A BP neural network model was established and trained to be applied in the evaluation of water quality in water source areas.The results indicate that the evaluation of water quality in water source areas for rural drinking water based on the BP neural network can avoid the disadvantages of other evaluation methods and improve the global optimization ability of the evaluation,and it can be adopted to rationally and fully evaluate the water quality in water sources areas.Meanwhile the model reflects the nonlinear characteristics of water quality in water source areas and improves the reliability and effectiveness of evaluation of water quality in water source areas.