Since BP neural network is easy to fall into local optimal solution and the poor dynamic performance of fuzzy control in photovoltaic soft MPPT algorithm, the BP neural network and adaptive fuzzy control photovoltaic MPPT algorithm (BP-FLC) is proposed. The deviation between the reference voltage and the photovoltaic cell voltage and the duty ratio D(n-1) at the previous time are used as the inputs of the fuzzy control. The constriction factor is employed to optimize D(n-1), and the duty cycle D(n) of the Boost circuit is controlled to achieve MPPT. Simulations on Matlab and experimental verification are carried out under atmospheric conditions, the results show that the proposed algorithm has better tracking speed and efficiency than the adaptive perturbation and observation, fuzzy control and particle swarm optimization algorithm.