Online monitoring and evaluating method and system for operation state of lightning arrester

Abstract

The invention discloses an online monitoring and evaluating method and system for an operation state of a lightning arrester. The online monitoring and evaluating method comprises the steps of: acquiring online and historical running state data of the lightning arrester; screening the online and historical operation state data to obtain lightning arrester defect characteristic data parameters andchange rules thereof; determining a lightning arrester operation state evaluation index; establishing a lightning arrester operation state evaluation model; and inputting the historical operation state data into the lightning arrester operation state evaluation model for model training, and inputting the online operation state data as a real-time input quantity into the lightning arrester operation state evaluation model for prediction, thereby realizing lightning arrester equipment operation state evaluation and development trend prediction. The online monitoring and evaluating method can bewidely applied to front-line sites such as intelligent substations, plays an important role in improving the test efficiency, enhancing the operation and maintenance level and preventing equipment faults, can be used for evaluating the operation state and the service life of the lightning arrester, can monitor potential hazards of equipment in time, reduces the fault risk of the lightning arrester, and brings huge social benefits.

Description

technical field [0001] The invention belongs to the technical field of monitoring the operating state of an arrester in a power system, and relates to an online monitoring and evaluation method for the operating state of an arrester. Background technique [0002] Metal oxide surge arresters (MOA) are widely used in the protection of lightning overvoltage and operating overvoltage in power systems, providing a reliable guarantee for the continuous and safe operation of the system. In addition to threats such as lightning overvoltage and operating overvoltage, MOA is also affected by external environmental factors such as temperature and humidity, chemical pollution, and pollution. With the increase of running time, MOA will experience aging and deterioration phenomena, usually manifested as a decrease in varistor voltage, an increase in leakage current, and a change in the volt-ampere characteristic curve. In severe cases, thermal runaway may occur, which seriously affects th...

AI Technical Summary

Problems solved by technology

[0005] In recent years, with the construction of smart substations, most power grid companies have adopted a certain number of online monitoring devices for arresters, but the current online monitoring data can only be uploaded to the auxiliary control system in the station, and only have a simple alarm function, and no effective analysis of the data , it is impossible to effectively grasp the important information such as the quality of the arrester manufacturer, the factors affecting the failure, and the source of potential risks.

At the same time, the online rate of existing online devices is less than 10%, and the data accuracy rate is low, and the data availability is poor, resulting in the inability to upload data to the background and the high rate of data mistransmission. The threshold selection method for status assessment and health diagnosis is not yet perfect. The reliability of online monitoring system data and system feedback results is poor

These specifications are usually given by deterministic algorithms. These algorithms rely on historical experience, lack theoretical basis, and cannot fully describe the laws of data. Therefore, the evaluation results are not satisfactory. Therefore, it is necessary to find more effective detection and evaluation methods to monitor The status of MOA remains a research hotspot

[0007] At present, the online monitoring system model based on big data is affected by many factors affected by the internal and external environment of MOA, and the influence of factors that have never appeared or appeared few times cannot be discovered in time. The online monitoring data and system feedback results are credible. Degree still needs to be improved

Images