IGBT Model Research Articles

Modeling and Analysis of IGBT Based on Two-Dimensional Charge Distribution

Modeling and Analysis of IGBT Based on Two-Dimensional Charge Distribution

On-line junction temperature identification of converter power devices

Abstract Traction converters often fail or age due to long operation hours, complex conditions, and frequent braking, mainly caused by overheating or temperature fluctuations. Monitoring the temperature at the junction of power device is crucial for system reliability. To achieve this, the electric-heat coupling models of IGBT are studied. Models for IGBT’s switching and conduction processes were created using its data sheet, yielding a power loss model linked to junction temperature. A Foster thermal network model for IGBT was then established, deriving the equation for temperature change when losses act as heat sources. Combining these models, an electric-heat coupling model was built on the SIMULINK platform, applied to predict and simulate junction temperature in a two-level inverter, generating a temperature-time image. Finally, an experimental test platform for a two-level SVPWM-modulated inverter was set up, using an infrared sensor to measure IGBT’s steady-state junction temperature. This was compared with simulation results to verify the model’s accuracy.

Loss Modeling and Testing of 800-V DC Bus IGBT and SiC Traction Inverter Modules

Loss Modeling and Testing of 800-V DC Bus IGBT and SiC Traction Inverter Modules

The influence of external conditions on the physical characteristics of Press-Pack IGBTs

In order to solve the problem of unclear factors affecting the physical characteristics of Press-Pack IGBT devices, this paper explores the influence characteristics of the two main external conditions of fixture pressure and heat dissipation capacity on the physical parameters of Press-Pack IGBTs by building IGBT models with different chip structure layout considering various physical characteristics of materials. The results show that the effect of fixture pressure on the stress characteristics of the device is more obvious than the heat dissipation ability of the device, and the water-cooling heat flux has more influence on the electrical-thermal characteristics of the device. The research results provide a reference for the external conditions and structure selection of Press-Pack IGBTs.

Improved Parameterization Methodology for a Field-Stop Trench-Gate IGBT Physical Model by Switching Feature Partitioning

Accurate simulation of IGBT characteristics is an essential task to predict its electrical behaviors. Numerous physics-based IGBT models have been developed and proven to produce satisfactory simulation results. However, unavailability of device parameters is the key roadblock to application of these physics-based models. Additionally, the coupling between device parameters renders parameter extraction very challenging and heavy. In this paper, a novel parameter extraction scheme by switching feature partitioning is proposed, which considerably relaxes the negative impact of parameter coupling during parameter extraction. The IGBT inductive turn-off process and static characteristics are divided into nine typical feature segments under an inductive load. Then, the physical principle of each feature segment is analyzed, and the corresponding influencing parameters of each feature segment are found. During the parameter extraction, the corresponding weights are assigned to the parameter search according to the error of each feature segment. The proposed method for parameter extraction is elaborated. Using the extracted parameters, the simulation results by the physics-based IGBT model are in excellent agreement with the experimental results of the tested IGBT, and the effectiveness of the proposed parameter extraction program is verified. Finally, the extracted parameters are evaluated and verified by global sensitivity analysis.

IGBT reliability analysis of photovoltaic inverter with reactive power output capability

When the PV power supply participates in reactive power regulation of distribution network, its output reactive power will affect the reliability of IGBT in the PV inverter. Aiming at this problem, this paper first qualitatively analyzed the influence of photovoltaic power supply participating in reactive power regulation of distribution network on the reliability of photovoltaic power supply. Then, a quantitative evaluation method of IGBT reliability based on data-driven was proposed. This method uses LightGBM machine learning model to replace the traditional thermoelectric coupling model, which effectively improves the calculation efficiency of IGBT junction temperature and reduces the dependence of IGBT reliability evaluation results on IGBT model parameters. Through this method, the reliability of core power electronic devices in photovoltaic inverters is quantitatively evaluated according to active power, reactive power, solar irradiance and ambient temperature. Finally, based on the IEEE 33 node distribution system, the reliability of IGBT in PV inverters participating in reactive power regulation of the distribution network was evaluated.

Numerical modeling of FS-trench IGBTs by TCAD and its parameter extraction method

The electrical behavior of the IGBTs is the key factors to the application of IGBTs. Technology Computer Aided Design (TCAD) is one of the most effective methods for characterizing of electrical behavior of IGBT. However, the great difficulties in extracting the model's parameters hinder its widely applications. In the present work, a high-precision TCAD model for FS-Trench IGBT was established and verified by experiments. Firstly, the parameters necessary for TCAD models were analyzed in theoretical and experimental. The key process parameters need to be determined in a value range table. A Convolutional Neural Network (CNN) was proposed and trained to form a relationship from characteristic curves to key process parameters. Subsequently, the numerical model's parameters were established by combining the trained CNN model with experimental data. Finally, the TCAD model was established and the simulations of IGBTs were carried out and verified by experiments. Results demonstrate that the proposed TCAD model is well consistent with experiments with a Root Mean Square Error (RMSE) of about 3.7 %. This work may provide a feasible method to obtain highly precise numerical models of power semiconductor devices. It may also promote the application of Artificial Intelligence in the device modeling and optimization design.

Multiple Points Measurement-Based Junction Temperature Estimation of IGBT Module

The junction temperature ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_j</i> ) estimation of IGBTs is critical for improving their reliability. To obtain <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_j</i> , the existing reference point temperature (RPT) measurement based methods require the power loss information, which is difficult to obtain accurately in the real-time applications. Also, these methods utilise the thermal model of IGBT which changes with ageing, resulting in an error in <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_j</i> estimation. In this paper, a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_j</i> estimation technique is proposed which utilises the temperature of multiple leads/terminals of an IGBT module. Unlike the existing methods, the proposed technique does not require the information of power losses in the power semiconductor chips (PSCs). Instead, the power losses in the PSCs are additional outcome of the proposed technique. Also, even for a degraded IGBT module, the proposed technique provides an accurate <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_j</i> estimation. The proposed technique also has the capability of estimating <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T_j</i> of multiple PSCs in an IGBT module simultaneously. A mathematical analysis of leads selection for the proposed technique is presented in this paper. Furthermore, a comparative analysis is presented between the proposed and the conventional method for a degraded IGBT module. The simulation studies of the proposed technique are carried out in ANSYS ICEPAK software. The proposed technique is also validated on an experimental setup developed in the laboratory.

Gate modulation drive and modeling of IGBT in MMC converters

The energy surplus caused by the alternating current (AC) system’s failure results in a severe overvoltage problem in Bai hetan-Jiangsu ±800 kV hybrid cascade direct current (DC) project. However, the insulated gate bipolar transistor’s (IGBT) off-state voltage is the crucial reason for the insufficient voltage capacity of modular multilevel converter (MMC). Therefore, analysis and suppression of the off-state overvoltage are of vital significance. The off-state model of IGBT in an MMC converter is established, and the process of IGBT’s turning off is analyzed. Then, based on the characteristics of IGBT’s off-state and traditional resistance switching strategy, the gate modulation drive strategy is proposed, and the equivalent topology of the MMC submodule with the strategy is established. Finally, MMC submodule models with gate modulation drive and resistance switching strategy, respectively, are built in simulation software, and two groups of data are compared under multiple groups of the control cycle and reference current slope given. The simulation results show that the gate modulation drive supresses the off-state overvoltage on IGBT better than the resistance switching strategy. Moreover, the shorter the control cycle and the less reference current slope, the better the overvoltage suppression.

Thermal cross-coupling effect analysis of multi-chips IGBT models considering chip size

With the increasing demand of high capacity of power converter, multi-chips IGBT modules are widely applicated. Taking thermal cross-coupling (TCC) effect among IGBT chips into account, the uneven temperature distribution should be investigated in-depth. Thus, in this paper, a steady-state finite-element model (FEM) of a conventionally utilized IGBT module in thermal domain is developed to analyse the effective heat transfer area from the chip to baseplate of the IGBT module. Moreover, the TCC coefficient is proposed in the light of effective transfer area to quantify the TCC effect. The influence of power loss distribution, heat dissipation ability and chip size on the TCC effect is discussed in the end. The results show larger power loss on neighbouring chips and smaller equivalent heat transfer coefficient (EHTC) cause larger TCC coefficient. Chip size has negligible influence on TCC effect at low EHTC, while larger chips size leads to severer TCC effect at high EHTC.

Comparison and Optimization of Datasheet-Driven Extraction of Gate-Drain Overlap Oxide Capacitance in IGBT Modeling

Extracting gate-drain overlap oxide capacitance ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">oxd</sub> ) is a necessary step for almost all kinds of insulated-gate bipolar transistor (IGBT) compact models. There are two datasheet-driven methods that are universally accepted, one is based on reverse capacitance voltage characteristics ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C–V</i> method), and the other uses gate charge characteristics ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Q–V</i> method). This letter is to reveal the underlying mechanism of these two methods. It is found that the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C–V</i> method has an inescapable shortcoming which will mistake the turn- <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">off</small> delay time forecasted by IGBT models. In this letter, the internal capacitance system of IGBT is modeled and demonstrated completely. Based on this model, the accuracy and feasibility of the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C–V</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Q–V</i> methods are analyzed. Moreover, the effect of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">oxd</sub> on IGBT compact models is studied and further verified by experiments and circuit simulations.

A Reliable Device Parameter Extraction Scheme for Physics-Based IGBT Models

The physics-based IGBT models have been proved to be very effective in accurately simulating the voltage, current, and carrier dynamics. However, the biggest challenge for those promising models is lack of detailed IGBT device parameters, which chip manufacturers generally seldom reveal. To extract precise and robust device parameters, a novel parameter extraction scheme based on the particle swarm optimization (PSO) algorithm is proposed in this article. The state-of-the-art Fourier-series-based (FSB) IGBT model is used to ensure the accuracy of the parameter extraction scheme. In the proposed scheme, IGBT parameters are simultaneously extracted by IGBT static and dynamic characteristic curves to account for global interaction between all those IGBT parameters during parameter optimization. The experiment is carried out to verify the proposed parameter extraction scheme. It should be noteworthy that only an experimental switching waveform under an arbitrary operating condition is required to implement global searching for the optimal parameters by our proposed scheme. The output characteristic curves and the switching waveforms of the FSB model under different operating conditions using the proposed parameter extraction scheme are in good agreement with the experimental waveforms. Results confirm the effectiveness and usefulness of the obtained device parameters by the randomly selected operating condition. Finally, the global sensitivity analysis is introduced to further demonstrate the reliability of the proposed parameter extraction scheme.

Physical Transient Model of IGBT in Forward Conduction Mode

Physical Transient Model of IGBT in Forward Conduction Mode

A SiC IGBT Model With Accurate Static and Dynamic Tracking Capability

A SiC IGBT Model With Accurate Static and Dynamic Tracking Capability

A Fully Coupled Model of Multi-Chip Press-Pack IGBT for Thermo-Mechanical Stress Distribution Prediction

In this article, a fully coupled multiphysical model of multichip press-pack insulated gate bipolar transistor (PPI) considering surface roughness is proposed. This model realizes the coupling of the electrical-, thermal- and mechanical domains through contact resistance. The simultaneous prediction of temperature and stress under various surface roughness is achieved. First, the fully coupled model and the contact resistance model are established. Second, both thermomechanical stress under the clamping phase and heating phase are analyzed. The influence of the surface roughness on the thermo-mechanical stress distribution is also investigated based on the proposed model. Results show that the surface roughness does not affect the thermomechanical stress distribution under the heating phase, while the amplitudes increase with surface roughness. Third, the correctness of the model is verified by the temperature measurement and deterioration observation results. The observed fretting scratches on the additional metallization area after the power cycling test validates the correctness of the deformation prediction. Finally, the strain-life fatigue model of PPI is established, it turns out that the surface roughness has an impact on the fatigue life of the chip metallization area.

Optimization of DC-Link Capacitance Based on the Power Loss Modeling of IGBT for Inverters Under Swing Bus Operation

The power density and reliability of the inverter can be improved with reduced DC-link capacitance by swing bus operation, etc. With insufficient DC-link capacitance, a double line frequency voltage ripple will be superimposed on the DC link, but how this voltage ripple affects the power loss of the IGBT in the inverter system has been seldom discussed. This paper analyzes the influence of voltage fluctuations on the power loss of IGBT and establishes the characteristic database of IGBTs under different operating voltages and currents using a dynamic test chamber. A switching profile-based power loss model has been proposed to evaluate the power loss of IGBT in the single-phase inverter under different DC link voltage ripples. All the theoretical analysis has been verified by a 6-kW inverter platform under swing bus operation (SBO).

Fault prediagnosis of power electronic devices in urban new energy system

The new energy system with low environmental load will more and more reflect its superiority in urban design and planning. The state monitoring of power electronic devices used for power conversion in new energy system is of great significance to the stable operation of the whole city. The failure mechanism and aging evolution process of IGBT modules are studied, and the failure types of IGBT devices are judged. An algorithm based on the thermal sensitive electrical parameter method and thermal resistance network method was proposed to identify aging types. Firstly, a healthy junction temperature model of IGBT based on the initial loss of turn-off was established in the double-pulse platform based on experimental data. Secondly, a thermal resistance network model of IGBT was established based on the datasheet. Finally, the accelerated aging of IGBT was carried out through the existing accelerated aging experimental platform of IGBT in the laboratory, and the aging parameters were imported into the above two models. The results show that the output results of the thermal resistance network model and the thermal parameter model are affected by the aging type, which provides a new solution for the aging type research of IGBT, and is of great value for the application of new energy system.

An IGBT junction temperature evaluation model for DC interruption application

IGBT is the core component of the hybrid direct current circuit breakers (DCCB), whose junction temperature is the key indicator for evaluating the ability of devices to withstand short-circuit current surges and ultimate breaking capacity. Thus, it is important to evaluate the junction temperature of IGBT under DC interruption application. At present, the main method for evaluating IGBT temperature is combining the finite element method (FEM) simulation and Cauer Thermal Network (CTN) model, which needs a large calculation amount and extracting IGBT physical structure parameters with complicated steps. In this paper, the IGBT model for DC interrupting is built, and a junction temperature prediction model based on a fast Forster-Cauer conversion algorithm is proposed. The IGBT power loss during DC interruption is taken as input, combined with junction temperature evaluation model to realize the rapid acquisition of temperature distribution between the physical layers of IGBT. The © results are shown to verify the feasibility of IGBT junction temperature evaluation model. It’s a kind of simple and fast simulation model without obtaining the detailed physical parameters in advance and suitable for temperature distribution research of IGBT core layers for turn-off transients under the large current surges.

A New Method to Measure the Parasitic Parameter Model of IGBT on Bias Voltage

With the increase in frequency, IGBT can generate serious electromagnetic interference (EMI) when it is turned on and off quickly. The variations in voltage and current produced by switching devices in the transient process are the main sources of high-frequency EMI. The ultrafast switching characteristics of IGBT (i.e., high du/dt or di/dt) will cause serious switching oscillation, EMI noise, and additional power loss. The speed of voltage and current variations in transient process is determined by the interelectrode capacitances of the semiconductor switch. Due to the structural characteristics of IGBT, the interelectrode capacitances will change based on the difference in external voltage. This article introduces an improved method to measure the parasitic inductance and resistance of IGBT. This method perfectly avoids the interference of ${Z}_{\text {ground}}$ which refers to the floating terminal impedance to the ground. The parasitic parameters of IGBT can be obtained by fitting the impedance of IGBT with genetic algorithm. The article also introduces a method to measure the parasitic parameters of IGBT on bias voltage with impedance analyzer and dc source. In the experiment, the capacitance is used to protect the impedance analyzer to avoid the dc current from the dc source. Finally, a double-pulse test circuit is used to verify the parasitic parameter model of IGBT measured in this article. By comparing the current and voltage waveforms when IGBT is turned on and off and the voltage frequency domain waveforms in multiple cycles, it is proved that the parasitic parameter model of IGBT is applicable to the circuit.

Predictive Model for Conducting Electromagnetic Interference by Bidirectional Excitation Controller

Bidirectional excitation controller is used in the excitation system of brushed DC motor. There are many monitoring sensors and weak current switches nearby. Therefore, it is necessary to study the conduction interference of the excitation controller. Firstly, based on the working principle of bidirectional excitation controller, the propagation path model and corresponding equivalent circuit of bidirectional excitation controller are established. Then, the parasitic capacitance parameters between the switch tube and the heat sink were extracted by ANSYS Q3D software, and the dynamic model of IGBT was established by using ANSYS Simplorer software. Based on ANSYS software, the prediction model of the equipment conducted electromagnetic interference was obtained. Finally an excitation controller conducting interference test platform was built, and the predicted results were compared with the measured interference results of the experimental platform to verify the accuracy of the prediction model.