Arc Fault Detection in Aviation Cables Based on Multi-Entropy Features and PSO-RF Algorithm

Aiming at the problem of high concealment and difficult detection of fault ares in aviation line systems, a method for detecting arc faults in aviation cables based on multi-entropy features and Particle Swarm Optimisation-Random Forest (PSORF) algorithm is proposed. Firstly, an experimental platform is built to simulate the arc fault of aviation cable and collect the current data, wavelet decomposition is used to decompose the arc fault current data, and then energy entropy, fuzzy entropy and approximate entropy are extracted from the current component to construct the feature vector. Finally, the PSO-RF model is used to classify the test samples. The results show that the method can identify the arc fault better, and the detection accuracy reaches $\mathbf{9 7. 0 \%}$, which is higher than the accuracy of the classical algorithm, verifying the effectiveness of the method.