The life support system (LSS) within a submarine ensures crew safety and sustenance during underwater missions. The Oxygen Injection System (OIS) within the LSS is crucial for supplying breathable air to the crew through various oxygen valves. Our model consists of two main components: firstly, a cohesive fault diagnostic for the oxygen valves, using a multimeter and pressure gauge to diagnose and monitor the functioning of two-way internally piloted solenoid valves in the OIS. This is implemented through MATLAB Simulink &Python for various time intervals in the operational range. Secondly, the estimation of the Remaining Useful Life (RUL) of the OIS is achieved through a Bayesian Convolutional Neural Network (BCNN) model & Kullback Leibler (KL) Divergence by using the data obtained after fault analysis. This comprehensive approach considers various ambient parameters & human parameters to estimate the RUL in hours & mins. Therefore, the OIS and LSS are integrated which then ensures smooth functioning of the submarine that allows detection of faults in a timely fashion, avoid leakage in O2 valves and provides time to time data on the RUL to the crew members.