A pattern recognition method for discriminating thermal structures in reservoirs
In this paper, taking different types of thermal structure in reservoirs as different modes, a pattern recognition method and its algorithm for the discrimination and prediction of reservoir thermal structures are put forward. Rather than directly describing the stratified mechanism of water temperature in reservoirs, the mode discrimination functions are derived by extracting and generalising the classification rules from practical category features of thermal structure in reservoirs using the perception approach. This approach, in which no assumptions were involved, totally depends on the basic reservoir features to get an appropriate discrimination function. As a case study, 29 reservoirs in the north of Hubei Province, China, are used to form the training mode samples set to derive the discrimination functions for reservoir thermal structure type. Another 6 reservoirs, which are not included in the training mode samples set, are used to validate the discrimination functions. The results show that all the thermal structure types of these 6 reservoirs are discriminated correctly. The approach of this paper has the great advantage of simplicity and can be widely applied.
reservoirs thermal structure pattern recognition discrimination prediction
Xiaohong CHEN
Department of Water Resources and Environment,Sun Yat-sen University,Guangzhou 510275,China
国际会议
The Four Conference of Asia Pacific Association of Hydrology and Water Resources(亚太地区水文水资源协会第4届科学大会)
北京
英文
216-222
2008-11-03（万方平台首次上网日期，不代表论文的发表时间）