Joint steady state and transient performance adaptation for aero engine mathematical model

Abstract

In the field of aero-engine control, it is valuable to build highly accurate component level models to meet the requirements of controller validation and model-based control. The accuracy of current performance map generation method is limited by the test data that always cannot cover the full envelope. The scaling-based performance map correction methods only focus on the adjustment of the steady-state performance. Therefore, a performance map segmentation-based joint steady-state and transient performance adaptation technique is proposed. Both the idle point and the design point are taken as precision references to scale the performance maps with large deviations. A scaling factor domain determination method is provided based on the characters of the speed lines. The performance map is first optimized with the steady state calculation based on the steady-state test data, and further optimized with the transient calculation based on the transient test data. Both the steady-state and transient performance adaptations are achieved by adjusting the performance maps. The proposed method was applied to a turbofan engine model. For continuous acceleration case, the average error is less than 1% at steady state points and less than 2% during transient operation. For large magnitude acceleration and rapid deceleration cases, the maximum errors for parameters around the idle point decrease by about one time (from 8.3% to 4.1%) and three and a half times (from 28% to 8%), respectively, by comparing with the model without transient performance adaptation. Thus, the effectiveness of the proposed method is demonstrated.