A novel high performance SiC asymmetric cell trench MOSFET with split gate (SG) and integrated junction barrier schottky (JBS) diode (SGS-ATMOS) is proposed for the first time. The shielding effect provided by the SG structure not only reduces the gate-drain capacitance ( C _gd ) but also alleviates the electric field crowding in the dielectric layer at trench corner. The integrated trench JBS diode bypasses the PiN body diode while obtaining good double protection from the SG and p-type shielding region. Therefore, not only the MOSFET but also diode performance is significantly improved for the proposed structure. Numerical analysis results show that compared with the conventional asymmetric cell trench MOSFET (Con-ATMOS), the high frequency figure of merit (HFFOM ¹ , R _on · C _gd ) is reduced by 92.5% and the Baliga figure of merit (BFOM, BV ² /R _on,sp ) is increased by 57.2%, respectively. In addition, the forward conduction voltage drop ( V _F ), reverse recovery charge ( Q _rr ) and peak reverse recovery current ( I _rrm ) of the diode are reduced from 3.10V, 1.95μC/cm ² and 68.0A for the Con-ATMOS to 1.56V, 0.97μC/cm ² and 35.9A for the proposed SGS-ATMOS, respectively. Compared with the Con-ATMOS, the turn-on loss ( E _on ) and turn-off loss ( E _off ) of the proposed device are reduced by 33.3% and 33.0%, respectively. The E _on and E _off of the proposed device are also 33.6% and 30.0% off compared with the Con-ATMOS with external JBS diode, respectively. The temperature characteristics of the SGS-ATMOS are also discussed and it is found that the proposed device exhibits good performance at high temperature.

A global and real-time control with embedded hardware artificial neural network (ANN) for imbalance current suppression of parallel connected IGBTs is firstly proposed in this paper. This method focuses on control strategy and control execution. The former one is realized by porting the ANN based PID (ANN-PID) strategy in the control loop to yield the real-time and adaptive characteristics without IGBT quantity limitation. The latter one is realized by designing the IGBT gate quantity of charge regulator (GQR) to execute the command from ANN-PID controller. The evaluation of ANN-PID algorithm results 0.023% mean error in IGBT current control that reveals the feasibility of the proposed method. A full prototype with FPGA based hardware accelerator for ANN-PID computing, including the designed GQR circuit, has been built for realization and qualification in a buck converter with parallel connected IGBTs. The experimental results show that the performance of proposed method in imbalance current suppression is improved about 3.5 to 5.5 times as the load increase from low to high with the advantage of immunity to load change and the current imbalance can be suppressed within 4%.

Composting is an important treatment method for Camellia oleifera shell and goat dung, which is crucial for the development of tea oil industry and goat breeding industry. Ventilation is the important regulatory factor in composting process, and high degree of maturity is the principal requirement for safe application of composting product. In the study, four treatments were designed as every 5 days turning (A1), every 7 days turning (A2), every 10 days turning(A3), and every 15 days turning (A4) for evaluating the maturity of Camellia oleifera shell co-composted with goat dung and optimizing turning frequency. During composting, TN, , GI, Solvita maturity index was increased along with composting process, while , C/N shown an opposite trend. For all treatments, the longest thermophilic period (over 50°C), highest total nutrient and lowest C/N ratio were observed in A2. A turning frequency of every 7 days for co-compost of Camellia oleifera shell and goat dung could enhance the composting product quality. According to the analysis of spearman correlation, TN, C/N ratio, and GI could be used to comprehensively evaluate the compost maturity.

The PNP bipolar junction transistors (BJTs) are susceptible to surface recombination in both the emitter–base depletion and neutral base regions when their passivation oxide layers are damaged by ionization irradiation, leading to an increase of base current. The previous model of excess base current proposed in 2015 omits the damage in the neutral base, and the impact of oxide charges, which results in an underestimation of bias sensitivity of excess base current. This work presents a physics-based improvement of the previous model, and it provides a much better fit to the experimental data for the bias dependence of excess base current for the 2N2907 vertical and gate controlled lateral PNP BJTs subjected to γ-ray irradiation. The model suggests that the damage in the neutral base can be significant for a lateral PNP with wide base, operated at high bias voltage, and exposed to a low dose. Furthermore, the model provides a simple and practical method to separate and estimate the total oxide charge and increased interface trap densities from excess base current readings on irradiated PNP BJTs. The extracted values of interface trap densities show less dependence on E-B bias voltage than with previous models, which provides evidence of its improved precision.