Sine-Sweep Input Generation With Minimum Energy Restriction for Obtaining a Precise Frequency Response Function in Industrial Servo Systems

This paper presents a sine-sweep input generation method to obtain a precise frequency response function (FRF) in industrial servo systems. Among the various system identification methods, the FRF method generally has rich information of the system especially in the high-frequency ranges which is important for resonance suppression and vibration control. Industrial servo systems are applied to many different types of plants that have many high-frequency components in broad frequency ranges. However, most servo drives have limitations in memory and computation power, which hinders using long excitation signals with various energies for different plants. This paper presents a method to obtain a precise FRF of the system with a novel sine-sweep input generation algorithm. In the proposed algorithm, the sine-sweep input is generated in real-time considering the energy of the input signal for each frequency. Using real industrial servo systems, it is demonstrated that the proposed sine-sweep input generation algorithm effectively obtains very accurate FRFs for various types of systems.