Realization Of Intelligent Power Distribution Monitoring System Based On Data Mining Technology

With the development of technology and the improvement of automation,the equipment controlled by computer systems is gradually increasing,and they are continuously being used in data centers and computer rooms such as telecommunications and government.With the continuous increase in the number of various types of electrical equipment With the increase,the failure rate of the power distribution system and the risk of safety accidents are increasing,and the difficulty of personnel maintenance is also increasing sharply.Therefore,a power distribution monitoring system that is matched with these high-performance equipment is required.In order to solve the above problems,this paper implements an intelligent power distribution monitoring system based on data mining technology.This system solves the shortcomings of traditional power distribution monitoring systems,improves the intelligent level of power distribution in the computer room,and makes the power distribution system more stable and Reliable,meeting the needs of future computer room power distribution.The main research contents of this article are:First,by consulting a large amount of literature,the research background and significance of the distribution monitoring system are explained,and the current research status of the distribution monitoring system,data mining technology and random forest algorithm is analyzed,and an improved random forest algorithm based on mixed sampling is proposed.,And based on the functional requirements and design principles of the intelligent power distribution monitoring system,the overall design plan of the system is given.Secondly,the software and hardware design of the system mainly combines the modular design idea to divide the whole system into several modules according to function.The hardware part discusses the realization of clock circuit,storage circuit,power circuit,communication circuit and voltage and current acquisition circuit;software;The part discusses the realization of the main program,storage program,communication program and human-computer interaction interface.Then,according to the requirements of the intelligent power distribution monitoring system for the fault warning function,a power equipment fault prediction model based on data mining technology is proposed.The fault prediction model is constructed with an improved random forest algorithm for the characteristics of the unbalanced data collected by the early warning platform.Experiments show that the fault prediction model constructed by the improved random forest algorithm has improved overall performance and can be used in the system.Complete failure warning work.Finally,through the overall test of the system,the hardware performance,measurement accuracy,communication and other functions of the system have all reached the overall design requirements.The intelligent power distribution monitoring system designed and implemented in this paper has reached the expected design goals,has the advantages of strong real-time,high security,and good stability,and meets the requirements of the use of power distribution in the computer room.