A novel method for classifying the combined power quality disturbance using Hilbert–Huang transformation and machine learning is proposed. At first the simple and complex power quality disturbances are generated using MATLAB. The Hilbert–Huang transformation is applied to the time-domain signals for feature extraction. The extracted features are fed to machine learning classifiers for classifying the disturbances. The proposed method demonstrates superior performance in distinguishing power quality disturbances. The outputs obtained from five different machine learning classifiers decision tree, random forest, Naive Bayes, K-nearest neighbor, and support vector machine are compared for their performance using various parameters. The results obtained from simulation validate that the support vector machine classifier has performed well surpassing other existing algorithms.