Multisystem Error Separation-Based Coaxiality Measurement for High Depth-to-Diameter Ratio Casing Cavities in Aeroengines

The coaxiality of the installation and machining processes of aero-engine components, including the combustion chamber casing, fuel gear pump, and intermediate casing, is a critical factor influencing service life and operational stability. During the machining and measurement of workpieces with large depth-to-diameter ratios, measuring equipment inevitably introduces additional errors. During measurement, systematic errors - including workpiece mounting eccentricity, sensor radial offset, sensor pitch angle deviation, sensor column inclination, and turntable axis tilt - are mutually coupled, resulting in a significant reduction in measurement accuracy. The proposed method is specifically developed for inner cavity surface measurements with large depth-to-diameter ratios, where the depth exceeds 400 mm, and the diameter is less than 100 mm. In the coaxiality measurement of inner cavities with large depth-to-diameter ratios, accessibility limitations and occlusion effects make it difficult for 3-D optical metrology systems to achieve reliable evaluation. Moreover, traditional measurement systems typically use the outer surface of stepped shafts as the measurement reference, making it difficult for probes to access deep inner cavities with large depth-to-diameter ratios; consequently, the corresponding measurement models are no longer applicable. In this study, a five-system-error measurement model is proposed based on a noncontact automated measurement device. The effects of the five systematic errors on measurement results are numerically simulated and analyzed. An optimization method based on the slime mold algorithm (SMA) is employed to estimate and separate systematic errors from measurement data obtained from shell components of varying sizes with large depth-to-diameter ratios. Simulation results indicate that the proposed model is more effective for workpieces with smaller dimensions and larger systematic error magnitudes. Finally, the validity of the proposed five-system-error model (5SEM) is experimentally verified. Measurements are conducted using a standard artifact with a nominal coaxiality of 11.64μ m. After error separation using the two-system-error model (2SEM), a coaxiality value of 22.9μ m is obtained, whereas a reduced value of 17.02μ m is achieved using the 5SEM. Overall, the proposed model improves measurement accuracy by 43.2%, corresponding to an improvement of 4.06μ m.