification data classification rule model, so as to obtain high prediction accuracy of identification model. Secondly, an identification process integrating input variable selection, model order determination and unbiased parameter estimation is proposed. Among them, the variance expansion factor method and the variable projection analysis method of partial least square regression are used to select the input variables of the identification model, and then the model parameters are unbiasedly identified based on the asymptotic identification method. The ML-combined identification method is applied to the deaerator water level system of thermal power unit, and its reliability and accuracy are verified by closed-loop simulation.
- Wang Q, Pan L, Lee K Y, Deep-learning modeling and control optimization framework for intelligent thermal power plants: A practice on superheated steam temperature[J]. Korean Journal of Chemical Engineering, 2021, 38(10): 1983-2002.Google ScholarCross Ref
- Peretzki D, Isaksson A J, Bittencourt A C, Data Mining of Historic Data for Process Identification[C] Proc. of the 2011 Aiche Meeting. 2011.Google Scholar
- Shardt Y A W, Huang B. Closed-loop identification condition for ARMAX models using routine operating data [J]. Automatica, 2011, 47(7):1534-1537.Google ScholarDigital Library
- Shardt Y A W, Huang B. Closed-loop identification with routine operating data: Effect of time delay and sampling time [J], Journal of Process Control, 2011, 21(7):997-1010.Google ScholarCross Ref
- Breiman L. Bagging Predictors[J]. Machine learning, 1996, 24(2): 123-140.Google Scholar
- Zhang L, Meng X.L. Data Center Cooling System Temperature Prediction Based on Bayesian Random forest [J]. Automation and Instrumentation, 2023, No.283 (05): 6-9. DOI: 10.14016/j.cnki.1001-9227.2023.05.006Google ScholarCross Ref
- Partha Pratim Roy, Kunal Roy. On Some Aspects of Variable Selection for Partial Least Squares Regression Models[J]. Molecular Informatics, 2010, 27(3):302-313.Google Scholar
- Gustavsson I, Ljung L, T. Söderström. Identification of Processes in Closed Loop-Identifiability and Accuracy Aspects[J]. 1977.13(1):59-75.Google ScholarDigital Library
- Lung L. System Identification: Theory for the User (2nd Edition )[M]. Prentice-Hall, Englewood Cliffs, N.J,1999.Google Scholar
- Zhu Y, Butoyi F. Case studies on closed-loop identification for MPC [J]. Control Engineering Practice, 2002, 10(4): 404-417.Google ScholarCross Ref
- Zhu Y.C Multivariable system identification for process control [M] Changsha: National University of Defense Science and Technology Press, 2005Google Scholar
- Zhu Y. Use of error criteria in identification for control [J]. IFAC Proceedings Volumes, 2000, 33(15):307-312.Google ScholarCross Ref
- Richalet J. Industrial applications of model based predictive control [J]. Automatica, 1274. The role of system1993. 29(5):1251Google Scholar
Index Terms
- A ML-combined closed-loop identification method for thermodynamic process
Recommendations
A laboratory experiment to teach closed-loop system identification
In this paper, a laboratory experiment, which was developed to teach undergraduate students at the University of Pretoria, Pretoria, South Africa, some of the main issues regarding closed-loop system identification, is discussed. In this experiment, a ...
Indirect closed-loop systems identification of distributed systems
In many cases it is impossible to remove the feedback during systems identification as it will make the system unstable. This paper presents an identification method for spatially interconnected distributed systems with identical subsystems operating in ...
Subspace identification for closed-loop 2-D separable-in-denominator systems
Identification for closed-loop two-dimensional (2-D) causal, recursive, and separable-in-denominator (CRSD) systems in the Roesser form is discussed in this study. For closed-loop 2-D CRSD systems, under feedback control condition, there exists some ...
Comments