Generally, an electro-mechanical actuator applied to a missile occupies little space but requires high power. This results in nonlinearity such as saturation factor, parameter variation owing to the operating environment, and disturbance due to the aerodynamic force in flight. To prevent this, rather than employ classical control theory, a robust controller that has excellent robustness is applied as position controller for an electro-mechanical actuator.
To design a robust controller, the velocity and position of the actuator are required. In general, a potentiometer is used as position sensor for the position information of an electro-mechanical actuator and a tachometer is used as velocity sensor for its velocity information. However, the use of these sensors has drawback such as increased cost, requirement of additional space for mounting, and decreased system reliability because of sensor failure.
Several methods have been studied to solve them. In previous research, the use of a position sensor for the electro-mechanical actuator and elimination of the velocity sensor has been attempted.
Typical methods for acquiring velocity information in the electro-mechanical systems are counting the pulses of latched type Hall-effect sensor and using mathematical observer. The first technique is useful in fast velocity range. However, this method becomes problematic in slow velocity range because the frequency of the pulses is proportional to the velocity and the velocity information can hardly be obtained in the low velocity range. The second one is an observer method. However, the velocity estimation method through such an observer is sensitive to the disturbance and parameter variation. There is a limitation that can not satisfy with the required performance and stability of the system.
In order to overcome the limitations of the existing studies as described above, the purpose of this study was set as follows.
First, it was attempted to implement a velocity signal to replace the velocity sensor of the electro-mechanical actuator by replacing the latch type hall-effect sensor which is essentially used for driving the BLDC motor of the electric drive with a linear type hall-effect sensor.
Second, it was attempted to eliminate both the existing position sensor and the velocity sensor by implementing the position signal as well as the velocity signal of the electro-mechanical actuator using the linear type hall-effect sensor. To this end, to prevent the magnetic flux interference caused by the stator current, a linear Hall sensor BLDC motor with a magnetic flux shield was designed/fabricated, and the nonlinear reduction ratio problem between the motor rotation shaft and the fin rotation shaft was derived through kinematic analysis.
In order to verify the excellence and validity of the proposed method, a comparative experiment with the case of applying the existing velocity/position sensor was performed, and experimentally verified that the proposed method can replace the existing velocity/position sensor.
In conclusion, by removing the existing velocity/position sensor through the method proposed in this paper, it is considered that it will contribute to miniaturization, cost reduction, and reliability improvement of the electro-mechanical actuator system.