Construction of Health Index for Aircraft Air-conditioning System Health Prediction

Constructing a reasonable health index(HI) to evaluate the health status of industrial equipment is the key to equipment fault prediction and health management. Nowadays, data-driven HI construction methods have been widely used. However, these methods usually require a large number of sample data and data for the whole life cycle. To solve this problem, in this paper, we propose a method to create a mapping between working condition parameters and fault parameters and then transform working condition parameters to the same specified working condition. We regard the transformed fault parameters as HI, then health status of future flights can be predicted by recurrent neural network(RNN) in terms of the constructed HI. The experimental results show that this method can effectively indicate and predict the health status of the aircraft air-conditionina system (ACS).