Neural modelling of steam boilers

被引:60
作者
Rusinowski, Henryk [1 ]
Stanek, Wojciech [1 ]
机构
[1] Silesian Tech Univ, Inst Thermal Technol, PL-44100 Gliwice, Poland
关键词
empirical modelling; neural network; steam boiler;
D O I
10.1016/j.enconman.2007.06.040
中图分类号
O414.1 [热力学];
学科分类号
摘要
The paper presents a method and example results of calculations of the neural modelling of steam boilers. Empirical models can be worked out based on the results of specially organised measurements or continuous measurements recorded in the computer system storing the operational performance. The introduction of operational measurement data for material and energy balances required the separation of stationary sub-periods of boiler operations. For each separated sub-period of stationary operation thermal calculations based on DIN 1942 have been carried out. The results of calculations are utilized to estimate the neural model of a steam boiler. This model describes the dependence of the main operational parameters of the boiler upon the flue gas losses and losses due to unburned combustibles. The parameters of the neural model have been estimated by means of the back-propagation method. (C) 2007 Elsevier Ltd. All rights reserved.
引用
收藏
页码:2802 / 2809
页数:8
相关论文
共 19 条
[11]  
LEWANDOWSKI J, 1999, PROBLEM THERM ENG ME, V181, P152
[12]  
McCulloch W.S., 1943, Bulletin of Mathematical Biophysics, V5, P115
[13]   Power plant condenser performance forecasting using a non-fully connected artificial neural network [J].
Prieto, MM ;
Montañés, E ;
Menéndez, O .
ENERGY, 2001, 26 (01) :65-79
[14]   Neural network for evaluating boiler behaviour [J].
Romeo, Luis M. ;
Gareta, Raquel .
APPLIED THERMAL ENGINEERING, 2006, 26 (14-15) :1530-1536
[15]  
Rusinowski H., 2003, IDENTIFICATION COMPL
[16]  
RUSINOWSKI H, 2006, INT CARP CONTR C ICC, P477
[17]  
Rusinowski H, 2006, ECOS 2006: PROCEEDINGS OF THE 19TH INTERNATIONAL CONFERENCE ON EFFICIENCY, COST, OPTIMIZATION, SIMULATION AND ENVIRONMENTAL IMPACT OF ENERGY SYSTEMS, VOLS 1-3, P389
[18]   Development of a neural network based integrated control system of 120 ton/h capacity boiler [J].
Saha, PK ;
Shoib, M ;
Kamruzzaman, J .
COMPUTERS & ELECTRICAL ENGINEERING, 1998, 24 (06) :423-440
[19]  
Swirski K, 2002, P 18 M THERM, V4, P1229