A sparrow search algorithm optimized support vector machine (SSA-SVM) based aviation arc fault detection method was proposed to address the issues of high concealment and difficulty in detecting arc fault in aviation line systems. Firstly, wavelet decomposition was used to decompose the arc fault current data, which can effectively overcome the problem of modal aliasing during empirical mode decomposition. From the perspective of signal disorder, energy entropy, fuzzy entropy, and approximate entropy were extracted from the current component, and feature vectors were constructed. Then, the sparrow search algorithm was used to optimize the weights of the support vector machine to obtain the optimal weights. Finally, the trained support vector machine was used to classify the test samples. In order to verify the effectiveness of the proposed method, an arc experimental platform was established to simulate the generation of arc faults in aviation line systems. AC series normal and arc fault current data were collected, and the SSA-SVM algorithm proposed was applied for arc fault detection. The results show that the proposed method can effectively identify arc fault, with a detection accuracy of 99.5%. Compared to particle swarm optimization or genetic algorithm optimized support vector machines, the detection accuracy of arc fault is 2.5% and 2% higher, respectively.