World is adapting sustainable energy solutions to the mainstream power grid and solar photovoltaic (PV) power generation is becoming one of the most prominent sources of energy. The generation of solar PV energy is entirely dependent on climate and weather conditions. Often, a solar PV plant is exposed to extreme environmental conditions. These conditions can result in a variety of faults that can disrupt their energy yield. This paper proposes an integrated approach for detecting electrical faults in solar PV plants in real time onload condition, classifying those faults, and precisely locating the faulty PV panel. The proposed method is tested using field data for faults like open circuit, short circuit as well as partial shading and degradation. The proposed approach is integrated and uses the trained ANN model based on field data. This makes the proposed method accurate and precise as well as appropriate for IoT-based applications.