This paper presents a hybrid method for detecting power quality events based on combined Hilbert transform and Fourier transform. The decision tree classifier has been implemented over the detection scheme to classify eight power quality disturbances. The proposed method extracts the envelope from the Hilbert transform (HT) and identifies the number of harmonic components present in the power quality (PQ) signal using the Fast Fourier transform (FFT) technique. To evaluate the proposed method, a standard database was created from the mathematical models and white Gaussian noise is added to the database with different noise levels. The relevant features, such as energy, kurtosis, minimum amplitude from the envelope, and the number of peaks from the Fourier magnitude spectrum are extracted. The features have been used to train the decision tree (DT) classifier and a classification accuracy of 99.30% is obtained. The robustness of the proposed algorithm is addressed on another synthetic database with different noise levels, and it achieves 99.27% overall classification accuracy. Finally, the proposed method is compared with recent techniques and achieves promising results in terms of classification accuracy. © 2021 IEEE.