This research presents a new methodology for the development of a controller based on Artificial Neural Networks and Direct control method in order to obtain the maximum available energy from Solar Photovoltaic (PV) Energy systems under different atmospheric changes of the solar insolation and ambient temperature. In this context, this research presents a new model for MPPT-ANN in order to track the Maximum Power Point of PV systems in Matlab/Simulink environment. The developed controller is based on Feed Forward Neural Network FFNN trained by Back-propagation algorithm of error to determine the optimal voltage operation of the system PV system at different atmospheric changes. This research also suggests, control algorithm based on the direct control method in order to determine the duty cycle, which used to control directly the operating of DCDC Voltage Converter, depending on a comparison of the difference between the output voltage of PV system and the optimal voltage output of the neural network. The developed controller MPPT-ANN based on a network FFNN, Characterized by fast speed to track of MPP point and achieve high efficiency for the PV system under the atmospheric changes. The simulation results completed in Matlab/Simulink environment, showed the best performance of developed controller MPPT-ANN by achieving a better dynamic performance and high accuracy when tracking the MPP, compared with the use of the another PI-ANN controller based on artificial neural network and the conventional Proportional-Integral Controller, and compared with the use of the conventional MPPTP& O based on Perturb and Observe (P&O) technique under different atmospheric changes.