Search

The magnetic flux linkage is an essential parameter for many control approaches of electrical machines. In most common control algorithms, this quantity is determined either by current-based look-up tables (LUTs) or by integrating the measured pulsewidth-modulated (PWM) voltage. But accurately measuring pulse-modulated voltage is challenging due to its steep slopes, especially for SiC and GaN voltage-source inverters (VSIs) and time delays during switching. Notwithstanding, for modern position sensorless drives, knowledge of instantaneous phase voltages, and thus the magnetic flux or the volt-seconds, respectively, is required, for example for flux linkage estimation realizing high-precision field-oriented control. Common measurement techniques to determine PWM voltages are based on the sample-and-hold method executed with successive-approximation analog-to-digital converters (ADCs). However, this measurement method has the disadvantage that low-cost ADCs cannot capture high-frequency voltage components. High sampling rates of these integrated circuits (ICs) can prevent this issue but lead to rising costs. This article presents a novel measuring technique for precise volt-second sensing that considers high-frequency components in the average value of a PWM period by measuring the volt-second continuously. Thereby, it combines the advantages of analog circuit technology with the performance of a field programmable gate array (FPGA). The presented sensor prototype for volt-second sensing is based on the concept of an analog synchronous voltage-to-frequency converter (SVFC). The concept of this measuring instrument is to save additional expensive components by utilizing existing control hardware. This work shows how these savings are implemented for various applications. For this purpose, the working principle and the characteristic curve of the sensor are demonstrated and illustrated with simulation results. The implementation in hardware and software is then presented in detail before the assumptions are validated and evaluated with measurements from the test bench. Furthermore, the dynamic behavior of the sensor prototype during measurements of PWM voltages is investigated, and an analysis algorithm for the instantaneous volt-second value is developed.

This article investigates evaluation methods of thermal transient measurements to obtain the internal thermal structure of semiconductor devices. First, the study uncovers the limitations of a widely accepted standard method that uses frequency-domain deconvolution. An important finding is that the sideband of the time constant spectrum by the standard method has no physical meaning despite it beeing interpreted as a continuous spectrum for a long time. Second, by understanding the limitations of the existing method, the article proposes an alternative method that remodels the frequency-domain deconvolution as a regularized least squares problem in the time domain. With the benchmark of the true values of several thermal networks based on simulation, the proposed sparsity-promoting method demonstrates several advantages, including a better ability to identify adjacent parameters in the time-constant spectrum and the obtained structure function reducing relative error by an order of magnitude. The influence of varying noise levels has also been evaluated. Finally, a proof-of-concept experiment using a commercial power semiconductor device validates its effectiveness.

Permanent magnet synchronous machines (PMSMs) are widely used in various applications as high-power-dense and highly efficient drives. For utilizing the unused potential of present PMSMs towards a reduced component size, light weight, and increasing power density, thermal management is key. With the goal of real-time temperature monitoring on embedded hardware, a workflow is introduced to automatically derive a reduced order model on the basis of the Craig-Bampton method using linear time parameter-varying systems. The time varying parameters are identified with an optimization approach based on experimental data. Finally, the reduced thermal model meets the target accuracy of 10 °C for several operation points and several cooling conditions such as variations in oil flow rate and oil temperature. Moreover, the real-time capability of the model is demonstrated for a low-cost micro-controller unit.

This article proposes a power hybridization factor-dependent DC supply grid layout and fault protection for hybrid electric aircraft. It is based on a regulatory review and an analysis of the grid components regarding their impact on grid layout and protection. The findings are applicable to medium-voltage DC and high-temperature superconductor-based electrical drive trains.

This article presents a novel model predictive method for torque ripple and source current ripple suppression in switched reluctance machines (SRMs) when the bus voltage fluctuates. Based on the measured rotor position, phase current, and phase voltage, the states, including bus voltage, capacitor current, and phase flux-linkage, are calculated for the preparation of the prediction. On this basis, the machine torque and inverter current for all the possible switching states are predicted to construct the cost function. The operating state with the minimum value of cost function is considered as the optimal switch signal and applied to the control power converter, thus achieving the effect of simultaneous reduction of torque ripple and source current ripple. Experimental results show the proposed method has a significant effect on the suppression of both torque ripple and source current ripple even with bus voltage fluctuation, which verifies the effectiveness of the proposed method. The influence of the key factors, such prediction horizon, parametric variation, and weight coefficient, on the machine performance are further investigated. The proposed method is easy to be implemented in the control logic as well as robust to bus voltage fluctuations, enabling to promote the competitiveness of SRMs in electric vehicle application.