Hybrid Model Predictive Control for Aircraft Engine Stall Margin Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Gas turbine engines face compressor stall issues due to unstable flow during acceleration and high altitude conditions, leading to increased turbine temperature and mechanical vibration, and reduced cooling air, which can result in turbine failure if not addressed in real time.
Innovation Solution
A Hybrid Model Predictive Control (HMPC) system is implemented, using power goals, operability limits, sensor signals, and a non-linear engine model to determine multi-variable control commands for actuator control, including stall margin estimation and dynamic optimization, to manage compressor stall margin in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If real-time control is implemented to prevent compressor stall, then reliability is improved, but device complexity increases due to computational requirements
Solution Approach 1:
The patent transforms the complex multi-variable control problem into a simpler form by changing parameters: it uses a reduced-order dynamic model that captures essential stall behavior, transforms the optimal control problem into a parameter optimization problem, and uses real-time parameter estimation to simplify the control law implementation while maintaining effectiveness in preventing compressor stall
Solution Approach 2:
The control system is segmented into distinct functional modules: a dynamic model of compressor stall, a parameter estimation module, an optimization module, and a control law module. This segmentation allows each module to handle specific computational tasks independently, reducing overall system complexity while maintaining real-time control capability
2Productivity
If computational efficiency is improved for real-time control, then productivity is improved, but measurement precision may deteriorate in stall margin estimation
Solution Approach 1:
The patent changes the estimation approach from direct complex model simulation to parameter optimization using a reduced-order dynamic model. This parameter transformation enables faster computation while maintaining accuracy by focusing on critical stall-related parameters rather than full system state estimation
Solution Approach 2:
The control system implements continuous feedback through real-time parameter estimation and optimization. The estimated parameters are fed back into the control law, which adjusts control inputs to maintain optimal stall margin. This closed-loop feedback ensures both computational efficiency and estimation accuracy by continuously correcting deviations
Data Source
AI summary
A control system for a gas turbine engine, a method for controlling a gas turbine engine, and a gas turbine engine are disclosed. The control system may include a hybrid model predictive control (HMPC) module, the HMPC module receiving power goals and operability limits and determining a multi-variable control command for the gas turbine engine, the multi-variable control command determined using the power goals, the operability limits, actuator goals, sensor signals, and synthesis signals. The control system may further include system sensors for determining the sensor signals and a non-linear engine model for estimating corrected speed signals and synthesis signals using the sensor signals, the synthesis signals including an estimated stall margin remaining. The control system may further include a goal generation module for determining actuator goals for the HMPC module using the corrected speed signals and an actuator for controlling the gas turbine engine based on the multivariable control command.