Title :
Data mining-based modeling and application in the energy-saving analysis of large coal-fired power units
Author :
Yang, Yong-Ping ; Wang, Ning-Ling ; Zhang, Zhi-wei ; Chen, De-gang
Author_Institution :
Key Lab. of Condition Monitoring & Control for Power Plant Equip., North China Electr. Power Univ., Beijing, China
Abstract :
The large-sized coal-fired power units characterizes as wide thermodynamic scale, huge equipment, large flow and mass, which results in distinct nonlinear feature in energy transmission, conversion and dissipation for specific equipment, system and process. There´s highly coupling and nonlinear correlation between the energy consumption in power generation and the external environment, resources and load demand. A data mining-based modeling methodology for complex system was proposed in this paper, reflecting the influences of boundary constraints and implementing the reconstruction of operation states. Based on this, a Spatial-temporal Distribution Model of Energy Consumption at Overall Conditions (SDMEC) for large coal-fired power units was built based on ε-SVR data mining and verified by the practical operation data of thermal power units. The result shows that the ε-SVR-based model is easy to implement and explicit to interpret with high accuracy.
Keywords :
coal; data mining; power engineering computing; regression analysis; steam power stations; support vector machines; ε-SVR data mining; coal-fired power unit; energy consumption; energy conversion; energy dissipation; energy transmission; energy-saving analysis; spatial-temporal distribution model; Biological system modeling; Data mining; Data models; Energy consumption; Power generation; Power systems; Support vector machines; Data mining; Energy consumption; Large coal-fired power units; Spatial-temporal distribution;
Conference_Titel :
Machine Learning and Cybernetics (ICMLC), 2010 International Conference on
Conference_Location :
Qingdao
Print_ISBN :
978-1-4244-6526-2
DOI :
10.1109/ICMLC.2010.5580941
