The Research And Application Of Fault Alarm System For Power Information Communication Equipment

With the increase of the number and types of communication equipment, the types and quantities of the faults have also begun to grow rapidly. How to analyze and process the fault data of the information communication equipment in order to obtain useful knowledge and help equipment maintenance has become a key issue. At the same time, with the increase of the number and types of communication equipment, network boundary continues to extension, and the weaknesses of various of systems and equipment cannot be accurately positioned. How to manage and configure the information and communication equipment and analyze all kinds of logs that generated by the operation of the equipment in order to accurately measure and control the risk which is also become a key issue. This has brought many problems to the traditional fault warning system, such as more source of the warning information, low system data relevancy, low processing ability of mass information, timeliness of processing, and inaccurate of location. The emerging technologies, such as effective equipment configuration management, big data analysis, are the key technologies to solve these problems. It not only can measure the weak link in the information communication equipment through the system of the equipment configuration management, but also can predict equipment failure and security risks from a large number of historical data and alarm logs, so it has a very high significance to improve the efficiency of maintenance management of information communication equipment.For the above problems, the first step is to build a platform of big data. The mainline of this platform are data acquisition, data storage, and data processing, and in the platform the data acquisition layer, data storage layer, data processing layer, application layer are loosely coupled architecture, internal each module is achieve the standardization of interface and integrated model, and based on the forcibly unified B/S service mode, the system present the operation of application and management through the web interface in order to create a integration platform of big data as the basis of the fault warning system.And then in this paper we depth study the outlier algorithm in the fault warning system. We select a partial outlier algorithm that based on the density which used as a priority content in the paper, and after analyzing the advantages and disadvantages of the algorithm and already exist neighborhood query optimization schema, the neighborhood query optimization has been done, and reduce the range of the neighborhood query, and reduce the data preprocessing time. And then, based on the study of MapReduce computing framework, we improve the algorithm for parallel and implement on the big data platform.Finally, we build a fault diagnosis system as application layer based on the big data platform to combine the system and the platform, and make full use of the advantages of big data technology in mass data processing to solve the bottleneck problem of mass data processing in fault diagnoses system.