In order to track nominal output power of photovoltaic (PV) battery effectively and consider the feature of non-linear output, researchers present a Radial Basis Function and Back Propagation (RBF-BP) combination neural network based on genetic algorithm (GA) optimization to use in PV maximum-power-point-tracking (MPPT).First, combination of double hidden layer RBFBP neural network is presented by researching the output feature of PV battery. In order to predict the maximumpower-point of PV battery more accurately, GA is used to optimize combination neural network. Illumination and temperature which is the main factors influencing the output of PV battery are treated as input to construct the prediction model, and simulate the model through MATLAB. Simulation shows that the system has advantages which increase the accuracy and efficiency of tracking the output maximum-power-point-tracking of PV battery effectively of high tracking accuracy, high speed rate and little iteration. KeywordPhotovoltaic (PV) battery; Maximum-powerpoint-tracking (MPPT); genetic algorithm (GA); Radial Basis Function and Back Propagation (RBF-BP); MATLAB simulation