Breakdown voltage (BV), on-state voltage (Von), static latch-up voltage (Vlu), static latch-up current density (Jlu), and threshold voltage (Vth), etc., are critical static characteristic parameters of an IGBT for researchers. Von and Vth can characterize the conduction capability of the device, while BV, Vlu, and Jlu can help designers analyze the safe operating area (SOA) of the device and its reliability. In this paper, we propose a multi-layer artificial neural network (ANN) framework to predict these characteristic parameters. The proposed scheme can accurately fit the relationship between structural parameters and static characteristic parameters. Given the structural parameters of the device, characteristic parameters can be generated accurately and efficiently. Compared with technology computer-aided design (TCAD) simulation, the average errors of our scheme for each characteristic parameter are within 8%, especially for BV and Vth, while the errors are controlled within 1%, and the evaluation speed is improved more than 107 times. In addition, since the prediction process is mathematically a matrix operation process, there is no convergence problem, which there is in a TCAD simulation.
Due to the complexity of the 2D coupling effects in AlGaN/GaN HEMTs, the characterization of a device’s off-state performance remains the main obstacle to exploring the device’s breakdown characteristics. To predict the off-state performance of AlGaN/GaN HEMTs with efficiency and veracity, an artificial neural network-based methodology is proposed in this paper. Given the structure parameters, the off-state current–voltage (I–V) curve can therefore be obtained along with the essential performance index, such as breakdown voltage (BV) and saturation leakage current, without any physics domain requirement. The trained neural network is verified by the good agreement between predictions and simulated data. The proposed tool can achieve a low average error of the off-state I–V curve prediction (Ave. Error < 5%) and consumes less than 0.001‰ of average computing time than in TCAD simulation. Meanwhile, the convergence issue of TCAD simulation is avoided using the proposed method.
Nowadays, the impact ionization coefficient in the avalanche breakdown theory is obtained using 1-D PN junctions or SBDs, and is considered to be a constant determined by the material itself only. In this paper, the impact ionization coefficient of silicon in a 2D lateral power device is found to be no longer a constant, but instead a function of the 2D coupling effects. The impact ionization coefficient of silicon that considers the 2D depletion effects in real-world devices is proposed and extracted in this paper. The extracted impact ionization coefficient indicates that the conventional empirical impact ionization in the Fulop equation is not suitable for the analysis of 2D lateral power devices. The veracity of the proposed impact ionization coefficient is validated by the simulations obtained from TCAD tools. Considering the complexity of direct modeling, a new prediction method using deep neural networks is proposed. The prediction method demonstrates 97.67% accuracy for breakdown location prediction and less than 6% average error for the impact ionization coefficient prediction compared with the TCAD simulation.