The 2021 robust-design study of adaptive cycle engine (ACE) performance under component performance uncertainty treated all twelve component map-scaling factors as having one uniform uncertainty level, considered only steady performance at three critical operating conditions, and never treated mode transition. Substantial adjacent progress exists — deterministic mode-transition control-law design, probabilistic component-margin allocation for a conventional aero-engine, interval-based deviation standards, and deterministic mission-level optimization — but no work allocates component performance tolerances inside an ACE robust-design optimization, treats mode-transition success and transition-point/control-law selection under component performance uncertainty, or extends robust ACE performance optimization to a complete flight mission. The problem remains open at its source generality, unified by the objective of a robust-design methodology for ACE performance over the engine's real operating envelope with component tolerances as design variables.

Background

An adaptive cycle engine (ACE) is a three-stream turbofan variant that can reconfigure itself in flight: through variable-geometry elements such as a mode select valve, variable-area bypass injectors, and a fan-on-blade (FLADE) stage on the fan hub, it shifts airflow between bypass ducts so that one engine behaves closer to a high-thrust turbojet for supersonic flight or a low-fuel-consumption turbofan for subsonic cruise. Its performance is computed by cycle (thermodynamic) models built from component maps — experimentally based tables giving each component's efficiency and flow capacity as functions of speed and pressure ratio. Component performance uncertainty means that the real engine's components deviate from their design-point maps because of manufacturing scatter and modeling error; a standard way to represent this is to scale the maps by uncertain factors (here twelve map-scaling factors, one per component performance parameter). Because such deviations shift thrust and specific fuel consumption (SFC, fuel flow per unit thrust), an engine designed only for nominal maps may miss its requirements with large probability. Robust design counters this by optimizing a performance statistic that accounts for the uncertainty — for example the probability of meeting a requirement (PMR), estimated by propagating the uncertain scaling factors through the cycle model, often by Monte Carlo sampling, and accelerated by surrogate models (cheap fitted substitutes for the expensive physics-based model, here a linear model in the scaling factors nested inside a quadratic response surface in the design variables). Zhang, Tang, and Chen (2021) established the first such robust-design framework for an ACE: they trained two-level surrogates, compared optimization targets (performance at a confidence probability versus Taguchi signal-to-noise ratio, a robustness statistic built from the mean and variance), and, in a two-stage process, first optimized six cycle parameters (turbine inlet temperature, overall and core-driven fan stage pressure ratios, and three bypass ratios) and then the variable-geometry control law, raising the PMRs of takeoff thrust, cruise thrust, and cruise SFC from roughly 46–50% to 93–99.9% under the assumed uncertainty. The study, however, assumed all twelve map-scaling factors to be independent and identically normally distributed with the same small standard deviation, evaluated only steady performance at three operating points (takeoff, subsonic cruise, supersonic cruise), and did not treat mode transition — the dynamic process of switching the engine between its airflow configurations, during which component matching and compressor surge margin (the distance to aerodynamic instability) constrain the admissible control schedules. The authors themselves identified three resulting research gaps, and adjacent literature has since addressed neighboring questions without closing them: interval analysis had earlier set per-mode allowable component-deviation standards on a fixed design (Chen, Zhang, and Tang, 2018), deterministic control-law design methods for ACE mode transition now exist (Zhang et al., 2022; Zhao et al., 2024; Fu et al., 2026), uncertainty propagation through a mode transition has been studied for a turbine-based combined-cycle propulsion system with inlet-side uncertainties (Lv et al., 2023), probabilistic component-margin allocation has been demonstrated for a conventional aero-engine (Yang et al., 2025), and mission-level operational optimization of an adaptive-cycle microturbofan is available in deterministic form (Palman et al., 2018) — but none of these couples component performance uncertainty with tolerance allocation, transition robustness, and full-mission design for an ACE.

Problem Statement

For an adaptive cycle engine whose component performance is uncertain — represented as deviations of component map-scaling factors (efficiencies and flow capacities) from their design values — extend the robust-design framework of Zhang, Tang, and Chen (2021) from its original scope (steady performance at isolated operating points under a single uniform uncertainty level assigned to all components) to the engine's real operating envelope, by resolving three linked questions the source explicitly leaves open: (1) Component tolerance allocation: replace the uniform uncertainty level with rationally allocated, component-specific allowable performance deviations, allocated according to how strongly each component actually affects engine-level performance so that tolerances can be set more economically, integrated into the robust-design optimization rather than fixed beforehand. (2) Uncertainty-aware mode transition: quantify how component performance uncertainty affects the mode transition process — whether the transition still succeeds (e.g., thrust and airflow continuity and adequate compressor surge margins) under uncertain component performance, how the transition point should be selected in its presence, and how the variable-geometry control law during transition should be designed to be robust to it. (3) Full-mission robust optimization: extend the robust optimization from a few representative steady operating points to performance over a complete flight mission, including the operating and transition segments between them. The common objective is a robust-design methodology for adaptive cycle engine performance under component performance uncertainty that treats component tolerances as design variables and covers the full mission including mode transitions. The three questions are linked and jointly posed — the allocation rule together with its integration into the optimization; transition success together with transition-point selection and control-law design; and the full-mission extension — and all three bear on that common objective. This statement fixes no ACE configuration; allocated tolerances, uncertainty-aware transition treatment, and the full-mission extension must each change the resulting design and its probability of meeting requirements relative to the uniform-tolerance, steady-point baseline.

References

  1. Zhang, Jiyuan; Tang, Hailong; Chen, Min (2021). Robust design of an adaptive cycle engine performance under component performance uncertainty. Aerospace Science and Technology 113:106704. DOI: 10.1016/j.ast.2021.106704. https://doi.org/10.1016/j.ast.2021.106704
  2. Chen, Min; Zhang, Jiyuan; Tang, Hailong (2018). Interval analysis of the standard of adaptive cycle engine component performance deviation. Aerospace Science and Technology 81:179–191. DOI: 10.1016/j.ast.2018.07.004. https://doi.org/10.1016/j.ast.2018.07.004
  3. Zhang, Jiyuan; Tang, Hailong; Chen, Min (2019). Robust design methodologies to the adaptive cycle engine system performance: preliminary analysis. Energy Procedia 158:1521–1529. DOI: 10.1016/j.egypro.2019.01.361. https://doi.org/10.1016/j.egypro.2019.01.361
  4. Zhang, Jiyuan; Dong, Pengcheng; Tang, Hailong; Zheng, Junchao; Wang, Jingkai; Chen, Min (2022). General Design Method of Control Law for Adaptive Cycle Engine Mode Transition. AIAA Journal 60(1):330–344. DOI: 10.2514/1.j060674. https://doi.org/10.2514/1.j060674
  5. Zhao, Xingyu; Li, Qiuhong; Pang, Shuwei; Xue, Zhaohui; Deng, Daming; Lu, Feng (2024). Research on Adaptive Cycle Engine Mode Transition Control Method. Energies 17(6):1276. DOI: 10.3390/en17061276. https://doi.org/10.3390/en17061276
  6. Fu, Yu; Song, Wenyan; Wang, Qiuyin (2026). Co-Optimized Flow Matching and Thrust Retention Control for an Adaptive Cycle Engine in Turbine-Based Combined Cycle Mode Transition. Energies 19(4):993. DOI: 10.3390/en19040993. https://doi.org/10.3390/en19040993
  7. Lv, Chengkun; Huang, Qian; Chang, Juntao; Wang, Ziao; Zheng, Jialin; Yu, Daren (2023). Mode transition path optimization for turbine-based combined-cycle ramjet stage under uncertainty propagation of integrated airframe-propulsion system. Energy 268:126718. DOI: 10.1016/j.energy.2023.126718. https://doi.org/10.1016/j.energy.2023.126718
  8. Yang, Kun; Sun, Yanghui; Tan, Zhiyong; Hou, Naixian; Pan, Xu (2025). A probabilistic approach for aero-engine component performance margins design. Proceedings of the Institution of Mechanical Engineers, Part G: Journal of Aerospace Engineering 239(16):2496–2510. DOI: 10.1177/09544100251362520. https://doi.org/10.1177/09544100251362520
  9. Palman, Michael; Leizeronok, Boris; Cukurel, Beni (2018). Mission Analysis and Operational Optimization of Adaptive Cycle Microturbofan Engine in Surveillance and Firefighting Scenarios. Journal of Engineering for Gas Turbines and Power 141(1):011010. DOI: 10.1115/1.4040734. https://doi.org/10.1115/1.4040734
  10. Chen, Min; Deng, Li; Zhang, Jiyuan; Tang, Hailong (2025). Review on the development status and performance deign [sic] of adaptive cycle engines. Propulsion and Power Research 14(1):35–63. DOI: 10.1016/j.jppr.2025.02.006. https://doi.org/10.1016/j.jppr.2025.02.006
  11. Zhang, Jiyuan; Tang, Hailong; Zheng, Junchao; Liu, Xin; Chen, Min (2025). Adaptive Cycle Engine's Performance Design Under Component Performance Non-Probabilistic Uncertainty. SSRN preprint. DOI: 10.2139/ssrn.5173342. https://doi.org/10.2139/ssrn.5173342