Research On Online Monitoring And Temperature Prediction Technology Of Overheat Fault In Ring Main Unit

Ring main unit(RMU)is an important power equipment in distribution network,which is widely used to realize regional closed loop power supply.However,due to various factors,the ring network cabinet is prone to abnormal temperature rise.If it evolves into a serious fault,it may lead to the disconnection of the ring network and affect the reliability of power supply.Most of the ring network cabinets are installed outdoors,with large number,wide distribution and narrow space in the cabinet.It is difficult to find and deal with the faults in time by the traditional monitoring method applied to the switch cabinet in the station.Therefore,it is of great significance to study how to accurately and comprehensively perceive the temperature and other important parameters of the key parts of the ring main unit,and predict the temperature change on this basis to evaluate the development trend of overheating fault.The main work and achievements are as follows:Firstly,based on the theory of heat transfer,the heating mechanism of RMU is studied,the advantages and disadvantages of various temperature measurement methods and wireless communication technology are analyzed and compared,and the overall structure of online monitoring system is designed according to the concept of ubiquitous power Internet of things.The causes of overheating and heat transfer process are analyzed,and the heat balance equation of key contact parts of the equipment is derived,and the main influencing factors of temperature change are summarized.According to the special working environment and the requirements of high and low voltage isolation in the cabinet,the non-contact small infrared temperature sensor is selected,and the load current and environmental temperature and humidity monitoring are integrated,so as to avoid the shortcoming of single monitoring state.The cloud platform of Internet of things is used as the background monitoring center,which improves the data transmission efficiency,interactivity and perception,and is more suitable for the application scenarios of RMU.Secondly,the software and hardware design of the on-line monitoring system of RMU is studied and completed.The modular scheme is adopted,which is mainly divided into two parts: wireless monitoring node and data concentrator.The hardware circuit connection schematic diagram is drawn,and the working principle of each part of the circuit is described.According to the functional requirements,the program flow of wireless monitoring node and data concentrator is compiled respectively.For the low-power function,the timing wake-up sampling,simplex based limited two-way transmission and other technical schemes are specially designed.According to the system structure,the software program of the IOT monitoring platform is designed,and the basic functions of the system are tested to verify the reliability and effectiveness of the software and hardware design.Finally,on the premise of completing the research of online monitoring of ring network cabinet,a method of temperature prediction is proposed by using LSTM network in depth learning to integrate multi-dimensional monitoring data.The time series characteristics of temperature data are analyzed,and the dynamic indexes of thermal fault discrimination and trend analysis based on predicted temperature are studied.The historical temperature,load current and ambient temperature are used as model inputs instead of single temperature history data,and more information is fully mined to optimize the prediction effect.The gate mechanism principle and error inverse delay calculation process of LSTM network are studied.The realization steps of prediction model are given and an example is analyzed under Matlab simulation environment.The model evaluation system is established and compared with the traditional BP neural network prediction results.The results show that the scheme of online monitoring system and the design of software and hardware in this paper are reliable and effective,and more suitable for the application scenarios of ring main unit.The proposed temperature prediction model has better prediction ability than traditional methods,which provides practical reference and theoretical support for solving the safe operation problems of RMU.This paper has 49 pictures,9 tables and 80 references.