This paper presents an efficient and cost-effective method for detecting open-circuit faults (OCFs) in single-phase H-bridge inverters. Such inverters are widely used in residential, renewable energy, and industrial applications for DC–AC power conversion. Since open-circuit faults in semiconductor switches account for nearly 38% of inverter-related failures, the proposed technique provides a practical and reliable diagnostic solution. The method employs a single Hall-effect current sensor along with Discrete Wavelet Transform (DWT)-based feature extraction and Random Forest (RF) classification. The measured current signal is processed using a five-level Daubechies-4 (db4) wavelet decomposition to capture time–frequency characteristics. Statistical features extracted from the wavelet coefficients are used for accurate fault classification. Simulation results demonstrate a classification accuracy of 92%, while an Arduino-based hardware prototype achieves an accuracy of 87%. The integration of db4 wavelet analysis, Random Forest classification, and single-sensor measurement ensures high diagnostic accuracy, low computational complexity, and suitability for real-time operation.