Data-Driven Modeling of High-Speed Centrifugal Compressors for Aircraft Environmental Control System

Abstract

The Environmental Control System (ECS) is the main consumer of non-propulsive power onboard aircraft, accounting for up to 3-5% of the total fuel consumption. The use of an electrically-driven Vapor Compression Cycle (VCC) system, in place of the conventional Air Cycle Machine (ACM), can lead to both a substantial increase of the Coefficient Of Performance (COP) at cruise conditions, and to a reduction of maintenance costs. The performance of the VCC system is highly affected by the design of its main components, namely, the compact heat exchangers and the high-speed centrifugal compressor. Therefore, the optimal system design requires the use of an integrated design methodology. This work documents the development of a data-driven compressor model based on Artificial Neural Networks (ANNs). The objective is to reduce the VCC model complexity, and the computational cost of the associated optimization problem. The model has been trained on a synthetic dataset composed of 165k unique centrifugal compressor designs generated with an in-house tool, validated with experimental data. The data-driven model has been coupled to an in-house integrated design framework for aircraft ECS, and it has been used to perform the multi-objective optimization of a VCC system for a single-aisle, short-haul aircraft, flying at cruise conditions. The results show that the number of function evaluations used to identify the Pareto front reduces by a factor of three, when leveraging the capabilities of the data-driven model. Moreover, the optimal solutions identified with the novel method cover a wider design space, due to the improved robustness of the VCC system model.