Inverse design of aircraft cabin ventilation by integrating three methods

To create a healthy and comfortable aircraft cabin, air-supply parameters of the cabin ventilation system must be designed appropriately. Several methods, such as the computational fluid dynamics (CFD)-based genetic algorithm, CFD-based adjoint method and CFD-based proper orthogonal decomposition (POD), have been developed in recent years for conducting an inverse design. The target environmental performance is specified first, and then the corresponding air-supply parameters are inversely solved with the use of a particular method. However, each method has its pros and cons in terms of efficiency and accuracy. To expedite the inverse design process, this study proposed to integrate the above three methods. The genetic algorithm was adopted first to circumscribe ranges of the air-supply parameters. Next, POD was applied to further narrow the ranges and estimate the optimal air-supply parameters for each design criterion. Finally, the estimated optimal design from POD was supplied to the adjoint method for fine tuning. The above strategy was applied to a five-row aircraft cabin to determine the air-supply opening sizes, directions and temperatures. Criteria that had been proposed specifically for aircraft cabins were used as design targets. Results show that the proposed integration was able to provide the optimal design for each design target. The integrated optimal design was superior to the design provided by each individual method. The bottleneck in further acceleration of the integrated design was the hundreds of design cases resolved by full CFD simulation.