Research And Development Of Power Transformer Vibration Feature Extraction Algorithm And Cloud Monitoring Platform Software Based On Internet Of Things

The power industry is an important basic energy industry that affects the national economy and the people’s livelihood.The stability and reliability of power transformers greatly affect the operational safety of the entire power grid system.The traditional power transformer fault diagnosis methods need to manually detect the transformer at the substation site,which brings a large labor cost,and the personnel safety and system easy maintenance cannot be guaranteed.Based on the long-term research and achievements and accumulation of transformer vibration characteristics,this paper further studies and discusses the vibration characteristics of the transformer and its extraction algorithm.The original vibration feature quantity and its extraction algorithm(frequency complexity)are Improvements in effectiveness and applicability have been made and a new vibration feature quantity(vibration vector variance)has been proposed.On this basis,using the increasingly mature network technology,database technology,server technology and mobile terminal application technology,the online cloud monitoring platform of power transformer based on the Internet of Things is designed and developed,and the improved and proposed vibration characteristics are applied online.Monitoring and diagnosis.Combined with the smartphone app and signal acquisition device(hereinafter referred to as real-time monitoring terminal),the cloud monitoring platform can realize 24-hour transformer vibration data storage,processing,real-time diagnosis,result display,user management,terminal equipment management,and the like.The research content of this topic is as follows:1、Transformer vibration characteristic value extraction algorithm.After in-depth study of the vibration characteristics of the transformer,the vibration eigenvalues(frequency complexity)proposed by the research group are discussed.The shortcomings and causes of the original feature algorithm are analyzed.The window function method is used to improve the probability distribution of the weight coefficients in the original algorithm.Improve the effectiveness and applicability of the feature algorithm.At the same time,through the study of the amplitude and phase relationship between multiple measuring points,a new vibration characteristic quantity(vibration vector variance)is proposed,and comparative experiments are carried out to verify its effect.2、Design of cloud platform data acquisition software.The cloud platform data collection software is the background server of the data collection device and is used together with the data collection device.Parameter configuration,start-stop control,system update,data upload and other operations can be performed on the data acquisition device.3、Design of cloud platform data application software.The cloud platform data application software is the background server of the mobile client software.The mature background development framework is used to realize functions such as user management,device status management,transformer information management,measurement waveform and result display,and log management.Through rigorous functional testing and field actual testing,the designed power transformer online cloud monitoring platform meets the design requirements in terms of functions and technical performance indicators,and has been put into practical use.