Quantum Machine Learning (QML) has emerged as a promising field at the intersection of quantum computing and Machine Learning (ML), offering new possibilities for enhanced data processing and classification tasks. In particular, quantum kernel methods have demonstrated potential advantages over their classical counterparts in high-dimensional feature mapping. In the transition toward more sustainable energy, Power Quality Disturbance (PQD) classification is crucial for ensuring grid stability amid growing renewable energy integration. Rapid and accurate detection of disturbances enables timely corrective actions to maintain reliable power supply. Support Vector Machines (SVMs), a class of supervised ML models based on kernel methods, have been widely used for PQD classification due to their strong generalization capabilities. In this paper, we integrate a quantum approach into the SVM framework and propose a novel quantum-classical dual kernel SVM that outperforms both purely classical and purely quantum kernel SVMs on an S-transform PQD dataset. By incorporating a weighting strategy between classical and quantum kernels, we develop a robust and highly accurate PQD classification model, achieving an average accuracy of 98.46% across all noise levels. To the best of our knowledge, this is the first application of a quantum-classical dual kernel SVM in power system applications, thereby demonstrating QML’s potential to enhance real-world classification.