On-line Monitoring Method For Equipment Status Of Distribution Internet Of Things Terminal

As an important part of the ubiquitous power Internet of Things,the distribution Internet of Things can further improve the automation and informatization level of the distribution network.As the sensing unit of the sensing layer of the distribution Internet of things,the low-voltage terminal unit,the reliability of its measurement data has a great impact on the fault processing and advanced applications of the distribution Internet of things.In the large amount of measurement data obtained by the low-voltage terminal unit in the actual distribution network,there are inevitably some sudden changes in the measurement data caused by various reasons,such as failures in the primary system of the low-voltage distribution network,secondary low-voltage terminal unit faults,data transmission or storage faults,etc.,the data changes caused by the primary system fault need to be reported to the master station for alarm and countermeasures should be taken in time,and the false alarm and omission of measurement data caused by the secondary monitoring equipment fault will seriously affect the decision-making of the power distribution Internet of Things reliability,and even lead to safety accidents,requiring timely maintenance.How to identify and extract these data from the measurement data of low-voltage terminal unit,and judge whether the working status of terminal equipment is normal by distinguishing the source of the change data is of great significance for improving the operational reliability of the power distribution Internet of Things system and the management level of system equipment.Aiming at the problem of data abnormality detection of power distribution Internet of Things low-voltage terminal unit,an online detection method of power distribution Internet of Things low-voltage terminal unit data abnormality based on clustering is proposed.First of all,in view of the problem that the traditional similarity measurement method has poor effect on the similarity measurement of high-dimensional data,the local similarity and global similarity of the measurement sequence are comprehensively considered,and the similarity distance measurement criterion of compound time series is proposed,which improves the accuracy of the input data when clustering.Then,in order to overcome the problem that the traditional DBSCAN algorithm is sensitive to the selection of global density parameters,a DBSCAN algorithm for adaptive generation of global density parameters is proposed.which uses the preprocessed historical measurement data of low-voltage terminal unit for a specified period of time to adaptively generate global density parameters,and obtain the core data points through clustering training to detect the mutation data in the measurement data.Compared with the actual simulation results,the improved DBSCAN algorithm has a lower false alarm rate and higher practical value when the detection rate is higher.Aiming at the problem of identifying the abnormal source of the mutation data of the low-voltage distribution network of the distribution Internet of Things,a fuzzy logic-based method for identifying the abnormal source of the data of the low-voltage distribution network of the distribution Internet of things is proposed.First of all,using the spatial correlation of D-Io T data,the sliding time window algorithm is used to calculate the spatial cross-correlation coefficient between the low-voltage terminal unit suspected of abnormal data and its neighboring low-voltage terminal unit,and the geometric features of the spatial cross-correlation coefficient are extracted as fuzzy logic.system input.Then,in view of the problem that it is difficult to obtain the evaluation of the spatial correlation of nodes through a certain quantitative calculation equation,a fuzzy logic algorithm that can process imprecise information based on fuzzy set theory is selected,and combined with the spatial and temporal correlation of the distribution Internet of Things data,a cascaded fuzzy logic system is designed to evaluate the degree of spatial correlation between low-voltage terminal unit.By evaluating the degree of spatial correlation between the low-voltage terminal suspected of abnormal data and its neighboring low-voltage terminal,it can confirm whether the changed data comes from a primary system fails or secondary monitoring equipment fails.If the monitoring equipment fails,the operation and maintenance personnel should be reminded to perform timely maintenance.Finally,the comparative simulation results show that the method has obtained high accuracy in the D-Io T network with different distribution densities of low-voltage terminal unit,and can correctly confirm the working status of end devices online.In summary,this paper studies the online monitoring method for equipment status of the distribution Internet of Things terminal,which provides a feasible method for the status monitoring of the distribution Internet of Things low-voltage terminal unit.