As a new achievement in the field of pattern recognition and machine learningdeep learning has broad prospects in the field of equipment fault diagnosis and health management.In this papera new method of fault diagnosis is proposed based on the characteristics of equipment fault big data and the advantages of deep learning theory.According to the principle of the denoising auto-encoderthe unsupervised feature learning of the training network is achievedand the structuring of the whole neural network is completed.According to the type of faultthe output layer is determined.Using the BP algorithmthe supervised fine-tuning of the whole network is carried outand thus the accuracy of fault classification is enhanced.By means of the above methodsthe module-level fault diagnosis of a communications station is completed through experiments.