Convolutional neural network approach for fault detection and characterization in medium voltage distribution networks

• This research presents a novel approach to identifying and characterizing faults in medium-voltage networks by utilizing Park's transformation to distinguish between normal and fault conditions through voltage waveform analysis. • Unlike most fault diagnosis studies that use current samples of Park's transformation, this paper uses voltage samples, which are simpler to implement and maintain in distribution networks. • The paper introduces a method that uses the dq0 transform (Park transform) for real-time fault monitoring, eliminating the need to store previous waveform samples, and thereby enhancing detection efficiency and speed. • By converting voltage waveforms into high-resolution images and using a sine-fitting algorithm to validate image classification results, this study sets itself apart from existing methods of fault detection. • This paper introduces a novel method for detecting, classifying, and locating faults in power systems through voltage waveform analysis using a convolutional neural network (CNN) integrated with the piecewise function put together (PFPT) algorithm for fault zone localization in a power distribution network. • Utilizing artificial intelligence and convolutional neural networks (CNNs) for fault prediction, the proposed technique achieves a remarkable 93.1% accuracy in identifying fault types, thereby enhancing the reliability of power system diagnostics. • In summary, this research presents a novel approach to identifying and characterizing faults in medium-voltage networks. The integration of CNN and PFPT allows for precise fault type classification and accurate fault zoning, enhancing the reliability and response time of power distribution systems. Future work will address the prediction of the exact fault location at the bus level using the PFPT algorithm. Power outages significantly impact the power industry by disrupting social welfare and economic stability. Still, existing methods for fault detection face challenges due to load and network topology, conditions, and installed equipment. However, recent advances in artificial intelligence (AI) are enabling researchers to create alternative approaches for fault detection and location strategies. Therefore, this paper introduces a novel method for detecting, classifying, and locating faults in power systems through voltage waveform analysis using a convolutional neural network (CNN) integrated with the Piecewise Function Put Together (PFPT) algorithm for fault detection and fault zone localization in a power distribution network. Utilizing Park's transformation, noise reduction PFPT sine fitting, and CNNs, the proposed method distinguishes between 'healthy' and 'faulty' conditions. Simulation results reveal that while the voltage Park's vector time behavior of a healthy system remains stable, it exhibits circular or mixed patterns under faulty conditions. These patterns enable the identification of four types of short circuit faults—single-line-to-ground (LG), line-to-line (LL), line-to-line-to-ground (LLG), and three-line (3L) faults—by analyzing 3D voltage Park's waveforms at network buses. The study validates fault type identification through the observation of rotating Park vectors from sine fitting of time-based voltage waveforms. By converting 3D voltage waveforms into high-resolution images, the method utilizes a CNN for fault recognition, achieving an accuracy of 93.1%. This innovative approach underscores the robustness and precision of combining traditional electrical engineering techniques with modern AI.