Design And Key Technology Research Of Intelligent Home Control System Based On Machine Learning

At present,the smart home scheme in the market usually only realizes the simple remote control of household appliances,has not yet fully meet the intelligent requirements.In general,smart home provides home users with an intelligent lifestyle that adapts to user habits,safety and energy saving.Based on the smart home hardware system developed by the team,this paper designs the i OS mobile client based on the MVC development framework and implements functions such as user management,device management,device control,extension functions and message reminders,and successfully submits to the Apple App Store.Based on the consideration of system energy saving,this paper proposes a method based on SVM and Markov model to predict the operating power of smart devices.At the same time,the model is encapsulated into an API interface.The i OS mobile client directly invokes this interface to obtain the predicted operating power of the smart device,and can push abnormal information of the device operating status to the user in real time to discover the smart device with abnormal power usage.Taking the forecast of the operating power of the intelligent new fan as an example,a single SVM model and a combination of the SVM and the Markov model are respectively used to predict the operating power of the intelligent new fan.The average absolute error and the mean square error are selected as the evaluation indexes to obtain the SVM and Markov combinatorial model prediction results more accurate conclusion.At the same time,the API interface response performance has been tested,the results show that the API interface meets the high real-time requirements.The key technology of smart home is also studied in this paper.The designed product realizes various network connection modes between i OS mobile client and smart devices,including Bluetooth,hotspot,LAN and Internet connection.The design of distributed storage of synchronous information realizes multi-client information synchronization.At the same time,using Weex cross-platform development technology to develop client pages,the code was written once,and the synchronization was deployed on i OS and Android mobile platforms.In order to solve the problem of slow loading of Weex pages,a solution to set up client caching mechanism can effectively reduce page loading time.Finally,the whole system has been functionally tested,performance tested and cross platform tested.The test results show that the entire smart home control system meets the design requirements and meets the needs of real life.