Abstract

The mechanical system of a drive can often be modeled as a two- or three-mass-system. The load is coupled to the driving motor by a shaft able to perform torsion oscillations. For the automatic tuning of the control, it is necessary to know the mathematical description of the system and the corresponding parameters. As the manpower and setup-time necessary during the commissioning of electrical drives are major cost factors, the development of self-operating identification strategies is a task worth pursuing. This paper presents an identification method which can be utilized for the assisted commissioning of electrical drives. The shaft assembly can be approximated as a two-mass non-rigid mechanical system with four parameters that have to be identified. The mathematical background for an identification procedure is developed and some important implementation issues are addressed. In order to avoid the excitation of the system with its natural resonance frequency, the frequency response can be obtained by exciting the system with a Pseudo Random Binary Signal (PRBS) and using the cross correlation function (CCF) and the auto correlation function (ACF). The reference torque is used as stimulation and the response is the mechanical speed. To determine the parameters, especially in advanced control schemes, a numerical algorithm with excellent convergence characteristics has also been used that can be implemented together with the proposed measurement procedure in order to assist the drive commissioning or to achieve an automatic setting of the control parameters. Simulations and experiments validate the efficiency and reliability of the identification procedure.

참고문헌 (11)

1. R. Isermann: Parameterschatzung dynamischer Systeme Ubersichtsaufsatz, Automatisierungs-technik, Heft 6, 1988 
2. G. Ellis: Autotuning Compliant Servo Systems using Optimized Bi-Quad Filter, PCIM 2004, Nuremberg, Germany 
3. J.M. Pacas, S. Villwock, T. Eutebach: Parameter-Identification of a Two-Mass-System in the Frequency Domain, PCIM 2004 Nuremberg, Germany 
4. Press, Flannery, Tenkolsky, Vetterling: Numerical Recipes, Cambridge University Press, 1986 
5. J.-K. Ji, S.-K. Sui: Kalman Filter and LQ Based Speed Controller for Torsional Vibration Suppression in a 2-Mass Motor Drive System, IEEE Transactions on Industrial Electronics, vol. 42, no. 6, 1995 
6. R. Isermann: Identifikation dynamischer Systeme 1, 1992 
7. R.G. Schwarz: Identifikation mechanischer Mehrkorpersysteme, Diss. Tech. University of Munich, Germany, 1980 
8. P. Eykhoff: System Identification - Parameter and State Estimation, 1977 
9. H. Unbehauen, G. P. Rao: Identification of continuous systems, North Holland, 1987 
10. J.O. Krah: Bode Plot based Auto-Tuning - Enahnced Solution for High Performance Servo Drives, PCIM 2004 Nuremberg, Germany 
11. F. Schutte: Automatisierte Reglerinbetriebnahme in elektrischen Antrieben mit schwingungsfahiger Mechanik, Diss., University of Paderborn, 2002 

이 논문을 인용한 문헌 (0)

1. 이 논문을 인용한 문헌 없음