Grid impedance affects the stability and control performance of grid-connected power electronic devices, such as inverters used to integrate wind and solar energy. Adaptive control of such inverters, to guarantee stability under different grid conditions, requires online measurement of the grid impedance performed in real time. Such online measurement can be performed by injecting a current perturbation from the inverter into the grid and by reading the grid voltage responses. To minimize the impact on the inverter operation, the injection must be kept as small as possible while producing enough voltage perturbation that can be reliably measured and processed to extract its various frequency components. This paper proposes the use of a discrete-interval binary sequence (DIBS) for this application to minimize the injection. The DIBS is a computer-optimized binary sequence, where the energy is maximized at specified harmonic frequencies based on the expected grid-impedance characteristics. Experimental results based on a three-phase grid-connected inverter are presented to demonstrate the effectiveness of the proposed methods.

Grid impedance is an important parameter for the operation and control of grid-connected inverters used for the integration of solar, wind, and other distributed generation resources. Since the grid impedance usually varies over time and with grid operation conditions, online measurement is required for adaptive control of grid-connected inverters. Existing online measurement methods based on impulse perturbation and Fourier analysis require large current or voltage injection that may interfere with normal operation of the inverter. This paper proposes the use of maximum-length binary sequence (MLBS) injection and averaging Fourier techniques to overcome the drawbacks of impulse injection. Experimental results based on a three-phase grid-connected inverter are presented and used to demonstrate the effectiveness of the proposed methods.

This paper introduces a feedforward charge control scheme, which controls the velocity of a piezoelectric actuator by the current fed to the actuator. The velocity-current relation, however, is not linear over the entire motion range, and therefore, information about the actuator voltage is also utilized. The required current is estimated using an actuator model, which is separated in two phases combining individual models for a motion phase and a static phase. The method is verified by a series of experiments, where a piezoelectric actuator is moved with variable velocities and displacements. During the experiments, the current is fed to the actuator using a current driver consisting of a voltage amplifier, a precise current meter, and a controller. The results show a significant improvement in comparison to open-loop voltage control; the hysteresis is less than 2% and the drift approximately 1%. This indicates that the motion of piezoactuators is a function of current and voltage, and does not depend considerably on the motion history. Therefore, a sensorless control scheme to overcome the hysteresis and drift would be feasible.

Switched-mode power converters are extensively used to power various electronic loads and processes. The use of low-cost components with high parameter variation together with the requirement of short time to market has set new challenges for the designers such as the necessity to characterize a large number of converters before production ramp-up. It has been shown that the frequency-domain characterization yields most useful information on the quality of a converter. The conventional frequency response analyzers can be used for measuring the responses, but the time required for one measurement is too long for characterizing hundreds or thousands of converters. Recent studies show that the time for single measurement can be drastically reduced by using broadband excitation signals such as the maximum length pseudorandom binary sequence (MLBS). The MLBS excitation signal is not well suited for measuring the voltage or current loop gains because of their sensitivity to the disturbance signals. This paper proposes a new method based on the use of MLBS but with a specified Fourier amplitude spectrum in order to avoid the said problem. The method is verified both with simulation and practical measurements, and the results are compared to the conventional MLBS technique.