Aimed at the problems in an electric-field coupled wireless power transfer (EC-WPT) system for underwater unmanned vehicle (UUV) such as poor anti-offset performance and large leakage field radiation, a UUV EC-WPT system with anti-offset and leakage field shielding characteristics is proposed. First, a novel five-stage plate coupler suitable for UUV is given, the ten coupling capacitors generated by the coupler are simplified, and an equivalent model of the coupler is established. Second, the EC-WPT system with dual LC compensation is taken as an example, a calculation method for system output parameters is formulated, and circuit simulation is also carried out. The electric field around the system is analyzed using the finite element method according to the electrode plate voltage obtained from simulations, so as to verify the leakage field shielding characteristics of the novel coupler. Results show that there is almost no leakage field inside the UUV in the system. Finally, an experimental device was built to verify the feasibility of the proposed scheme for wireless power supply to UUV from the aspects of energy transmission performance and anti-offset performance. In the experiment, 74 W power transmission was achieved, with an efficiency of 85.6%. When the energy receiver was offset from three dimensions, the system’s output power and efficiency changed little.

In the drive system for a permanent magnet synchronous motor (PMSM) used in electric vehicles, optimization strategies such as maximum torque per ampere (MTPA) are often applied to optimize the motor’s operating efficiency. However, the optimization methods based on offline data look-up table and motor parameter identification are difficult to cope with the challenges of motor parameter saturation and permanent magnet temperature change at the same time. In this paper, a method of synchronous compensation of electromagnetic torque and optimal current angle based on data-driven model is proposed. First, a response surface model of flux linkage is established based on the multi-temperature offline data, and a wide-temperature torque model of the motor is obtained by combining the online status information with the offline model through the least square identification of flux linkage error parameters. Then, the electromagnetic torque compensation and online current angle optimization are realized based on this model. Finally, experimental results show that under the premise that the response speed of the proposed method is not inferior to that of the lookup table method, the novel method allows the PMSM to obtain better electromagnetic torque accuracy and torque per ampere over the wide temperature range.

The active neutral-point-clamped (ANPC) three-level structure is more advantageous in high-power motor drive systems because it can optimize the loss distribution in switching devices. To improve the performance of a drive system under operation conditions with a low switching frequency, a multi-mode synchronous space vector pulse width modulation (SSVPWM) switching strategy is studied in this paper. Aimed at the problem of current and torque impact caused by the switching between different modes, a multi-mode SSVPWM switching strategy based on pulse insertion is proposed for ANPC three-level inverters. By reconfiguring the stator flux linkage from the motor, the stator flux linkage deviation at the switching position is calculated online, the selection of ANPC redundant zero-state is combined at the switching moment, and the smooth switching between multiple modes is achieved with compensation by inserting the voltage pulse. The correctness and effectiveness of the proposed strategy were verified through simulation and experimental results.

Flying-start is one of the key technologies used in the traction electric-drive system of a sensorless permanent magnet synchronous motor, and one of the common methods is to short-circuit the motor through triggering zero-voltage vector pulse to realize the flying-start below the rated speed. To accurately select the short-circuit pulse time and the parameters of short-circuit pulse time interval, a flying-start method with adaptive adjustment of parameters at different rotation speeds is proposed, which estimates the motor’s approximate rotation speed by injecting a test zero-voltage vector pulse under various free rotation conditions. According to the obtained approximate rotation speed, the short-circuit pulse time and the parameters of short-circuit pulse time interval are adjusted adaptively. As a result, the contradiction that the traditional fixed parameter method is unable to strike a balance between the inverter overcurrent and the estimation accuracy of rotor position is solved, and the estimation accuracy of rotor position during restarting is significantly improved. Finally, the correctness and effectiveness of the proposed method was verified by simulation and experimental results.

In maglev long-stator linear synchronous motor drive systems, multiple sources of disturbances during their operation can lead to significant fluctuations in motor parameters, thereby reducing the control effect of model predictive approaches. To address this issue, a robust predictive control algorithm based on error feedback correction is proposed. Under this strategy, the error between the predicted and measured values of current is used as a feedback signal. By designing an appropriate feedback structure to adjust the output voltage of the conventional deadbeat predictive controller, the proposed method can effectively mitigate steady-state tracking errors and instability problems caused by mismatches in inductance, resistance and flux linkage. Finally, a maglev linear synchronous motor platform was constructed for comparative verification, and experimental results validated the effectiveness of the proposed algorithm.

To enhance the robustness of rotor radial magnetic suspension, a rotor magnetic suspension control strategy for a bearingless flux-switching permanent magnet motor (BFSPMM) is proposed, which is based on the combination of the nonlinear feedback linearization theory and internal model control method. Under this control strategy, the rotor dynamic model of the BFSPMM is taken as the internal model, the difference between the actual motor and the internal model output is expanded into a novel nonlinear error system by a second-order filter, and the given suspension current is calculated according to the input-output feedback linearization theory. Simulation and experimental results show that the proposed control strategy can effectively realize the high robust operation control of rotor radial magnetic suspension.

With a large number of renewable energy devices being integrated into DC distribution network, the factors such as the random fluctuation of photovoltaic power generation and unpredictability of loads will have an impact on the system stability. During its operation, the system has characteristics of being dominated by power electronic equipment and time-varying. To ensure the efficient and stable operation of the DC distribution network, it is crucial for obtaining the impedance characteristics of power electronic equipment in the distribution network accurately in real time. A wide-band impedance measurement method based on discrete interval binary sequence (DIBS) with optimized spectrum is proposed, and the phase optimization formula and Nelder-Mead algorithm are used for optimizing the phase of each frequency component of the DIBS generating signal. In addition, the methods of generating signal and amplitude selection are improved, so that the signal spectrum is focused on the target frequency band, and the measurement efficiency and accuracy are improved. To verify the effectiveness of the proposed method, the time-domain waveforms and spectra of the multi-sinusoidal signal and optimized DIBS signal are compared, and simulations are carried out in Matlab/Simulink to verify the method. Results show that the injected voltage amplitude of DIBS is smaller and can accurately measure the impedance characteristics of power electronic equipment with the same signal spectral energy. Finally, passive load impedance measurement experiments were carried out on a hardware platform to verify the effectiveness of the proposed method.

To study the explosion risk of thermal-runaway gas of ternary lithium-ion batteries, an experimental platform for the thermal-runaway gas explosion of lithium batteries was independently built. Based on the risk assessment theory, the safety of lithium batteries in different states-of-charge (SOHs) was quantitatively evaluated. The probability of thermal-runaway gas explosion was represented by the product of the ignition parameters of lithium batteries and the risk degree of gas explosion, the result of gas explosion was characterized by the explosion severity, and the risk of thermal-runaway gas explosion was evaluated comprehensively through experiments. Experimental results show that the battery with a high SOC is more likely to cause thermal runaway in a more intense manner, its thermal-runaway gas explosion range is wider, and the corresponding risk degree is greater. Within the explosion limit, the batteries with higher SOCs at the same location have higher risks. With an increase in the thermal-runaway gas concentration in the mixed gas produced by batteries with different SOCs, all the maximum explosion temperature, maximum pressure, maximum pressure rise rate and explosion risk show a changing rule of first increasing and then decreasing, and the parameters reach their maximums near the middle value between the upper and lower explosion concentrations.

Buck converters in DC microgrid are frequently used to transform the DC voltage of high-voltage bus, and the electrolytic capacitor therein is the most vulnerable component. The analysis of capacitance degradation parameters shows that the equivalent series resistance (ESR) is the best health indicator for electrolytic capacitors, and it is of significance for online monitoring the variations in ESR values in real-time. A non-intrusive online ESR monitoring method for the output capacitor of a Buck converter is proposed, which is based on the output voltage. In this method, the AC component of inductive current is fitted by the drive signal, and the real-time online calculation model of ESR is derived by combining the AC component of output voltage. Under this scheme, the additional current sensors and high-precision trigger sampling devices are not required, and the accurate calculation of the output capacitance ESR of the Buck circuit under different working conditions can be achieved only by sampling the output voltage and combining the drive signal of the switch tube. An experimental platform for the ESR monitoring system was built, and experimental verification was also carried out. Experimental results verified the effectiveness of the proposed method.

The thermal characteristic modeling of silicon carbide (SiC) power devices is a key aspect in their design and development. The structure function method is crucial for characterizing the devices’ thermal characteristics, focusing on acquiring the time-constant spectrum of input data via deconvolution algorithms. However, the environmental noise inevitably affects actual tests, thereby demanding more sophisticated deconvolution algorithms for the structure function. Therefore, a deconvolution structure function algorithm based on improved Wiener filtering is proposed, effectively addressing the noise problem in the measurement data during the transient thermal analysis. First, a structure function model based on Wiener filtering deconvolution is put forward, and the algorithm’s effectiveness is validated through analyzing the TDIM-Master benchmark problem. Second, different levels of noise are added to the benchmark problem to explore the signal quality and peak thermal resistance of the time-constant spectrum across various noise conditions. The accuracy and superiority of the improved method when dealing with noises are verified through a comparison with the conventional algorithm. Finally, the transient thermal resistance of two different types of SiC power devices were measured under conditions with or without thermal silicone grease, and the structure function analysis was also performed, thus verifying the stability and applicability of the proposed algorithm under real measurement environment.