Fast Simulation Approach Research Articles

A highly efficient computational approach for fast scan-resolved microstructure predictions in metal additive manufacturing on the scale of real parts

In metal additive manufacturing (AM), fast and efficient simulation approaches are essential to explore the full potential of these promising processes, particularly in generating components with tailored microstructures via laser powder bed fusion (LPBF). Due to the inherent multiscale nature of LPBF, existing approaches often need to resort to strong simplifications, such as layer-wise heating models, to make part-scale simulations feasible. In contrast, the present article proposes a scan-resolved approach, which consistently resolves the laser scan path in a coupled thermo-microstructural model of LPBF. Building on a high-performance computing model for the thermal problem, we propose a highly efficient implementation of a recently developed microstructure model for Ti-6Al-4V with three main constituents: stable αs-phase, martensitic αm-phase and β-phase. The implementation is tailored to modern hardware features using vectorization and fast approximations of transcendental functions. A performance model and selected numerical examples of LPBF manufacturing of parts on the centimeter scale are studied to verify the high degree of optimization. Depending on the specific example, results were obtained with moderate computational resources in a few hours to days. We demonstrate how the proposed scan-resolved model allows us to predict the correlation between scan strategy and resulting microstructure composition, an aspect that layer-wise heating models cannot capture. The numerical examples include scan-resolved thermo-microstructure simulations of the full NIST AM Benchmark cantilever specimen. It is shown that varying the build plate temperature by only 100K can significantly change the microstructure composition from αm- to αs-dominated.

A Fast Simulation Approach to Switched Systems

A Fast Simulation Approach to Switched Systems

A fast hybrid simulation approach of ion energy and angular distributions in biased inductively coupled Ar plasmas

In this work, a two-dimensional hybrid model, which consists of a bulk fluid module, a sheath module and an ion Monte-Carlo module, is developed to investigate the modulation of ion energy and angular distributions at different radial positions in a biased argon inductively coupled plasma. The results indicate that when the bias voltage amplitude increases or the bias frequency decreases, the ion energy peak separation width becomes wider. Besides, the widths of the ion energy peaks at the edge of the substrate are smaller than those at the center due to the lower plasma density there, indicating the nonuniformity of the ion energy distribution function (IEDF) along the radial direction. As the pressure increases from 1 to 10 Pa, the discrepancy of the IEDFs at different radial positions becomes more obvious, i.e. the IEDF at the radial edge is characterized by multiple low energy peaks. When a dual frequency bias source is applied, the IEDF exhibits three or four peaks, and it could be modulated efficiently by the relative phase between the two bias frequencies. The results obtained in this work could help to improve the radial uniformity of the IEDF and thus the etching process.

Fast and Efficient Simulation of the FEBID Process with Thermal Effects

Focused electron-beam-induced deposition (FEBID) is a highly versatile direct-write approach with particular strengths in the 3D nanofabrication of functional materials. Despite its apparent similarity to other 3D printing approaches, non-local effects related to precursor depletion, electron scattering and sample heating during the 3D growth process complicate the shape-true transfer from a target 3D model to the actual deposit. Here, we describe an efficient and fast numerical approach to simulate the growth process, which allows for a systematic study of the influence of the most important growth parameters on the resulting shape of the 3D structures. The precursor parameter set derived in this work for the precursor Me3PtCpMe enables a detailed replication of the experimentally fabricated nanostructure, taking beam-induced heating into account. The modular character of the simulation approach allows for additional future performance increases using parallelization or drawing on the use of graphics cards. Ultimately, beam-control pattern generation for 3D FEBID will profit from being routinely combined with this fast simulation approach for optimized shape transfer.

An approach for fast cascading failure simulation in dynamic models of power systems

The ground truth for cascading failure in power system can only be obtained through a detailed dynamic model involving nonlinear differential and algebraic equations whose solution process is computationally expensive. This has prohibited adoption of such models for cascading failure simulation. To solve this, we propose a fast cascading failure simulation approach based on implicit Backward Euler method (BEM) with stiff decay property. Unfortunately, BEM suffers from hyperstability issue in case of oscillatory instability and converges to the unstable equilibrium. We propose a predictor–corrector approach to fully address the hyperstability issue in BEM. The predictor identifies oscillatory instability based on eigendecomposition of the system matrix at the post-disturbance unstable equilibrium obtained as a byproduct of BEM. The corrector uses right eigenvectors to identify the group of machines participating in the unstable mode. This helps in applying appropriate protection schemes as in ground truth. We use Trapezoidal method (TM)-based simulation as the benchmark to validate the results of the proposed approach on the IEEE 118-bus network, 2383-bus Polish grid, and IEEE 68-bus system. The proposed approach is able to track the cascade path and replicate the end results of TM-based simulation with very high accuracy while reducing the average simulation time by ≈10−35 fold. The proposed approach was also compared with the partitioned method, which led to similar conclusions.

A New Analytical Approach for Fast and Accurate Dynamic Simulation of MMC-HVDC System

The Speed of execution along with acceptable accuracy of the simulation is crucial for Modular Multilevel Converter based High Voltage Direct Current (MMC-HVDC) systems in offshore applications. The high number of power electronic components in the MMC-HVDC system poses a challenge to solve them using the existing methods, based on average models, with required speed and accuracy. This paper presents a method to simulate all the submodules of the MMC-HVDC system, which enables to study of the dynamics in the subsea cable, submodules, and at the system level. MMC-HVDC system in a standard three-phase CIGRE benchmark configuration is modeled to validate the analytical method. It is compared with PSCAD simulations (based on Dommel’s method) and validated both in steady state and in transient conditions of the MMC-HVDC system. The results indicate that the analytical method is accurate, and the execution time is lower compared to PSCAD based on Dommel’s method.

Residual stresses in deep-drawn cups made of duplex stainless steel X2CrNiN23-4

Residual stress development in deep drawing processes is investigated based on cylindrical cups made of duplex stainless steel sheet. Using a two-scale approach combining finite element modelling with a mean field homogenization scheme the macro residual stresses as well as the phase-specific micro residual stresses regarding the phases ferrite and austenite are calculated for steel X2CrNiN23‑4 for various drawing depths. The simulation approach allows for the numerical efficient prediction of the macro and phase-specific micro residual stress in every integration point of the entire component. The simulation results are validated by means of X‑ray diffraction residual stress analysis applied to a deep-drawn cup manufactured using corresponding process parameters. The results clearly indicate that the fast simulation approach is well suited for the numerical prediction of residual stresses induced by deep drawing for the two-phase duplex steel; the numerical results are in good agreement with the experimental data. Regarding the investigated process, a significant influence of the drawing depth, in particular on the evolution of the residual stress distribution in drawing direction, is observed. Considering the appropriate phase-specific strain hardening, the two-scale approach is also well suited for the prediction of phase specific residual stresses on the component level.

A fast simulation approach to assess draft gear loads for heavy haul trains during braking

Train traction and brake dynamics simulations can be used to improve the safety and efficiency of heavy haul transportation. Virtual models can be used to analyze many operational conditions that are expensive or inviable to be measured due to costs and security risks. This paper presents a low computational cost simulation model that encompasses the longitudinal dynamics equations and the pneumatic brake system behavior. The model is used to evaluate the effects of the locomotive traction/dynamic-brake and of the pneumatic brake delays in the draft gear forces. A real-world scenario was analyzed, in which a standard train control procedure was applied for 168 ore freight cars propelled by two locomotives. For this analysis, the model was able to perform up to 8 simultaneous parallel simulations with a processing time of 607 s to conclude a route of 2884 s, which is very suitable for this kind of investigation.

GeantV

Full detector simulation was among the largest CPU consumers in all CERN experiment software stacks for the first two runs of the Large Hadron Collider. In the early 2010s, it was projected that simulation demands would scale linearly with increasing luminosity, with only partial compensation from increasing computing resources. The extension of fast simulation approaches to cover more use cases that represent a larger fraction of the simulation budget is only part of the solution, because of intrinsic precision limitations. The remainder corresponds to speeding up the simulation software by several factors, which is not achievable by just applying simple optimizations to the current code base. In this context, the GeantV R&D project was launched, aiming to redesign the legacy particle transport code in order to benefit from features of fine-grained parallelism, including vectorization and increased locality of both instruction and data. This paper provides an extensive presentation of the results and achievements of this R&D project, as well as the conclusions and lessons learned from the beta version prototype.

Performance analysis of three iteration-free numerical methods for fast and accurate simulation of thermal dynamics in district heating pipeline

Numerical methods can provide sufficient information for thermal dynamics of district heating (DH) network. However, high computation burden restricts its application, especially for large-scale DH networks. This dilemma can be overcome by establishing fast numerical calculation procedure and properly selecting the calculation steps without reduction of numerical accuracy. The first-order implicit upwind method is an iteration-free numerical approach for fast simulation of DH pipeline thermal dynamics, while the first-order precision restricts its numerical performance. To further improve the simulation performance, two iteration-free numerical methods with high-order precision, the second-order implicit upwind method and third-order semi-implicit QUICK method are developed. Validations of the three numerical methods are conducted with the measured data of a real DH pipeline. Preferred calculation steps of the three methods are studied via comprehensive numerical experiments. Based on the preferred step size, simulation comparisons of the numerical methods are performed. Results show that the outlet temperature fluctuations of DH pipeline can be feasibly predicted by these methods with satisfying accuracy. The second-order implicit upwind method perform best considering least computation burden (0.003 s) within acceptable error. With the application of the second-order implicit upwind method, simulation analyses on DH pipeline under varying flow velocities are performed.

Parameter Identification through Physics-Based Impedance Models Using Impedance Spectra for Lithium-Ion Batteries

Electrochemical impedance spectroscopy (EIS) is a useful characterization tool due to its ability to separate physicochemical processes by their intrinsic time constants. EIS is extensively used to study diffusion1 and kinetic2 processes, investigate passive layer growth3, and characterize aging4 in lithium-ion batteries. However, experimental EIS data is traditionally fit using equivalent circuit models with circuit elements that are representative of the physical processes happening inside a battery. Such models may lose physical meaning and interpretability as processes and parameters are lumped together.Physics-based impedance models mathematically describe physicochemical processes based on fundamentally derived equations and are thus better alternatives for providing direct insight than their equivalent circuit analogs5. These models are often highly coupled and complex, and traditional solution methods make it difficult to solve them in a reasonable time frame suited for multi-parameter identification. Recent work on a hybrid analytical-collocation approach demonstrated significant reduction in computation time (two orders of magnitude improvement compared to commercial software), making multi-parameter estimation a more feasible optimization problem6.Estimation of parameters is an important step to using physics-based models in battery management systems (BMS) to monitor, predict, and control batteries, since such physics-based models inherently require knowledge of specific battery parameters. This work builds on the previously developed fast simulation approach for physics-based impedance models for a full cell consisting of two porous electrodes, by conducting a parameter reduction study through clustering of physical parameters. We present a method to demonstrate the identification of parameters and explore the ability to estimate parameters for the impedance spectra at different states of charge of a cell and at various states of aging.

A Fast Simulation Approach to the Thermal Recovery Characteristics of Deep Borehole Heat Exchanger after Heat Extraction

Necessary intermittence after heat extraction for a deep borehole heat exchanger (DBHE) is beneficial for sustainable operation. This paper centers on the fast simulation for thermal recovery characteristics of DBHE under intermittent condition. First of all, in view of the existing temperature gradient and multi-layer heterogeneity of rock underground that could never be ignored for DBHE, we extend the classical finite line source model based on heat source theory and superposition principle to account for the vertical heat flux distribution varying along depth and heterogeneous thermal conductivities in the multi-layer rock zone. Moreover, a fast simulation approach for heat transfer analysis inside the borehole coupled with the extended finite line source model is put forward to depict the transient thermal response and dynamic thermal recovery of rock outside borehole. To the authors’ knowledge, no such algorithm for deep BHE has yet been suggested in the previous literature. This approach has proven to be reliable and efficient enough for DBHE simulation under the intermittent condition. Simulation results show that at least 65 days of intermittence for the model in study should be spared after the heating season to achieve sustainable heat extraction in the next cyclic operation. Compared to the detailed solution based on full discretization numerical schemes, the relative error for borehole bottom temperature was 0.79%. In addition, comparison of the simulation results for thermal performance during the heating season in a three-year cyclic operation with 205 days intermittence shows that both the outflow temperature and heat extraction rate in the subsequent cycle after intermittence are in good agreement with the full 3D numerical solution in the reference (with a relative error of 6.36% for the outflow temperature and 9.3% for the heat extraction rate). Regarding the calculation speed, around a 13 times acceleration can be achieved. Finally, it is also promising to be applicable for thermal recovery simulation after heat extraction of vertical closed loop borehole heat exchangers at arbitrary length from shallow to deep.

GPU-accelerated transient lattice Boltzmann simulation of bubble column reactors

Design parameters for Bubble Column Reactors such as backmixing and light/dark cycles of microalgae cells are pertinent to transient flow dynamics and mesoscale coherent structures. Transient simulation is therefore of practical significance. Traditional CFD solvers are generally implemented on multi-core CPU workstations, which is time-consuming and difficult to meet the requirement of industry applications. We developed a GPU-accelerated transient LBM simulation of gas-liquid flow, achieving 5000 times acceleration in comparison with the CFD simulation. A notable feature of this method, which allows a tunable and smaller dimensionless relaxation time, is the remarkable decrease of required mesh number in LBM simulation by several orders of magnitude. The solver is validated against experimental data, resolving two mechanisms of mesoscale structure, i.e., vortex rotation and radial migration, in the intensification of macromixing. The computation acceleration and resolution of mesoscale structures demonstrate the potential of this approach for fast simulation of industrial applications.

Voltage and Frequency Recovery in an Islanded Inverter-Based Microgrid Considering Load Type and Power Factor

This paper introduces simple and complete under voltage-frequency load shedding schemes to determine proper load shedding amounts (LSAs) in an islanded inverter-based microgrids (IBMGs) considering power and power factor (PF) of various load types, such as constant power, constant current, and constant impedance loads. The proposed method employs a fast time-step simulation approach based on the complete state-space model of IBMG in order to obtain the transient and steady state responses of the system. The required power-voltage curves are obtained considering load PF and system topology. Ultimately, LSA is obtained according to the under voltage load shedding (LS) and under frequency LS values. To demonstrate the effectiveness of the proposed simple and complete LS methods, the corresponding performances are analyzed in a test islanded IBMG modeled in MATLAB. The simulation results show that unlike the other compared methods, the proposed LS approach can accurately determine less LSA while maintaining the power system stability.

Simple and fast simulation approach to investigate the NBTI effect on suspended gate MOS devices

ABSTRACT In this paper, we investigate the negative bias temperature instability (NBTI) on conventional P-type metal-oxide-semiconductor field effect transistors (PMOSFET) using on-fly bulk trap technique (OTFBT). The extracted NBTI induced interface (ΔNit ) and oxide traps (ΔNot ), using OTFBT, are modelled and used to simulate the NBTI effect on N-type suspended gate metal-oxide-semiconductor devices (N-type SG-MOS), which could be manufactured by the same fabrication process as conventional PMOSFET. The used approach to simulate the NBTI effect is performed by combining, in the same simulation program, the N-type SG-MOS devices model with the NBTI induced ΔNit and ΔNot models. This approach allowed us to simulate and predict rapidly the lifetime of the N-type SG-MOS devices subjected to the NBTI degradation. The simulation shows that the degradation of N-type SG-MOS devices due to the NBTI is the same as that of conventional PMOSFET. However, the extracted lifetime of N-type SG-MOS devices (stiction of the suspended gate) is longer than that of conventional PMOSFET.

Modelling Approach for Fast Simulation and Optimization of Large District Heating Networks

Modelling Approach for Fast Simulation and Optimization of Large District Heating Networks

Direct equations to retention time calculation and fast simulation approach for simultaneous material selection and experimental design in comprehensive two dimensional gas chromatography

Column selection and experimental design are recognized to be wear-and-tear processes to obtain a good fingerprinting result with comprehensive two dimensional gas chromatography (GC × GC). The processes involve a large number of experiments for analysis of each sample due to the optimized conditions depending on the selected column sets. In this study, the analytical solutions of time summation model combined with temperature dependent linear solvation energy relationship (LSER) were derived for constant flow separation in GC × GC. The derived equations allow calculation of analyte retention time in first and second dimensional separation (1tR and 2tR), under temperature program separation employing any column combination with known LSER database. As a result, optimization software was developed enabling simulation of several hundred thousand GC × GC results (fingerprints) for separation of each sample. Good correlations between our predicted results and the results obtained with the previously established numerical approach, were obtained with the R2 of 1.000 and 0.998 for simulation of 1tR and 2tR, respectively. The developed approaches were further applied to simulation of 96,000 individual fingerprints in temperature programmed GC × GC, with the focus on application of 16 column sets including non-ionic liquid and ionic liquid (IL) stationary phases for separation of 678 model compounds. These approaches resulted in the computational time of 1 day compared with 1 year provided by the numerical method. Best column sets and experimental conditions (secondary column lengths and temperature programs) could then be extracted according to maximizing number of separated peaks in separation, which represents the quality of each fingerprinting.

Block Markov Superposition Transmission of BCH Codes With Iterative Erasures-and-Errors Decoders

In this paper, we present the block Markov superposition transmission of BCH (BMST-BCH) codes, which can be constructed to obtain a very low error floor. To reduce the implementation complexity, we design a low complexity iterative sliding-window decoding algorithm, in which only binary and/or erasure messages are processed and exchanged between processing units. The error floor can be predicted by the proposed genie-aided lower bounds, while the waterfall performance can be analyzed by the density evolution method. To evaluate the error floor of the constructed BMST-BCH codes at a very low bit error rate (BER) region, we propose a fast simulation approach. Numerical results show that, at a target BER of $10^{-15}$ , the proposed BMST-BCH code with hard-decision can achieve a net coding gain (NCG) of 10.55 dB with 25% overhead, while a soft-decision design can yield an NCG of 10.74 dB. The construction of BMST-BCH codes is flexible to trade off latency against performance at all overheads of interest and may find applications in optical transport networks as an attractive candidate.

3D convolutional GAN for fast simulation

Machine Learning techniques have been used in different applications by the HEP community: in this talk, we discuss the case of detector simulation. The need for simulated events, expected in the future for LHC experiments and their High Luminosity upgrades, is increasing dramatically and requires new fast simulation solutions. We describe an R&D activity aimed at providing a configurable tool capable of training a neural network to reproduce the detector response and speed-up standard Monte Carlo simulation. This represents a generic approach in the sense that such a network could be designed and trained to simulate any kind of detector and, eventually, the whole data processing chain in order to get, directly in one step, the final reconstructed quantities, in just a small fraction of time. We present the first application of three-dimensional convolutional Generative Adversarial Networks to the simulation of high granularity electromagnetic calorimeters. We describe detailed validation studies comparing our results to Geant4 Monte Carlo simulation. Finally we show how this tool could be generalized to describe a whole class of calorimeters, opening the way to a generic machine learning based fast simulation approach.

Multi-level timing and fault simulation on GPUs

In CMOS technology first-order parametric faults during manufacturing can exhibit severe changes in the timing as well as in the functional behavior of cells. Since these faults are hard to detect by conventional tests, the accurate simulation of these low-level faults plays an important role for test validation. However, pure low-level fault simulation approaches impose a high computational complexity that can quickly become inapplicable to larger simulation problems due to limitations in scalability.In this paper, the first parallel multi-level fault simulation approach on graphics processing units (GPUs) is presented. The approach utilizes both logic level and switch level descriptions concurrently in a mixed-abstraction timing simulation. The abstraction is lowered in user-defined so-called regions of interest that locally increase the modeling accuracy enabling low-level first-order parametric fault injection. Resulting signal waveforms are transformed between the different abstractions transparently. This way a fast, versatile and efficient multi-level fault simulation approach on GPUs is created that scales for designs with millions of cells while achieving high simulation throughput with runtime savings of up to 84% compared to full switch level simulations.