In recent years, the problem of aviation safety has been paid more and more attention. The ac fault arc caused by the destruction of the aerial cable is one of the key factors causing the problem of aviation safety.

The material of the insulation layer of the airport cable is affected by friction, ground stress and other factors, which leads to the physical damage of the cable sheath. In addition, temperature, damp and cold, organic chemical corrosion and other reasons will lead to the decomposition of the cable sheath aging, weakening of the insulation layer, and the probability of electric breakdown expanding, thus causing intermittent fault arc.

With the continuous increase of the volume of the power supply and distribution system at the airport, many changes of the total number of cables and the natural environment of electrical equipment with complicated structure are generated. The electricity safety problem caused by fault arc becomes more and more frequent. The secrecy of fault arc makes it difficult to be tested.

Therefore, it is of practical significance to scientifically study the characteristics of fault arcs and to define a reasonable identification method for ac fault arcs in order to ensure the safety of aviation electricity.In this paper, the characteristics of the fault arc caused by the change of the line current, the characteristics of the current extraction analysis from different perspectives, and the identification of the aviation ac fault arc scientific research.

First, carry out aviation fault simulation experiment according to the code, obtain all normal current statistics and fault arc current statistics under different loads necessary for analysis, and prepare for feature extraction in advance;Secondly, the transformation of the current characteristics of the front and back and left and right lines caused by arc is studied scientifically from three aspects: the difference of current wave pattern, the concept of data signal model and the feature extraction of special frequency segment.

The results found that the fault arc occurred, the line current you in three main levels and all the normal work of the current is not the same characteristics;The difference between arc current characteristics and all normal current characteristics is affected by load type, current level, switching power frequency and other factors.The current frequency conversion and sampling rate conversion are analyzed for the reliability of the authentication feature extraction method.

Finally, for the problem of fault arc identification under different working standards, a multi-characteristic fault arc identification technology is proposed.The multi-characteristic statistical data vector is built, and the extreme children learning machine is introduced to classify and identify the fault arc, and the characteristics of the classification entity model are analyzed.

Based on the statistical analysis of the evaluation data, it is concluded that the classification entity model established in this paper has a high identification success rate and can provide reliable reference for the identification of aviation ac fault arc.

In the low-voltage power supply and distribution system, transmission lines are often caused by series fault arcs due to virtual connection, aging insulation layer of cable raw materials, poor quality of components and damage caused by external forces.But the arc temperature is very high, can ignite the cable insulation layer skin and other things, easy to cause fire accidents and other safety accidents, resulting in damage to machinery and equipment and accidental injury.

Because the current amplitude of the serial-type fault arc is lower than or equal to the normal current of the load, and is within the rated voltage range of the isolator, it is not easy to be detected by the relay protection device, thus expanding the concealment of the fault arc.