Fault-identification method for voltage equalize ring base on improved secondary target detection network wire clamp

Abstract

A fault-identification method for voltage equalize ring base on improved secondary target detection network wire clamp is disclosed. The method includes such steps as making image data set of wire clamp or voltage equalizing re, classifying and cleaning image data samples, and labeling the samples as initialization training data. 2, generate an Anchor box according to different detect data sets;S3, initializing a new target detection network by using migration learning, inputting data into the model to start training, and stopping training after the model converges; S4, solidifying the model, removing the training variables contained in the model, and reserving the weight value for detection; Step S5: The model after the data is cured generates a primary and a secondary detection map, and adopts a computer vision algorithm to judge whether a fault occurs for the wire clamp or the voltage equalizing ring. The invention reduces the precision of the secondary detection on the conditionthat the model parameters can reach two models, and at the same time, the detection speed is greatly improved.

Description

technical field [0001] The invention belongs to the field of electric power inspection fault identification, and in particular relates to a fault identification method for unmanned aerial vehicle electric power inspection image line clips and voltage equalizing rings based on an improved secondary target detection network. Background technique [0002] With the increasing size of the transmission network, the workload of power grid operation and maintenance is increasing rapidly, coupled with the complexity of the accompanying terrain and traffic, which has brought great difficulties to manual inspection. The new UAV inspection technology has become a popular technology at present. UAVs are flexible and low-cost in line erection traction and line inspection. It can also find problems such as corrosion of hardware, missing cotter pins and nuts, which are difficult to find by manual inspection. [0003] The fault detection of UAVs applied to power components is generally dire...

AI Technical Summary

Problems solved by technology

However, the process of UAV line inspection will generate a large number of pictures, and it is bound to be inefficient and inaccurate to manually judge whether the equipment is faulty.

On transmission lines, grading rings and clamps become extremely difficult in fault detection due to their small size

In the case of clamp faults, the suspension angle generally exceeds the allowable value of the suspension clamp, resulting in an increase in the additional bending stress, causing the overhead line to be damaged at the exit of the clamp, resulting in broken strands or loose strands

In the voltage equalizing ring fault, since there are many insulators in the high voltage line, but the voltage on each insulator is uneven, so the first insulator close to the wire needs to withstand a large voltage, and the voltage equalizing ring It is the ring fitting for improving the voltage of the insulator. If this component fails, it will cause great harm to the insulator string and cause damage to the power transmission line of the power grid.

Images