Research On The Recognition Method Of Electrical Equipment In Building Network Based On Machine Learning

With the development of the Internet of things technology of the Internet of everything,the development speed of the Internet of things of building equipment is accelerating day by day.The goal is to connect all electrical equipment to the network to achieve measurable and controllable.However,the uncertainty of connecting electrical equipment through smart modules such as smart sockets brings difficulties to equipment management.How to automatically identify the type of equipment in the building networking system is an urgent need for the development of building networking.This paper focuses on the classification and identification of electrical equipment in the office building networking environment,and realizes automatic real-time identification of equipment types according to the electrical parameter characteristics of connected equipment in the building networking environment.Firstly,this paper analyzes the development status of equipment classification and identification methods and the Internet of things system at home and abroad.Combined with the development concept of Internet of things technology and the data basis needed for equipment identification,a set of equipment data acquisition and analysis system based on Internet of things is researched and designed.On this basis,the characteristics of electrical parameters of common equipment are analyzed,and the similarities and differences of electrical parameters of different equipment are explored.Secondly,the realization ideas of existing equipment classification and identification methods are summarized and their advantages and disadvantages are compared.On the basis of studying the architecture of building equipment Internet of Things,a set of data acquisition and analysis system of electrical equipment based on building network is designed and developed,and the electrical parameter characteristics of common equipment and different equipment are studied.This paper studies the combination of the improved K-means clustering algorithm and kNN algorithm to mine the potential features in the data and realize the method of classification and identification of building electrical equipment.Then,the construction process of the equipment classification model using k-means clustering algorithm twice and the equipment recognition model using kNN algorithm is described in detail.K-means is adopted for the preliminary clustering to realize the data of the total harmonic distortion rate of the clustering current.Later clustering improves the similarity distance measurement method by using multiple parameters,and the marking center method is used to optimize the clustering results and improve the classification accuracy.The result of equipment classification model is used as the training data of equipment identification model,and kNN algorithm classification decision rule is used to realize equipment type identification.Finally,the experiment was carried out with the historical data of five common equipment collected by the networked building system built by the team,including water dispenser,desktop computer,microwave oven,heat fan and electric frying pan.The validity of the above algorithm is verified.Compared with traditional methods,the improved classification method has higher classification accuracy.The equipment recognition method can make a quick judgment on the recognition object and has a high recognition rate.