A multi-source-information-constrained method for identifying component performance dispersion in mass-produced aero-engines

During the mass production of aero-engines, variations in manufacturing processes, material properties, and assembly precision inevitably cause deviations in component performance, leading to noticeable dispersion in overall delivered engine performance. However, the limited number of measurement stations and the insufficient information content of single-engine test data make it difficult for traditional methods to reliably identify component-level deviations or characterize performance dispersion across production batches. To address this challenge, this study proposes a multi-source-information-constrained identification method for quantifying component performance dispersion in mass-produced aero-engines. The method integrates condition-number-based screening of measurement parameters, mean - value rationality constraints derived from multi-engine statistics, and prior dispersion estimates from a geometry–component correlation model. These constraints collectively mitigate the effects of measurement noise and systematic bias and enhance the robustness of the identification process.A twin-spool mixed-flow turbofan engine is used as the demonstration case. Performance-dispersion identification is performed for 100 production engines, and the results are validated against low-speed air-flow measurements. The final methodology achieves a mean deviation of only 0.34% and improves the accuracy of standard-deviation prediction by approximately 26 times relative to traditional approaches.The proposed framework provides a reliable and scalable solution for dispersion identification in mass-produced aero-engines, offering an important quantitative basis for performance evaluation, production quality assurance, and future health-management applications.