Font Size: a A A

Research On Residential Water Behavior Recognition Methods Of Rural Area

Posted on:2018-09-26Degree:MasterType:Thesis
Country:ChinaCandidate:W H GuFull Text:PDF
GTID:2348330533465913Subject:Electronic and communication engineering
Abstract/Summary:PDF Full Text Request
With the development of economy and the improvement of residents' living standard, the contradiction between water supply and demand is becoming more and more prominent, such as population growth,environmental pollution and urbanization. The researches of rural residents' water consumption can improve the water saving consciousness and make the current situation of water resource management better. A method of identifying rural residents' water consumption is put forward in this paper. It can accurately identify the behavior of rural residents' water consumption and enhance the water infrastructure. The paper studies the method through the analysis the flow characteristics of several typical water use behavior of residents. The specific works are as follows:The flow characteristics of different water use behavior were extracted from the trainig set. The left-right Hidden Markov Model (HMM) was selected to establish the identification model of different types of residents' water. The test data were input into the trained model to identify the water consumption of residents/ Then, according to the water flow of residents, the water events can be identified at this time. In order to improve the recognition result of HMM,the HMM model and the time probability function were combined to obtain the recognition result of this method.First of all, The Artificial Neural networks (ANN) algorithm was selected. The structure of BP neural network (Back Propagation, BP) network was designed and the BP network training parameters were determined. Then the BP neural network was used to establish the identification model of water consumption. Finally, the test data were input into trained model to identify the residents of water consumption and the results were obtained.The results show that the combination model of HMM and time probability function can be used to obtain more accurate identification results for different flow patterns; the BP neural network model can be used to identify the water behavior in the water flow with similar flow pattern.
Keywords/Search Tags:Residential water, Behavior recognition, Hidden Markov Model, BP Neural Network
PDF Full Text Request
Related items