An overview on the deep learning based prognostic

The systems of prognostic and health management (PHM) aim to locate all kinds of anticipated faults by determining the remaining useful life of a certain system. Existing traditional methods try to find a solution to the problem of detecting faults and predict the future behavior of the machine. However, these methods suffer from some limits such as features extraction manually. To bypass those limits, Deep Learning has the ability to analyse and process large data using helpful techniques. This paper presents the utility of Deep Learning in prognostics. In addition, the present paper displays the difference between the mention methods with the existing ones through a comparison.