Fault Location Mechanism Of Integrated Data Network Based On Interactive Active Detection

In the smart grid,the integrated data network carries a variety of power information management services.With the continuous improvement of the information level of power grids,the integrated data network of electric power has gone deep into various departments and positions in electric power production,operation and management.The requirements for operational support are also getting higher and higher,especially for the detection and positioning of faults.Most traditional network fault management uses the northbound interface to collect the alarm information from the passive collection device,then establishes the alarm-failure association model to analyze the root cause of the fault.Passive waiting for the alarm results in poor timeliness of fault location,and once the alarm information is false,lost and redundant,cannot guarantee the accuracy of fault location.In order to make up for the lack of traditional network management,active detection technology is gradually introduced to the operation and maintenance of data communications networks.Therefore,it is of great significance to carry out researches on fault location of power integrated data network by using active probing technology and to improve management and operation of power integrated data network.In view of the accuracy and efficiency requirements on fault location in the integrated data network for power system,an interactive fault location mechanism is proposed.To resolve the fault diagnosis set selection problem in the mechanism,an interactive probing choice algorithm(IPCA)is proposed.A Bayesian model is built for the nodes of power integrated data network and the candidate set of location probes.The set is divided into several subsets with the help of conditional independence characteristic of Bayesian networks.The detection value is introduced to measure the diagnostic capacity of the probe,and the time complexity of the fault diagnosis set selection is reduced using the submodule characteristic in the interaction update process.Simulation results show that IPCA shortens the location time by 20%on average while ensuring the accuracy of fault location.In the process of fault locating using active detection method for the power integrated data network,in order to reduce the Noisy interference of the fault node identification and improve the efficiency of fault recognition,a fault recognition algorithm based on the dynamic Bayesian model is proposed.First,filter the Noisy to weaken the impact of Noisy as well as reduce the scale of calculation.Then,establish the dynamic Bayesian model with the network nodes and detections.Finally,based on this model,a fault node recognition algorithm by the greedy idea named DBFRA is designed.Simulation results show that the proposed algorithm can reduce the false positive rate of fault recognition while satisfying the real-time locating requirements.In this paper,we study the interactive active probing fault location mechanism,especially the fault diagnosis set selection algorithm and fault identification algorithm,improves the accuracy and timeliness of fault location,which is of great significance to intelligent and efficient fault management of power integrated data network.