A Method Of Online Condition Monitoring Of Electrical Equipment Based On Multiple Change Points Detection And Template Matching

With the advent of the era of big data and the development of smart grid,electricity consumption in various industries is gradually becoming intelligent.The online condition monitoring technology of electrical equipment is a technology to study the running state of electrical equipment,which can realize the fault diagnosis of electrical equipment and the analysis of user’s electrical behavior,and it is becoming a research hot spot in the field of smart power.At the present stage,the research on this technology has been relatively mature,but there is still room for further optimization in the aspects of algorithm time consumption and detection accuracy.To solve the above problems,The multiple change points detection technology and template matching strategy are used to analyze the electricity consumption data of electrical equipment,in order to quickly and accurately obtain its running state and state switching time.Firstly,an online condition monitoring method for single electrical equipment is proposed.This method uses two dimensional template matching strategies of buffer and sliding window to realize the running state identification and state switching time location of electrical equipment respectively.In order to further improve the accuracy of state monitoring,the random overlapping strategy of buffer and sliding window is introduced to solve the problem of missing detection of change points located at the edge of data segments.The feasibility of the strategy is verified by conducting the online condition monitoring experiment on the household refrigerator.Secondly,a template matching method based on finite state machine and autonomous regulation mechanism is proposed.In this method,the state machine model of electrical equipment is established to describe the running state of electrical equipment and the transfer relationship between the states.On this basis,the autonomous adjustment mechanism of buffer size and sliding window size is constructed.The buffer size and sliding window size can be adjusted with the change of the running state of the electrical equipment,which makes the template matching strategy more flexible and efficient.The effectiveness and universality of the autonomous regulation mechanism are verified by conducting comparative experiments.Finally,a parallel online condition monitoring method for multiple electrical equipment based on bus method is proposed.In this method,the power data at the bus end is analyzed to determine whether any of the electrical equipments connected to the bus has changed its running state,and if so,the condition monitoring is performed for each electrical equipment connected to the bus.The effectiveness of the this method is verified by conducting comparison experiments of parallel online condition monitoring on multiple devices.