Solar energy has become one of the most important renewable energies in the world. With the increasing installation of power plants in the world, the supervision and diagnosis of photovoltaic systems have become an important challenge with the increased occurrence of various internal and external faults. Indeed, this work proposes a new solar power plant diagnosis based on the artificial neural network approach. Therefore, the artificial network model was built based on measurements collected on the studied system, with the aim of increasing the electricity production of the studied power plant located in Tamanrasset in the south of Algeria, with different weather conditions in terms of radiation and ambient temperature. The neural network approach is becoming one of the most used approaches to diagnose, detect, localize, and isolate faults in photovoltaic systems. The proposed approach starts by building a reference model with real measurement data for three days of normal operation and then comparing the new days of operation to generate the residual characteristics and to process and evaluate the performance of the new days. The proposed intelligent approach has the ability to prevent the nonlinear behavior of abrupt time changes. Neural networks have shown interesting results with high accuracy.