The method of recognition of regional feature and prediction of the flow data in drainaae oiDe network
The flow variation ofsewer pump stations was influenced by wastewater quantity in upstream drainage network,the catchment area character and rainfall etc．The recognition and prediction ofquantity change character ofregion can assist the measurement and operation of drainage pipe network．The study on method ofrecognition ofregional feature and prediction of variation ofwastewater quantity between the pump stations ofsewer was carded out in the researc The result indicated that it is feasible to recognize the region feature ofthe drainage system with spectrum analysis and statistic analysis．Cluster analysis．neural network pattern recognition and neural network clustering can effectively classify the pump station according to the flow data． Using the GA．BP neuraI network can get a better performance than using the BP neural network only in predicting the flow change of WWTP and pump station．the neural network time series call also forecast the flow and the relative error is within the acceptable class．
drainage system region feature neural network wastewater collection
Mingkai Zhang Yanchen Liu He Jing Yong Qiu Zhiqiang Wang Hanchang Shi
School of Environment, Tsinghua University, 100084, Beijing, China
国际会议
北京
英文
109-110
2011-11-24（万方平台首次上网日期，不代表论文的发表时间）