Multiple Sliding Modes with Override Logic: Limit Management in Aircraft Engine Controls

G AS turbine engines are subjected to several operational constraints, both at steady state and in the transient regime. These constraints correspond to mechanical limits such as speed and acceleration; limits in thermodynamic variables such as pressures and temperatures; and limits preventing engine components from undergoing dynamic instabilities, mainly surge, stall and flame blowout. Engine control systems incorporate measures to protect the engine from crossing these limits at all times. Achieving limit protection while ensuring fast transient response represents a design challenge, which was summarized by Spang and Brown in their 1999 paper [1]: “. . .much of the complexity of the control comes from the need to operate the engine as close as possible to its limits.” This paper presents a novel and very effective way to achieve such maximal exploitation of the available limits to enable fast responses in thrust or other relevant variables. The proposed technique is based on thewidely usedmultiregulator architecture with max–min limit protection logic [1,2], which uses linear compensators in its original form. As shown in Fig. 1, this architecture uses a single control input (fuel flow) to drive the desired output (typically fan speed or engine pressure ratio) to a setpoint through a control loop with integral action and a linear compensator. Additional compensators are set up to control limited variables to their permissible values, still using fuel flow as the control input. A max–min selection logic stage decides whether the main output regulation task or any of the limit protection tasks should be made active at anygiven time. Integral action is used to achieve type 1 loops for offset-free regulation. The regulators, thus, provide control input rates. Define two index sets as L f1; 2; . . . lg and H fl 1; l 2; . . . hg, where L contains the indices of the regulators associated with the min selector, while the elements of H are the indices of the regulators associatedwith themax selector. No index is defined for the output of the min selector. As shown in Fig. 1, the control rate applied to the integrator is selected according to the formula