A controller using fuzzy probabilistic wavelet neural network (FPWNN) for grid-connected photovoltaic (PV) system during grid faults is proposed in this study. The FPWNN controller is developed to control the output reactive and active power of the grid-connected inverter which also considers the ratio of the injected reactive current to meet the low voltage ride through (LVRT) regulations. Moreover, the balance of the active power between the PV panel and the grid-connected inverter during grid faults is controlled by the dc-link bus voltage. Furthermore, to reduce the risk of over-current during LVRT operation, a current limit is predefined for the injection of reactive current. The main contribution of this study is the design of the FPWNN controller for reactive and active power control that provides LVRT operation with power balance under various grid fault conditions. Finally, some experimental tests are realized to validate the effectiveness of the proposed control scheme.