Failure Modes and Modeling of Flight Actuators

As long as flight actuators operate normally, the desired deflections can he driven exactly as the FCS has mandated. However, when actuators fail to operate, the control commands cannot be completed as required, and catastrophic consequences may be induced. Based on the fault analysis of a hydraulically driven actuator in Ref. [30], the ith actuator fault model can be presented as:

where the factor /, is declared to indicate the effectiveness of the ith actuator. More particularly, (1) /, = 1 describes that the ith actuator is working under a normal condition; (2) 0 < l, < 1 denotes a partial loss of effectiveness in the ith actuator; and (3) U = 0 represents that a complete failure occurs in the ith actuator. A diagonal matrix L describes the effectiveness in any of the actuators in Eq. (7.24) as:

As a result, the aircraft longitudinal model with actuator faults and limited control authority can be represented as:

Remark 7.2. It is worth mentioning that actuator stuck and runaway are typical failures in hydraulic-type actuators as well. When an actuator is lost due to stuck failure, the faulty actuator is “frozen” at a specific position, which can no longer respond to the applied command. Runaway failure is characterized as when the surface locks at its limited deflection amplitude. It is seen as the worst case of jammed failures. In comparison to the investigated fault in Eq. (7.26), the jammed and runaway failures can impose an external disturbance on the faulty aircraft, deteriorating the safety. Nonetheless, the strategy against jammed and runaway failures is beyond the scope of this chapter.

Failure Modes and Modeling of Flight Sensor Gyros

According to either the law of conservation of momentum or Sagnac effect, there are two sorts of gyros: mechanical and optical gyros. Conventional spinning gyro and vibrating gyro fall into the mechanical type, while ring laser gyro (RLG), and fibre optic gyro are recognized as the optical type. Conventional spinning gyros are still used for the Boeing 747 transport aircraft [182], albeit with the growing configuration of RLGs in civil aircraft.

As can be observed from [183], the rate gyro consists of (1) the spin motor and rotor; (2) the gimbal; (3) the torsion spring; and (4) the pickoff (output signal generator). The motor converts electrical energy to mechanical energy, which activates the inertia wheel rotating at a constant speed. The motor and the inertia wheel are contained in the gimbal. The springs can provide the rotational resistance to balance the output torque caused by the gyro. The pickoff determines an AC voltage proportional to the rotating speed resolved by the gyro.

Quoting from [58], gyros have the highest failure rates among the crucial components in an IMU. Gyro malfunctions are pertinent to bearing failure induced by the instrument ingesting dirty air, and/or impact damage to the sensitive gyro rotor and gimbal bearings. Inadequate vacuum or pressure system air filtration can accelerate bearing wear. In faulty cases, the sensed angular rates are no longer reliable for FCS and FHMS.

The aircraft output vector can be expressed as:

where h (x) : 7?.^:i —» TZ² is a smoot h vector field of x.

When sensors suffer faults, Eq. (7.28) can be further described as:

where Г = diag {71,72} and 7 = [tq, v^'. The typical sensor faults are represented by Eq. (7.29) [57]. (1) When 7j = 1 and Vj = a constant offset value, a bias fault occurs in the j^th sensor; (2) when qq = 1 and Vj = a time-varying offset factor, a drift fault takes place in the j^th sensor; (3) when 7j = 0 and vj = 0, the j^th sensor suffers from signal loss; and (4) when 7j = 0 and vj = a constant value, the j^th sensor encounters a stuck failure.

Problem Statement

As the main contributions of this work, two fault accommodation strategies are developed in response to different fault cases. In regard to the first architecture, the objective is stated as follows:

• 1) Stabilize the aircraft and retain the tracking performance in the event of actuator failures;
• 2) The amplitude and rate limits of healthy actuators are not violated when overcoming the deleterious effects of actuator failures.

The second algorithm is designed such that:

• 1) The aircraft can be stabilized and the tracking performance can be preserved in the case of both actuator failures and gyro failures;
• 2) The amplitude and rate limits of healthy actuators are respected during the fault accommodation procedure.