Research On Visual Monitoring Of Synchronous Generator Based On Spark Streaming

With the continuous increase of power grid capacity in China,more and more power equipments are connected,which brings more challenges to the stability of power grid.Synchronous generator is an important equipment for stable and safe operation of power supply system.How to use its monitoring data for accurate and efficient visual display is of great significance to the real-time control of generator and power plant operation management.The traditional synchronous generator monitoring system is simple in visualization form,and the monitoring data is only displayed in simple tables or curves,from which the monitoring personnel cannot get useful information.In particular,when the generator fails,many monitoring data change,which often requires professional personnel to analyze the monitoring data in order to make fault judgment,and the monitoring personnel cannot understand the actual operating conditions of the generator unit.In view of the above problems,this paper studies the structure,common faults and visual monitoring methods of synchronous generator,and designs and implements a new visual monitoring system based on Web 3D technology.The system visualizes the generator set in the way of three-dimensional solid model,so that the monitoring personnel can intuitively understand the operation conditions of the generator set,and solve the problem that the visual monitoring system cannot view the real scene.For the three-dimensional modeling of synchronous generator set,this paper creatively uses the method of Echart GL to build the model,which simplifies the three-dimensional model,simulates the main parts of the generator model with curved surface,and renders directly on the web page,which not only reduces the workload of model rendering,but also improves the dynamic adjustment speed of the model when the data changes.The three-dimensional model includes the main parts of the generator set such as stator,rotor,collector ring and main shaft.The real-time working condition data is mapped to the color,action and prompt box text of the corresponding module of the three-dimensional model.When the monitoring data changes,the model is dynamically adjusted,so that the monitoring personnel can view the operation condition of the synchronous generator through the model at any time.The paper also carries on the dynamic visualization of various charts for the monitoring data,including the visualization of real-time data,operation characteristics and statistical data,so as to realize the full display of the monitoring data,enrich the diversity of the visualization methods of the monitoring system,and further improve the practicability of the visualization monitoring system.This paper uses the data flow processing framework based on spark streaming to process the monitoring data.This framework uses flume system to collect the monitoring data,Kafka distributed system to gather and transmit,spark The processing core transforms and stores the monitoring data,which can process the massive real-time data at high speed.It not only realizes the real-time collection and processing of the monitoring data,but also meets the high real-time requirements of the synchronous generator visual monitoring system.Finally,the paper tests the real-time performance of the visual monitoring system in local and remote,including the rendering time of graphics and the dynamic adjustment and refresh speed test.The results show that the system meets the real-time requirements of visual monitoring.