Local model based turbo-shaft engine gas turbine speed signal reconstruction method

By establishing a local model and estimating health parameters, the problem of high failure rate of gas turbine speed sensor in turboshaft engine was solved, and high-precision and real-time signal reconstruction was achieved, which is applicable to steady-state and transient processes of turboshaft engine.

CN117521506BActive Publication Date: 2026-07-21NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
Filing Date
2023-11-10
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

The existing turboshaft engine gas turbine speed sensor has a high failure rate, which leads to instability in the control system and affects the accuracy of guide vane position control. Furthermore, the existing signal reconstruction method cannot meet the requirements of real-time performance and engineering applications.

Method used

The method for reconstructing the gas turbine speed signal of a turboshaft engine based on a local model combines component-level modeling and effective sensor signals. By establishing a local model and designing training modes to estimate health parameters, it achieves high-precision reconstruction of the gas turbine speed signal.

Benefits of technology

It improves the reconstruction accuracy and real-time performance of gas turbine speed signals, providing high-precision signal reconstruction results in both steady-state and transient states. It is highly practical and does not require additional sensors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117521506B_ABST
    Figure CN117521506B_ABST
Patent Text Reader

Abstract

The application discloses a local model-based turboshaft engine gas turbine speed signal reconstruction method. First, through turboshaft engine flow path analysis, relevant components and sensor information for calculating the gas turbine speed ng are selected, and an aerodynamic thermodynamic method is used to establish a ng local model. When the ng signal does not fail, the model enters a training mode, decoupling of health parameters is realized through theoretical analysis, appropriate health parameters are selected, based on the error between the model calculation result and the measured data, the health parameters of components are estimated based on the method of solving nonlinear equations and K-means clustering, and the health parameters are introduced into the ng local model as adjustable parameters, so that online correction of the local model is completed. The results show that the steady-state error of the ng reconstructed signal is not more than 0.25%, and the transition state error is not more than 0.33%, indicating that the signal reconstruction method has high precision. Moreover, the method does not need to increase additional sensors, has good real-time performance and strong engineering practicability.
Need to check novelty before this filing date? Find Prior Art