Quantum Support Vector Machines for Aerodynamic Classification

Aerodynamics plays an important role in the aviation industry and aircraft design. Detecting and minimizing the phenomenon of flow separation from scattered pressure data on the airfoil is critical for ensuring stable and efficient aviation. However, given that it is challenging to understand the mechanics of flow-field separation, aerodynamic parameters are emphasized for the identification and control of flow separation. The mechanics of flow-field separation have been extensively investigated using traditional algorithms and machine learning methods such as support vector machine (SVM) models. Recently, growing interest in quantum computing and its applications in various research communities has shed light on the use of quantum techniques to solve aerodynamic problems. In this study, we applied qSVM, a quantum SVM algorithm based on the quantum annealing model, to identify flow separation and compared its performance to that of the widely used classical SVM. We demonstrated that our approach outperforms the classical SVM with an 11.1% increase in accuracy, from 0.818 to 0.909, for this binary classification task. We further developed a multiclass qSVM based on a one-against-all algorithm and applied it to the classification of multiple angles of attack on the wings, where its advantage over its classical multiclass counterparts was maintained with a 17.9% increase in accuracy, from 0.67 to 0.79. Our study demonstrates a useful quantum technique for classifying flow separation scenarios and may promote the investigation of quantum computing applications in fluid dynamics.