122
Unauthenticated
Download Date | 3/31/16 1:19 PM
acta mechanica et automatica, vol.8 no.3 (2014), DOI 10.2478/ama-2014-0021
where: I(f) – amplitude of current in the electromagnet coil at the frequency f, (f) – amplitude of the PWM factor at the frequency
It appears that the maximal frequency of variations of the command signal is 188 Hz, which means that the minimal sam-pling frequency of the control algorithm should be equal to 376 Hz. The set-up for ACS testing is shown in Fig. 4.
Fig. 3. Transmissibility characteristics of the power interface for the VI-MR
Fig. 4. Set-up for ACS testing
3.2. Results
Testing was done on the ACS system in the open loop config-uration (ACS-O) and in the closed-loop configuration (ACS-PID). The applied control signal was current i in the electromagnet coil. Electrical parameters of the electromagnet coil were: resistance 1.059 Ω, induction 1.13 mH.
The PI version of the PID controller, used in the ACS-PID sys-tem, is governed by the equation (2).
Tab. 2. Time required for reaching the preset current level in the control coil
where: u(t) – control, u0 – control value in the steady state, ε(t)
– control error, i0 – preset current level in the coil, Kp – controller
gain, Ti – integration time.
The parameters of the PI controller chosen using the paramet-
ric optimisation methods are: K p = 49.3; T = 1.74 × 10−4.
i
The integration function of the PI controller is approximated
by the rectangle method.
Fig. 5. Time patterns of current : a) i0=1 A, b) i0=3 A, c) i0=5 A
Testing of the ACS-O and ACS-PID system was done for the pre-defined PWM signal frequency (5 kHz) and the duty cycle was varied from 0 to 1 with the step 0.1. As it was mentioned previously, the primary function of the ACS-PID system is to stabi-lise the current level in the electromagnet coil. This configuration of the system was tested for the preset current levels i0: 1, 2, 3, 4, 6 A. The patterns of the PWM control and current in the coil
123
Unauthenticated
Download Date | 3/31/16 1:19 PM
Maciej Rosół, Bogdan Sapiński
Autonomous Control System for a Squeeze Mode MR Vibration Isolator in an Automotive Engine Mount
registered for the ACS-O and ACS-PID configurations are shown in Fig. 5, for i0=1 A, i0=3 A and i0=5 A. The current i(t) plots show that the required current level i0 was reached in the elec-tromagnet coil, both in the ACS-O and the ACS-PID configuration. Application of the PI controller allowed a 30% reduction of the time needed to reach the preset current level (see Tab. 2). One has to bear in mind, however, that for smaller values of i0 these time intervals can become decidedly shorter.
4. SUMMARY
This paper summarises the structure design of an MSP430-based ACS and laboratory test data are provided. The engineered ACS was tested in the laboratory conditions in two configurations: the open loop and in the feedback loop configuration (with a PID controller). Testing was done under the loading applied to the power interface VI-MR through an electromagnet coil. Experi-mental results show that application of the PI controller allowed a 30% reduction of the time needed to reach the preset current level and force produced by VI-MR in real application.
It is worthwhile to mention that the selected microcontroller MSP430 has a sufficient computational power to implement algo-rithms and allows for expanding the ACS structure such that it should incorporate extra sensors and actuators. ACS tests will be repeated once the final version of the vibration isolator VI-MR has been developed.