An Analysis Method of the Energy Consumption Characteristics of the Power Units Based on the ε-SVR

Article Preview

Abstract:

According to multi-borders, nonlinear, time-varying characteristics of the thermal system of large coal-fired power units, the relationships between the operating parameters and the energy consumption characteristics are very complex. The key operating parameters which influenced the standard coal consumption rate are obtained based on rigorous theoretical analysis. On this basis, features are extracted from the characteristics to be used as inputs of ε-SVR for training and testing. Energy consumption distribution model under full conditions of large coal-fired power units based on aforesaid method has high precision.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

2860-2865

Citation:

Online since:

October 2013

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2014 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] Diyong Jie, Wang Guopeng, Han Pu, system identification based on support vector machines [J]. Computer simulation, 2004, 21 (11), 39 ~ 41.

Google Scholar

[2] Wu Hongxiao, PROCEEDINGS, Liu Yong, et al. Immune clustering-based radial basis function network model of short-term load forecasting [J]. Chinese Society for Electrical Engineering, 2005, 25 (16): 53 ~ 56.

Google Scholar

[3] Chen Jian Hung, Li Wei, Sheng Deren. Line performance calculation of a thermal power units in data fusion [J]. Chinese Society for Electrical Engineering, 2002, 22 (5) : 152-156.

Google Scholar

[4] Wang Huijie thermal power units operating parameters of energy consumption sensitivity analysis [J]. Chinese Society for Electrical Engineering, 2008, 28 (29) : 6-10.

Google Scholar

[5] Tat Yan, Hu Qinghua, Bowen. Integration of rough sets and fuzzy clustering, continuous data knowledge discovery [J]. Electrical Engineering of . 2004, 24 (6) : 205-210.

Google Scholar

[6] JWHines, REUhrig, DJWrest. Use of Autoassociative Neural Network for Signal Validation [J]. Journal of Intelligent and Robotic Systems, 1997. 143.

Google Scholar

[7] Zhou Shuxiao Based on Wavelet Denoising and neural network theory of gas-solid circulating fluidized bed particle concentration prediction [J]. Petrochemical . 2003, 32 (3): 224 ~ 229. [J]. Chinese Society for Electrical Engineering . 2004, 24 (7) : 253-256.

Google Scholar

[8] The mathematical model of the virtual sensors in Bo, Hu Nien Su, Zhou Yuyang. Steam Turbine Monitoring and Diagnosis System [J]. China Electrical Engineering . 2004, 24 (7) : 253-256.

Google Scholar

[9] SI Feng-qi, XU Zhi-gao. Based on the self-associative neural network and the measurement data since the calibration test method [J]. Chinese Society for Electrical Engineering . 2002, 22 (6) : 153-155.

Google Scholar

[10] Liu Fuguo. Power plant coal into the furnace element analysis and heat of the soft sensor real-time monitoring [J]. Electrical Engineering of . 2005, 25 (6) : 139-146.

Google Scholar

[11] Zhang Zhiming, Chenghui Tao, Hong Xu, Hu, three high in the neural network ensemble forecasting model and its hair in the steam turbine unit state of repair [J]. Electrical Engineering of . 2003, 23 (9): 204 ~ 206.

Google Scholar

[12] Chen Huayou combination forecasting method and its application: [doctoral thesis], University of Science and Technology, 2002, 9.

Google Scholar

[13] Dongxue Yu, Hu China, Xu Zhi-gao. Power plant performance analysis of sampled data reliability test method [J]. Power Engineering . 1998, 18 (2): 16 to 19, 74.

Google Scholar

[14] Li Jianqiang. Optimization theory and applications of data mining-based power plant operation: [doctoral thesis], North China Electric Power University, (2006).

Google Scholar

[15] Toda, Hiromichi, Yamanaka, Planning and operation performance of 600 MW coal and oil dual-fired boiler for a supercritical sliding pressure operation, [J]. Technical Review - Mitsubishi Heavy Industries, Vol 22, No. 3, pp.225-233, Oct, (1985).

Google Scholar