Font Size: a A A

Research On Technologies Of User Behavior Prediction And Temperature Setting Decision Of Electric Water Heater

Posted on:2018-06-30Degree:MasterType:Thesis
Country:ChinaCandidate:W Y KongFull Text:PDF
GTID:2392330572965699Subject:Control engineering
Abstract/Summary:PDF Full Text Request
In recent years,rise of the concept of intelligent home appliance reflects the growing requirements of users for comfort in family life.This concept continually put forward the new technical requirements for the home appliance industry,at the same time,pointed out the new development space and development direction for them.For the electric water heater,predicting the user using water behavior,is the key to march toward intelligence.By user behavior prediction,settting temperature for the users' water demand,it overcome the existing shortcomings of longer heating time,larger energy consumption,promoted the users'experiences.It is significance for the study of energy-saving and comfort of intelligent home appliances.This paper studies the user behavior prediction and temperature settting decision of electric water heater.The main research contents and key innovations are summarized as follows:Firstly,analysed the character of user behavior,pre-treated the collected data,put the user behavior into four rank by habits form strong to weak.Used a way based on rules for user behavior pattern recogniton and prediction.Predicted all the rank,and compared the prediction results in different rules.Secondly,reaserch on the prediction of Hide Markov Model,screening the users'historical data based on similarity,predicting the next user behavior state base on viterbi.And compared the predicting results of the four rank of user behavior,verified the stability and the reliability of the Hide Markov Model.Thirdly,using Weighted Markov Chain based on ordered clustering method and mean-standard deviation to predict the using water time.Evaluated the two ways between theory and results.The experimental results showed that the Weighted Markov Chain based on ordered clustering method is better.Fourthly,creating mathematics of the electric water heater in different process,and verified the model,on this foundation,designed two kinds of heating strategy of electric water heater,simulating in physical truth according to different strategies,the simulation results show that after optimised the way of setting temperature,its energy consumption was better.
Keywords/Search Tags:user behavior, pattern recognition, hide markov model, order clustering method, weighted markov chain, temperature setting
PDF Full Text Request
Related items