Non-intrusive Electric Safety Monitoring System Base On Internet Of Things And Cloud Platforms

In recent years,the level of electricity consumption in our country has been greatly improved.The large-scale application of electric energy has brought great convenience to people's life,but it also causes a large number of electric fire safety hazards.The illegal application of electrical appliances and the arc fault caused by circuit failure are the main causes of electric fire.In order to prevent electrical fire,scholars have done a lot of research on electrical monitoring and fault identification.But,most of the research is limited to theory of algorithm.At the same time,few of algorithms are put into practice to build a complete system used in actual production and life.In the aspect of arc fault monitoring,there is arc fault circuit interrupter(AFCI)put into use,but its function is single and the interrupter does not have the networking function so that it cannot realize remote monitoring.So,combined with cloud platform,based on internet of things(IOT),cloud computing and non-intrusive monitoring technique,an electrical safety monitoring system,which used for remote monitoring for indoor electric and fault arc,was designed and developed in this paper.This paper has mainly done the following several work:First,aiming at the identification and monitoring of electrical appliances,this paper adopted Fourier harmonic analysis feature extraction algorithm to extract the 15 harmonic of electrical appliances as characteristic parameters of electrical appliances and designed cascade classification algorithm used for electrical appliances identification by combining back propagation(BP)neural network algorithm and k-nearest neighbor(KNN)algorithm;Second,aiming at the fault arc monitoring of the circuit,this paper put forward the autoregressive(AR)model arc fault characteristic parameter extraction algorithm after in-depth study of current waveform characteristics of common electric appliances in the normal working and fault arc.Based on the extracted fault arc characteristic parameters,the paper proposed the fuzzy fast fault arc recognition algorithm and the fault arc detection algorithm based on the support vector machine(SVM).Third,a non-intrusive electrical intelligent monitoring terminal based on STM32 microcontroller,which mainly used for signal preprocessing and control on the line and a cloud monitoring platform used for algorithm analysis,data storage and query are developed in this paper after completing the design and implementation of the algorithm.Fourth,the terminal and the cloud monitoring platform are integrated in the system,and the system has been functionally tested in this paper.Finally,the test results are analyzed and summarized by the paper.The system design and test results meet the expected goals.The electrical safety monitoring system designed in this paper is a new practical exploration of using IOT technique,electrical appliances identification technique and arc fault detection technique in in actual production and life.The monitoring system has been put into application.The system has been installed in dozens of dormitories of my university for pre running pilot.At present,the system run stably and received high praise,which determined the application value of the system.