Abstract:To address the issue of high freewheeling losses caused by SiC Metal-Oxide-Semiconductor Field-Effect Transistors(MOSFETs) during the freewheeling process in power modules, an integrated Low-Barrier Diode SiC MOSFET(LBD-MOS) structure is proposed. The total power consumption of LBD-MOS and the conventional SiC MOSFET(CON-MOS) under the same area is investigated. Simulation results show that the freewheeling voltage drop(U_F) of LBD-MOS is 1.6 V, which is 50% lower than that of CON-MOS; the switching loss (E_switch) of LBD-MOS is 187.3 μJ, which is 6% lower than that of CON-MOS. Under operating conditions with a frequency of 10 kHz and a duty cycle of 50%, the total power consumption of LBD-MOS is reduced by 22.6% compared to that of CON-MOS. LBD-MOS is suitable for applications where the freewheeling ratio is higher than 50% and the switching frequency does not exceed 1 MHz.

Abstract:The heteroepitaxial process of Gallium Nitride High Electron Mobility Transistors (GaN HEMTs) leads to the presence of a trapping effect in GaN HEMT devices. This effect causes dynamic changes in the on-resistance of the devices under continuous transient operating conditions, known as dynamic on-resistance, which is higher than the theoretical value under static conditions. This dynamic on-resistance can pose a threat to the stability of power systems. Therefore, it is necessary to investigate efficient and accurate testing methods for the dynamic on-resistance of GaN HEMT devices. The mechanism of dynamic on-resistance generation in GaN HEMT devices is introduced in this paper. In combination with practical testing requirements, a novel clamping circuit based on an ultra-high-speed voltage feedback operational amplifier is designed. The Pspice simulation tool is employed to simulate this new clamping circuit and compare it with other commonly used existing clamping circuits. The results show that this circuit can more rapidly and accurately read the drain voltage of the device after it transitions from the off-state to the on-state. It also enables the characterization of the device's on-resistance under different bias voltages and frequencies.

Abstract:In recent years, significant progress has been made in the development of domestic 6-inch SiC-based Gallium Nitride High Electron Mobility Transistors(GaN HEMTs). This paper investigates the multi-layer dielectric stress modulation technique and high-consistency backside etching technique, which are integrated into the 6-inch process. When operating at 48 V, the 0.5 μm process achieves an output power density of 8.6 W/mm at 3.5 GHz, with a power gain of 15 dB and a Power Added Efficiency(PAE) of 58.5%. When operating at 28 V, the 0.25 μm process achieves an output power density of 5.5 W/mm at 10 GHz, with a power gain of 8.7 dB and a PAE of 55.2%. The reliability of GaN devices is evaluated through High-Temperature Operating Life (HTOL) and High-Temperature Reverse Bias(HTRB) tests, with the saturation output current of the devices changing by less than 10% after 1 000 hours. The 20 W and 40 W power transistors, as well as X-band Monolithic Microwave Integrated Circuit (MMIC) power amplifiers, are fabricated to validate the process technology, with measured on-wafer yields of 90%, 86%, and 77%, respectively. The results indicate that domestic6-inch SiC-based GaN HEMTs have application potential below the Ku-band.

Abstract:The working principle of a nanosecond-level high-voltage pulse generation circuit based on a Drift Step Recovery Diode(DSRD) is introduced. The circuit is modeled, and the key circuit parameters that affect the pulse output characteristics are discussed based on the model. In the experiment, a high-voltage Silicon Carbide(SiC) DSRD device developed in the author's laboratory is employed to generate a nanosecond-level pulse voltage with a peak value of 2.27 kV and a rise time of 1.846 ns on a standard load of 50 Ω. By changing the key parameters in the circuit, the variation trends of the pulse voltage peak obtained from the tests are consistent with the analysis from the model, which validates the rationality of the model. Considering the voltage overshoot issue at the drain-source terminals of the switch during the turn-off process, a buffer capacitor is connected in parallel across the drain-source terminals. The parameters of the buffer capacitor are adjusted experimentally to reduce the overshoot voltage at the drain-source terminals without affecting the peak voltage of the DSRD pulse discharge.

Abstract:Based on the GaAs substrate Enhanced/Depletion-mode pseudomorphic High Electron Mobility Transistor(E/D pHEMT) process, a three-bit adjustable 1 400 ps Digital-Controlled Delay (DCD) chip operating in the 0.5~6 GHz frequency range has been developed. The chip measures 3.60 mm ×4.00 mm ×0.07 mm and integrates a three-bit digital-controlled delay line and a 3-bit parallel port drive circuit. Within the 0.5~6 GHz range, the DCD chip exhibits insertion loss of less than 11 dB, with insertion loss variation of less than ±0.5 dB. The Voltage Standing Wave Ratio(VSWR) for both input and output is less than 1.5 across all states. The 1 400 ps delay error can be internally adjusted to ±4 ps, achieving a delay quantity at the nanosecond level. By incorporating additional adjustable units and bonding cut-off methods, the delay accuracy is enhanced to 3‰. The chip features broadband operation, high precision, large delay quantity, and a compact size, making it well-suited for applications in antenna systems.

Abstract:A novel Avalanche Triggered MOS-Controlled Thyristor(AT-MCT) is proposed, which achieves high current peak, high current rise capability (di/dt), and non-activation protection in the non-operating state during capacitor pulse discharge. The device incorporates a highly doped N Avalanche Layer(N-AL) buried in a Pbody, with an N+ region near the cathode separated from the MOS structure. When a gate voltage is applied, the channel generated by the MOS transfers the potential of the N-drift region to the N-AL. The highly doped N-AL experiences avalanche due to the electric field peak, and the generated electron-hole pairs serve as the base current of the thyristor, enabling the AT-MCT to rapidly establish a self-feedback mechanism. Meanwhile, the positive feedback process established by the avalanche significantly improves the two-dimensional transient carrier transport effect, increases the effective conduction area of the cell during transient turn-on, and thus achieves more efficient energy conversion. The AT-MCT exhibits a 40% increase in current peak and a 31% increase in di/dt capability compared to Cathode-Shorted MCT(CS-MCT). Moreover, by designing the doping concentration of the N-AL, non-activation protection in the non-operating state can be realized, thereby enhancing the reliability of the pulsed power system.

Abstract:The performance and defect detection of Low-Emissivity(Low-E) glass are of great significance in its production. Currently, for Low-E glass with single, double, or multiple silver coatings, defect detection mainly relies on the analysis of visible light images, which presents certain limitations such as insufficient contrast and low identification efficiency. Terahertz waves can penetrate most non-conductive materials, making them highly effective for internal defect detection and structural analysis. Therefore, in this study, Terahertz Time-Domain Spectroscopy(THz-TDS) and scanning imaging technology based on this principle are employed to detect the coating structure and surface scratches of Low-E glass. By utilizing reflective THz-TDS and scanning imaging techniques, the waveform and intensity of the reflected terahertz pulse signals from glasses with different coating structures are investigated and the defect imaging is successfully achieved. This study demonstrates the feasibility of THz time-domain spectroscopy in the identification of glass coatings and defect detection.

Abstract:Based on an ultrashort H-plane coupler and a cut-off structure, a high frequency structure of a W-band Periodically Cusped-Magnetic(PCM) field focused wideband sheet-beam Traveling-Wave Tube(TWT) is shortened by 27.7 mm. This approach allows the TWT to generate more than 200 W of power within a bandwidth of 13 GHz(90~103 GHz). The interaction between the electron beam and microwave in the ultrashort cut-off structure is investigated. The study shows that the ultrashort cut-off structure can mitigate the divergence of bunched electrons at the cut off structure when the first section of the Slow-Wave Structure(SWS) is long enough, thereby significantly enhancing the output power and bandwidth of the TWT. Therefore, long cut-off structures, high input signal voltages, and low electron gun voltages should be avoided when a wideband TWT is designed.

Abstract:To enhance the angular stability of Frequency Selective Surface(FSS) absorbers, a broadband absorber is designed by using an Indium Tin Oxide(ITO) resistive film in combination with a dielectric compensation layer. This absorber features flexibility, transparency, polarization insensitivity, and high angular stability. The 90% absorption bandwidth of the absorber covers the frequency range of 5.66 to 22.98 GHz, achieving a relative bandwidth of 121.0%. When the incidence angle of the electromagnetic wave varies from 0° to 60°, the absorber maintains an absorption rate of over 80% in the frequency band of 7.00 to 22.86 GHz. Fabrication and experimental testing have shown that the designed FSS absorber exhibits good transparency and flexibility, with test results matching well with simulation results, thereby validating the rationality of the design. Moreover, when the sheet resistance of the resistive film and the relative permittivity of the dielectric layer fluctuate within certain ranges due to manufacturing processes, the absorber structure can still achieve broadband absorption and maintain high angular stability. This significantly enhances the engineering practicability of the designed absorber in applications such as stealth and electromagnetic interference suppression for the windows and doors of aircraft and vehicles.

Abstract:A dual-band amplitude and phase dual-regulation metasurface is proposed to achieve arbitrary beamforming shapes at two preset frequencies. By combining two resonators of different sizes in the metasurface unit, independent amplitude and phase control can be realized at the two frequency points. Under linear polarization excitation, amplitude regulation is achieved by rotating the resonators, while phase regulation is realized by changing the size of the resonator openings. An improved array synthesis algorithm, combining Taylor synthesis and genetic algorithm, is employed to obtain far-field patterns that closely match the desired shapes. The resulting phase range is relatively small, which is conducive to achieving high-efficiency array beamforming. A metasurface array that can generate a flat-top beam at 6.25 GHz and a cosecant-squared beam at 15 GHz is designed. The full-wave simulation results show that the beams generated by the designed metasurface array match the target beams very well, with low sidelobe levels. This work provides a new approach to far-field array beamforming using dual-band metasurfaces.

Abstract:To investigate the potential damage caused by High-Power Microwave Weapons(HPMW) attacking radar systems through exposed cable backdoors, a cable model between radar units is established. The effects of different cable types, installation heights, electric field directions, and cable lengths on coupling signals are compared and analyzed. The characteristics of coupling signals are measured through irradiation experiments, and the impact of coupling signals on radar is analyzed. Corresponding protective circuits are designed, and other protective measures are discussed. Theoretical analysis and experimental results indicate that the peak voltage and waveform of the coupled signal are greatly affected by factors such as cable type, installation height, and electric field direction, while the resonant frequency of the coupled signal is greatly affected by cable length. The threat of coupling signals can be reduced through technical protection measures and reasonable cable layout.

Abstract:High-gain DC/DC converters have a promising application prospect in new energy generation and DC microgrids, and good dynamic characteristics are the foundation of their applications. Compared with traditional switching converters, high-gain converters face challenges such as high computational complexity and high model order in modeling. A new modeling method is proposed for high-gain converters based on system identification. It analyzes the working principle of the three-winding Boost?Forward converter and establishes a small-signal model of the converter using the state-space method. The correctness of the established model is verified through simulation. The sources of modeling errors in the state-space method are analyzed. The small-signal model of the converter is initially extracted using the least squares method, with a system model accuracy of 91.43%. Subsequently, an improved Particle Swarm Optimization(PSO) algorithm is employed to accurately extract the small-signal model, achieving a system model identification accuracy of 94.62%. Numerical experimental results demonstrate the correctness of the proposed identification method. The results have high reference value for the modeling and control loop design of complex converters.

Abstract:Aiming at the problem that existing jumping robots cannot precisely control their jumping posture angles, a posture control method for jumping robots based on flywheel reaction force regulation is proposed. During the jump, the reaction force generated by the continuous rotation of the flywheel can be utilized to adjust the robot's posture angle. Each flywheel regulator is controlled using a fuzzy adaptive Proportional-Integral-Derivative(PID) controller. Compared with traditional PID controllers, fuzzy adaptive PID can more quickly adjust the robot's posture angle with smaller tracking errors.Simulation results show that, compared with traditional posture control methods, using flywheel regulation can achieve real-time control of the jumping robot's posture angle more quickly. Especially during continuous jumping, the flywheel can better absorb intermittent impact momentum, allowing the robot's posture angle to remain stable over a larger range. In future work, a physical prototype of the jumping robot will be considered for construction to conduct obstacle-crossing tests in both indoor and outdoor environments.

Abstract:To address the issue of high implementation complexity in compensating for nonlinear distortion in short-range optical fiber communication systems using Intensity Modulation and Direct Detection(IMDD), a method based on an Adaptive Artificial Neural Network(ANN) equalizer is proposed. By considering the characteristics of optical communication networks, the number of nodes in the input layer and hidden layer of the ANN, as well as the number of training samples, are determined. Taking into account the noise in the optical fiber system and the distortion caused by dispersion, the training samples are expanded to enhance the generalization capability of the ANN equalizer. The weights of the ANN equalizer are adaptively adjusted, and small changes in the weight values are utilized to track channel fluctuations, thereby alleviating the problem of weight offset caused by variations in the fiber channel parameters due to changes in environmental conditions. Experimental results show that compared to a non-adaptive ANN, the proposed adaptive ANN has advantages in terms of overall gain, computational complexity, and memory requirements. The model demonstrates strong robustness in optical fiber communication under noisy conditions and has practical value.

Abstract:Aiming to address the issue of poor new energy accommodation capacity caused by single-objective optimization algorithms in the current coordinated dispatch process of source-grid-load-storage, a multi-objective optimization algorithm is combined to propose an optimized solution method for the coordinated dispatch problem of source-grid-load-storage. With the goals of minimizing dispatch costs and maximizing renewable energy accommodation, a multi-objective optimization function for coordinated dispatch of source-grid-load-storage is defined. Reasonable constraints are set from four aspects: energy components, main grid energy procurement, flexible load response, and energy storage devices. With the assistance of rough set theory, the weight coefficients of each dispatch optimization objective function are determined. An improved whale optimization algorithm with nonlinear weights is introduced to solve the multi-objective optimization function and derive the optimal coordinated dispatch plan. Experimental results show that after applying the dispatch optimization plan generated by the proposed method, the new energy accommodation percentage of the active distribution network reaches 97.25%, significantly enhancing the new energy accommodation capacity of the power system.

Abstract:In response to the issue of poor authentication effectiveness in traditional encryption and authentication systems, an Internet of Things(IoT)-based 5G power business user identity encryption and authentication system is proposed. In terms of hardware, an encryption chip, a communicator, and an identity recognizer are designed. On the software side, leveraging the information transmission model of the Internet of Things, the identity information of power business users are input to ensure the integrity of the identity information. Using cryptographic Hash function, the Hash values of different data information are calculated to encrypt the user identity information. The authentication of the encrypted user identity is completed through the authentication code generated by the terminal device. Through the above hardware and software designs, the 5G power business user identity encryption and authentication system is completed. In simulation experiments, compared with previous 5G power business user identity encryption and authentication systems, the proposed IoT-based 5G power business user identity encryption and authentication system has an average memory leak value of 32 MB, with better authentication performance and greater application value.