Performance Optimization of Aero Turboshaft Engine Based on Bayesian Network

The aero turboshaft engine is mainly used in helicopters. As a power unit that drives the rotor to generate lift and propulsion, it has been rapidly developed in recent years. When the power of the turboshaft engine meets the conditions of use, the key section temperature often exceeds the threshold. As another important indicator of engine performance, it will affect the safety of the whole machine. This situation has become the primary problem for the current turboshaft engine manufacturers. In this paper, based on the collected data of a certain type of turboshaft engines, according to the manufacturer's suggestions, three component size variables are extracted firstly. They have been confirmed to affect the engine power and the key section temperature. Then, based on Bayesian network, the engine performance models are established for power and the key section temperature respectively. Finally, after validity verification, the production optimization table and transition optimization matrix are proposed. From them, some effective suggestions are also given for the optimization of engine performance.