High-voltage circuit breaker base screw loosening fault detection method

Abstract

The invention relates to a high-voltage circuit breaker base screw loosening fault detection method, which comprises the following steps of: introducing two vibration sensors, and constructing a weighted segmented vibration difference operator to measure the vibration difference between the two sensors; and further measuring the distance between the vibration difference data and the high-voltage circuit breaker in a normal operation state to judge whether the high-voltage circuit breaker base screw has a loosening fault or not. According to the invention, the installation positions of the twovibration sensors and the material of the high-voltage circuit breaker connecting base are not strictly required, and the limitation of equipment assembly conditions is reduced; meanwhile, only the data of the vibration sensor during normal operation need to be collected in terms of data and cannot be bound with a specific base screw loosening sample, the detection success rate is increased, and safe and stable operation of the high-voltage circuit breaker is guaranteed.

Description

technical field [0001] The invention relates to a fault detection method for a high-voltage circuit breaker in a power system, in particular to a fault detection method for a loose base screw of a high-voltage circuit breaker, and belongs to the technical field of safety protection devices for a high-voltage power system. Background technique [0002] High-voltage circuit breakers can cut or close high-voltage lines. As one of the core components of the power system, it plays a key role in the control and protection of lines. The safe operation, maintenance and overhaul of high-voltage circuit breakers are closely related to the normal operation of the entire power grid system. The operating mechanism of the high-voltage circuit breaker is prone to mechanical failures, such as the vibration of the base screws during long-term operation; the loosening of these screws will further enhance the vibration of the high-voltage circuit breaker and reduce its service life, and even a...

AI Technical Summary

Problems solved by technology

The operating mechanism of the high-voltage circuit breaker is prone to mechanical failures, such as the vibration of the base screws during long-term operation; the loosening of these screws will further enhance the vibration of the high-voltage circuit breaker and reduce its service life, and even affect the cutting or closing of the high-voltage circuit breaker The operation causes the switch to refuse to move, so it is necessary to check whether the base screws of the high-voltage circuit breaker are loose in a timely and accurate manner

[0003] Because the location of the high-voltage circuit breaker is dangerous, it is usually not suitable for high-frequency manual inspection; for the automatic detection of the loosening of the base screw of the high-voltage circuit breaker, the current mainstream method is to use the vibration sensor to collect the vibration of the base screw data, and use artificial intelligence models to predict based on the characteristics of the vibration data. This kind of method is usually more successful in laboratory simulation experiments, but problems will arise in the actual implementation process: (1) Due to the vibration and the direction of the installation position of the high-voltage circuit breaker , altitude, air pressure, and the material of the fixed base are all related, and the frequency and amplitude of the vibration of the high-voltage circuit breaker are also related to the location of the vibration sensor. Therefore, to make the method obtained in the laboratory effective, it is necessary to strictly require the assembly location and location of the equipment. process; when a certain condition affecting the vibration changes, the traditional intelligent model needs to build an experimental data set to adapt to it; and the laboratory simulation capability is limited, it is impossible to exhaust all the possibilities that arise during installation, so the traditional model In the actual use process, it will often fail, and there will be a large number of false positives and false negatives.

(2) The traditional method needs to construct loose and non-loose samples of high-voltage circuit breaker base screws in the laboratory. Since the samples of loose screws cannot exhaust all loose situations, neural networks, decision trees, and support vector machines are used in this case. Such shallow artificial intelligence models may cause the model to be bound to specific loose samples, which leads to low predictive ability in practical applications; some deep neural network models add high-level feature extraction functions to prevent binding, but such The neural network needs relatively powerful computer or server support; and the location of the high-voltage circuit breaker is usually difficult to place these servers, and the high-frequency vibration information is uploaded to the server. Additional network and storage devices are required, and these devices increase the overall maintenance and construction of the power grid cost

Images