Repetitive Simulation Research Articles

A dynamic importance function for accidental scenarios generation by RESTART in the computational risk assessment of cyber-physical infrastructures

The Computational Risk Assessment (CRA) of Cyber-Physical Systems (CPSs) calls for the analysis of accidental scenarios emerging from the complexities and interdependencies typical of CPSs. Generating these scenarios via crude Monte Carlo Simulation (MCS) is impractical due to the high computational demand of simulation codes of CPSs, considering the combinatorial number of possible scenarios. In this paper, we tailor the use of Repetitive Simulation Trials After Reaching Thresholds (RESTART), a rare-event simulation method of literature, to efficiently generate relevant accidental scenarios. The tailored RESTART is guided by a dynamic Importance Function (IF) originally introduced here to dynamically characterize the relevance of the scenarios with reference to the current topology of the CPS and the susceptibility of its components. Two case studies of increasing complexity are considered: a single power grid and a CPS consisting of an Integrated Power and Telecommunication (IP&TLC) infrastructure. Results show that RESTART mines out more relevant scenarios than crude MCS for a number of different IFs based on vulnerability metrics of literature, and thus particularly efficiently when the novel IF is adopted.

Enhancing subsurface multiphase flow simulation with Fourier neural operator

Accurate numerical modeling of multiphase flow in subsurface oil and gas reservoirs is critical for optimizing hydrocarbon recovery. However, traditional physics-based algorithms face substantial computational hurdles due to the need for fine grid resolution and the inherent geological heterogeneity. To overcome these challenges, data-driven surrogate models solving the flow governing partial differential equations (PDEs) offer a promising alternative to enhance the efficiency of hydrocarbon production operations. In this study, we employ the Fourier Neural Operator (FNO) to extract spectral information from the reservoir properties, thereby facilitating the solution of coupled porous flow PDEs. Our focus is on two-phase flow dynamics, specifically exploring how water injection enhances reservoir pressure and displaces oil. This scenario involves solving a set of nonlinearly coupled PDEs with highly heterogeneous coefficients. Numerical results demonstrate that the developed FNO accurately predicts the reservoir pressure distributions. We further observe that the FNO's zero-shot super-resolution capability is sensitive to abrupt local changes in the reservoir pressure near injection and production wells. To enhance its accuracy, we propose a multi-fidelity FNO model that exhibits better adaptability across various grid configurations. After moderate training on graphics processing units (GPUs), the FNO achieves a speedup of three orders of magnitude compared to traditional numerical PDE solvers. Our experiments confirm the FNO's potential to replace repetitive physics-based simulations, significantly advancing computational efficiency in the uncertainty quantification of reservoir performance.

An automated framework for material property calibration in loudspeaker simulation model

The shift to virtual meetings, online classes, and remote work has established a new norm, leading to a surge in the use of virtual communication platforms such as Zoom and Microsoft Teams. This shift has increased the demand for high-quality headsets and speakerphones, emphasizing the need for clear, superior audio quality. The process of calibrating material properties typically relies on repetitive simulations guided by experts' intuition, presenting challenges in establishing new Finite Element Models (FEMs) of loudspeakers, as it requires the repeated identification of material property values. We present a systematic framework for calibrating the mechanical material properties of loudspeaker drivers, a crucial prerequisite for developing accurate FEMs of loudspeakers. Specifically, we propose a statistically-driven approach to replace the conventional manual calibration process, which typically relies on multiple simulations guided by expert intuition. Efficient Global Optimization (EGO) is applied to address the expensive optimization problems of loudspeaker simulation. To tackle the curse of dimensionality, the objective function is decomposed into several functions based on effective parameter groups using Global Sensitivity Analysis (GSA) results. The parameters of the FEM are then calibrated to the reference data from the Lumped Parameter Model (LPM) using the decomposed-reduced objective function, providing the calibrated parameters for the loudspeaker simulation. By implementing this novel approach, even individuals without prior knowledge or experience in loudspeaker material properties can effectively and reliably obtain the necessary data for finite element modeling.

Groundwater LNAPL Contamination Source Identification Based on Stacking Ensemble Surrogate Model

Groundwater LNAPL (Light Non-Aqueous Phase Liquid) contamination source identification (GLCSI) is essential for effective remediation and risk assessment. Addressing the GLCSI problem often involves numerous repetitive forward simulations, which are computationally expensive and time-consuming. Establishing a surrogate model for the simulation model is an effective way to overcome this challenge. However, how to obtain high-quality samples for training the surrogate model and which method should be used to develop the surrogate model with higher accuracy remain important questions to explore. To this end, this paper innovatively adopted the quasi-Monte Carlo (QMC) method to sample from the prior space of unknown variables. Then, this paper established a variety of individual machine learning surrogate models, respectively, and screened three with higher training accuracy among them as the base-learning models (BLMs). The Stacking ensemble framework was utilized to integrate the three BLMs to establish the ensemble surrogate model for the groundwater LNAPL multiphase flow numerical simulation model. Finally, a hypothetical case of groundwater LNAPL contamination was designed. After evaluating the accuracy of the Stacking ensemble surrogate model, the differential evolution Markov chain (DE-MC) algorithm was applied to jointly identify information on groundwater LNAPL contamination source and key hydrogeological parameters. The results of this study demonstrated the following: (1) Employing the QMC method to sample from the prior space resulted in more uniformly distributed and representative samples, which improved the quality of the training data. (2) The developed Stacking ensemble surrogate model had a higher accuracy than any individual surrogate model, with an average R2 of 0.995, and reduced the computational burden by 99.56% compared to the inversion process based on the simulation model. (3) The application of the DE-MC algorithm effectively solved the GLCSI problem, and the mean relative error of the identification results of unknown variables was less than 5%.

A methodology for computationally generating phase space files for Monte Carlo simulations applied to treatment plans for medical linear accelerators

The application of Monte Carlo (MC) simulation in the context of Medical Linear Accelerator (LinAc) treatment planning has increased in recent decades owing to its accuracy in dose calculations. The Multileaf Collimator (MLC), a component of the LinAc heads, plays an essential role by allowing the precise customization of the treatment beam's shape, effectively targeting the tumor area while minimizing radiation exposure to surrounding healthy structures. During a jaws static treatment, this MLC configuration varies for each angular position of the accelerator gantry, while the rest of the LinAc geometry remains static. To address the operation of the accelerator during the treatment planning avoiding the need of computationally intensive and repetitive simulations, phase-space files (PSF) are widely employed in LinAc MC calculations. This study aims to introduce and validate a methodology for the computational generation of PSFs specifically defined just below the MLC. These PSFs will assume the appropriate shape corresponding to the angle at which the accelerator gantry is positioned during treatment. The core of this methodology involves the creation of a comprehensive database containing precalculated probability functions derived from MC. In order to validate the methodology, simulated dose values of a LinAc emitting a 6 MV photon beam have been employed. The results have been tested against experimental data, and compared with the results from complete simulations, yielding PSFs of varying shapes. Finally, this study reveals that the results obtained by comparing organ dose values computed from the generated PSFs to those from the reference PSF are statistically compatible. This innovative method not only offers a valuable tool for optimizing LinAc treatment planning but also underscores the importance of accurate dose calculations in the field of radiation therapy.

Mechanical Quantities Prediction of Metal Cutting by Machine Learning and Simulation Data

Metal cutting is an important process in industrial manufacturing. Using the mechanical quantities of metal cutting to optimize process design is helpful to improve productivity. However, it is expensive to obtain these quantities due to the complexity of the cutting process, including material nonlinearity, geometric nonlinearity, state nonlinearity and their interactions. In this paper, a prediction model is constructed by combining machine learning (ML) and simulation data to quickly acquire multi-difficult-to-obtain metal cutting mechanical quantities to solve this problem. First, Adaptive Smoothed Particle Hydrodynamics (ASPH) is used to generate a simulation dataset of 2000 metal cutting cases. Based on the simulation data, six machine learning (ML) methods are employed to establish two prediction models, single-task learning and multi-task learning, to predict the mechanical quantities of metal cutting. The experimental results demonstrate that the ML method can predict abundant reference data efficiently after understanding the relationship between simulation parameters and mechanical quantities from simulation data, which is expected to replace some similar and repetitive simulation work. The Multilayer Perceptron (MLP) model under the multi-task setting provides the best prediction performance, fastest prediction time efficiency, and stable model behavior. Additionally, input erasure experiments reveal that the prediction of maximum equivalent plastic strain is significantly affected by particle spacing, and cutting speed plays a vital role in predicting maximum velocity. This work highlights the promotion of the data-driven ML method in quickly obtaining abundant reference data for the metal cutting process, and provides an auxiliary means for process optimization.

Predicting multiphase flow behavior of methane in shallow unconfined aquifers using conditional deep convolutional generative adversarial network

Numerical modeling is an essential tool for geoscience applications involving multiphase flow behavior. Performing numerical simulation is, however, computationally intensive due to multi-scale, non-linear, and multi-physics nature of mechanisms controlling multiphase flow dynamics. Additionally, the computational challenges associated with traditional numerical simulations limit their applicability for repetitive simulations required in inverse modeling, uncertainty analysis, and optimization tasks. To overcome these limitations, surrogate modeling has recently emerged as a viable solution. A surrogate model, which functions as a regression model, is trained using numerical simulation data. It approximates the input–output relationships within a dynamic fluid environment. We design surrogate models, based on conditional deep convolutional generative adversarial network (cDC-GAN), to map the cross-domain between input and output pairs in a multiphase multicomponent system, focusing on methane (CH4) migration in shallow unconfined aquifers. The cDC-GAN technique is selected due to its ability to generate high-fidelity surrogate models that effectively capture complex spatial distributions and heterogeneities. This technique excels in handling high-dimensional data, learning the intricate relationships between inputs and outputs, and producing realistic representations of subsurface phenomena. The trained cDC-GANs consider capillary entry pressure heterogeneity as input because this geologic attribute places first-order control on CH4 migration in subsurface. The outputs include CH4 saturation, water saturation, residual saturation of CH4, CH4 mole fraction, and CH4 molality. For each output, a separate cDC-GAN model is trained. Once trained, cDC-GANs can at any given time determine CH4 distribution in a shallow unconfined aquifer, both qualitatively and quantitatively. The cDC-GAN approach can be considered as a valuable complement to numerical simulations for predicting multiphase flow behavior in other geoscience-, energy-, and environmental applications such as contaminant transport, hydrocarbon production, and geological carbon dioxide and hydrogen storage.

Accelerating regional-scale groundwater flow simulations with a hybrid deep neural network model incorporating mixed input types: A case study of the northeast Qatar aquifer

ABSTRACT This study presents the ‘Dual Path CNN-MLP’, a novel hybrid deep neural network (DNN) architecture that merges the strengths of convolutional neural networks (CNNs) and multilayer perceptrons (MLPs) for regional groundwater flow simulations. This model stands out from previous DNN approaches by managing mixed input types, including both imagery and numerical vectors. Such flexibility allows the diverse nature of groundwater data to be efficiently utilized without the need to convert it into a uniform format, which often leads to oversimplification or unnecessary expansion of the dataset. When applied to the northeast Qatar aquifer, the model demonstrates high accuracy in simulating transient groundwater flow fields, benchmarked against the well-established MODFLOW model. The model's efficacy is confirmed through k-fold cross-validation, showing an error margin of less than 12% across all examined locations. The study also examines the model's ability to perform uncertainty analysis using Monte Carlo simulations, finding that it achieves around 1% average absolute percentage error in estimating the mean hydraulic head. Errors are mostly found in areas with significant variations in the hydraulic head. Switching to this ML model from the conventional MODFLOW simulator boosts computational efficiency by about 99%, showcasing its advantage for tasks like uncertainty analysis in repetitive groundwater simulations.

Can cerebellar theta-burst stimulation improve balance function and gait in stroke patients? A randomized controlled trial.

The cerebellum is a key structure involved in balance and motor control, and has become a new stimulation target in brain regulation technology. Interference theta-burst simulation (iTBS) is a novel simulation mode of repetitive transcranial magnetic simulation. However, the impact of cerebellar iTBS on balance function and gait in stroke patients is still unknown. The aim of this study was to determine whether cerebellar iTBS can improve function, particularly balance and gait, in patients with post-stroke hemiplegia. This study is a randomized, double-blind, sham controlled clinical trial. The study was carried out at the Department of Rehabilitation Medicine in a general hospital. Patients with stroke with first unilateral lesions were enrolled in the study. Thirty-six patients were randomly assigned to the cerebellar iTBS group or sham stimulation group. The cerebellar iTBS or pseudo stimulation site is the ipsilateral cerebellum on the paralyzed side, which is completed just before daily physical therapy. The study was conducted five times a week for two consecutive weeks. All patients were assessed before the intervention (T0) and at the end of 2 weeks of treatment (T1), respectively. The primary outcome was the Berg Balance Scale (BBS), while secondary outcome measures included the Fugl Meyer Lower Limb Assessment Scale (FMA-LE), timed up and go (TUG), Barthel Index (BI), and gait analysis. After 2 weeks of intervention, the BBS, FMA-LE, TUG, and BI score in both the iTBS group and the sham group were significantly improved compared to the baseline (all P<0.05). Also, there was a significant gait parameter improvement including the cadence, stride length, velocity, step length compared to the baseline (P<0.05) in the iTBS group, but only significant improvement in cadence was identified in the sham group (P<0.05). Intergroup comparison showed that the BBS (P<0.001), FMA-LE (P<0.001), and BI (P=0.002) in the iTBS group were significantly higher than those in the sham group, and the TUG in the iTBS was significantly lower than that in the sham group (P=0.002). In addition, there were significant differences in cadence (P=0.029), strip length (P=0.046), gain velocity (P=0.002), and step length of affected lower limb (P=0.024) between the iTBS group and the sham iTBS group. Physical therapy is able to improve the functional recovery in hemiplegic patients after stroke, but the cerebellar iTBS can facilitate and accelerate the recovery, particularly the balance function and gait. Cerebellar iTBS could be an efficient and facilitative treatment for patients with stroke. Cerebellar iTBS provides a convenient and efficient treatment modality for functional recovery of patients with stroke, especially balance function and gait.

Developing a cost-effective emulator for groundwater flow modeling using deep neural operators

Current groundwater models face significant challenges in their implementation due to heavy computational burdens. To overcome this, our work proposes a cost-effective emulator that efficiently and accurately forecasts the impact of abstraction in an aquifer. Our approach uses a deep neural operator (DeepONet) framework to learn operators that map between infinite-dimensional function spaces via deep neural networks. The goal is to infer the distribution of hydraulic heads in a confined aquifer in the presence of a pumping well. We successfully tested the DeepONet framework on multiple problems, including forward time-dependent problems, an inverse analysis, and a nonlinear system. Additionally, we propose a novel extension of the DeepONet-based architecture to generate accurate predictions for varied hydraulic conductivity fields and pumping well locations that are unseen during training. Our emulator’s predictions match the target data with excellent performance, demonstrating that the proposed model can act as an efficient and fast tool to support a range of tasks that require repetitive forward numerical simulations or inverse simulations of groundwater flow problems. Overall, our work provides a promising avenue for developing cost-effective and accurate groundwater models.

Preparing for practice, the effects of repeated immersive simulation on the knowledge and self-efficacy of undergraduate nursing students: A mixed methods study

AimThe aim of this study was to compare the effects of two immersive simulation-based education instructional designs, immersive simulation with team deliberate practice and immersive repeated standard simulation, when delivered over the same time on the knowledge and self-efficacy of nursing students. BackgroundImplementing immersive simulation-based education is not without its resource challenges, making it prohibitive for simulation educators to include it in their curricula. Subsequently, there is a need to identify instructional designs that meet these challenges. DesignA two-stage mixed methods approach was used to compare the two instructional designs. MethodsIn stage one, data were collected using questionnaires and differences estimated using analysis of covariance. In stage two, data were collected from two focus groups and analysed using a qualitative content analysis approach. Data were collected as part of a doctoral study completed in 2019 and was analysed for this study between 2022 and 2023. The justification for this study was that the identification of effective designs for immersive simulation remains a key research priority following the increase in allowable simulation hours by the Nursing and Midwifery Council. ResultsIn stage one, there was no statistical significance in the participant’s knowledge or self-efficacy between the models. In stage two, four themes were identified: vulnerability, development of knowledge, development of self-efficacy and preparation for placement. In contrast to stage one, participants reported that the repeated nature of both designs reinforced their knowledge base increased their self-efficacy, reduced their anxiety levels, and helped them to prepare for placement. ConclusionThe results inferred that both designs had a positive impact on the participants. Overall, participants reported that it helped them prepare for placements. Based on the findings, wherever possible, repeated immersive simulation-based education designs should be used and not a standalone immersive simulation-based education scenarios. If resources allow, this could be either a repeated scenario, or if there are resource constraints to use, over the same time, immersive simulation with team deliberate practice, or a similar model. Thus, giving a potential return on investment, one that supports simulation educators making those sensitive decisions regarding the inclusion of immersive simulation with team deliberate practice in their curriculum. Further research is needed into this area to ascertain the design features that maximise this impact and support a move away from standalone scenarios to an approach that uses repetitive immersive simulation.

Neuromodulation for Post-Stroke Motor Recovery: a Narrative Review of Invasive and Non‑Invasive Tools.

Stroke remains a leading disabling condition, and many survivors have permanent disability despite acute stroke treatment and subsequent standard-of-care rehabilitation therapies. Adjunctive neuromodulation is an emerging frontier in the field of stroke recovery. In this narrative review, we aim to highlight and summarize various neuromodulation techniques currently being investigated to enhance recovery and reduce impairment in patients with stroke. For motor recovery, repetitive transcranial magnetic simulation (rTMS) and direct current stimulation (tDCS) have shown promising results in many smaller-scale trials. Still, their efficacy has yet to be proven in large-scale pivotal trials. A promising large-scale study investigating higher dose tDCS combined with constraint movement therapy to enhance motor recovery is currently underway. MRI-guided tDCS studies in subacute and chronic post-stroke aphasia showed promising benefits for picture-naming recovery. rTMS, particularly inhibitory stimulation over the contralesional homolog, could represent a pathway forward in post-stroke motor recovery in the setting of a well-designed and adequately powered clinical trial. Recently evidenced-based guideline actually supported Level A (definite efficacy) for the use of low-frequency rTMS of the primary motor cortex for hand motor recovery in the post-acute stage of stroke based on the meta-analysis result. Adjunctive vagal nerve stimulation has recently received FDA approval to enhance upper limb motor recovery in chronic ischemic stroke with moderate impairment, and progress has been made to implement it in real-world practice. Despite a few small and large-scale studies in epidural stimulation (EDS), further research on the utilization of EDS in post-stroke recovery is needed. Deep brain stimulation or stent-based neuromodulation has yet to be further tested regarding safety and efficacy. Adjunctive neuromodulation to rehabilitation therapy is a promising avenue for promoting post-stroke recovery and decreasing the overall burden of disability. The pipeline for neuromodulation technology remains strong as they span from the preclinical stage to the post-market stage. We are optimistic to see that more neuromodulation tools will be available to stroke survivors in the not-to-distant future.

Evolving reliability assessment of systems using active learning-based surrogate modelling

Risk tracking involves monitoring and assessing evolving event risks in complex infrastructure systems. As these systems change over time, so do the associated risk profiles, requiring reliability updating principles. Traditional methods rely on repetitive simulations of complex models, while recent surrogate-based techniques offer efficiency and accuracy without relying on observed data. However, these approaches inadequately address time-dependent reliability updating and surrogate modeling integration. To overcome these limitations, this study presents the Time-Variant Reliability Updating method (TVRU), which seamlessly integrates Kriging-based time-dependent reliability with parallel learning. TVRU estimates prior failure probability by leveraging the adaptive training capabilities of a time-variant surrogate Kriging model. This model incorporates random variables and time as inputs, using the performance function's responses as outputs. By classifying design samples, the prior failure probability is accurately computed, and a meticulous Kriging model estimates the posterior probability of failure. The TVRU method is demonstrated using four numeric models of varying complexities, showcasing exceptional efficiency and accuracy. This novel approach enhances risk assessment in complex systems, facilitating informed decision-making.

Exploring the Role of Simulation Training in Improving Surgical Skills Among Residents: A Narrative Review.

The role of simulation in medical education is crucial to the development of surgeons' skills. Surgical simulation can be used to improve surgical skills in a secure and risk-free environment. Animal models, simulated patients, virtual reality, and mannequins are some types of surgical simulation. As a result, feedback encourages students to reflect on their strengths and weaknesses, enabling them to focus on improvement. Healthcare simulation is a strong educational instrument, and the main goal of this is to give the students an opportunity to do a practical application of what they have learned through theory. Before taking it to the patients, they will already have certain tools they have previously acquired during the practice. This makes it easier for students to identify the knowledge gaps that they must fill to improve patient outcomes. Moreover, simulation brings a wonderful opportunity for students to acquire skills, gain confidence, and experience success before working with real patients, especially when their clinical exposure is limited. The use of simulation to teach technical skills to surgical trainees has become more prevalent. The cost of setting up a simulation lab ranges from $100,000 to $300,000. There are several ways to evaluate the effectiveness of simulation-based surgical training. Repetitive surgical simulation training can improve speed and fluidity in general surgical skills in comparison to conventional training. Few previous studies compared learners who received structured simulation training to a group of trainees who did not receive any simulation training in single-center randomized control research. Significantly faster and less time-consuming skill proficiency was noticeable in simulated trainees. Despite being anxious in the operating room for the first time, simulated trainees completed the surgery on time, demonstrating the effectiveness of surgical simulation training. Traditional surgical training involves senior-surgeon supervision in the operating room. In simulation-based training, the trainees have full control over clinical scenarios and settings; however, guidance and assessment are also crucial. Simulators allow users to practice tasks under conditions resembling real-life scenarios. Simulators can be compared with traditional surgical training methods for different reasons. For example, intraoperative bleeding may occasionally show up not only visibly on the screen but also by shaking the trocars erratically. Without haptics, training on virtual simulators can cause one's pulling and pushing forces, which are frequently greater than what the tissue needs, to be distorted. A good method of simulation training is using virtual reality simulators with haptics and simulated patients. The availability of these facilities is limited, though, and a typical session might include an exercise involving stacking sugar cubes and box trainers. The degree of expertise or competency is one area that needs clarification as medical education transitions to a competency-based paradigm. The article aims to provide an overview of simulation, methods of simulation training, and the key role and importance of surgical simulation in improving skills in surgical residents.

Augmenting Deep Residual Surrogates with Fourier Neural Operators for Rapid Two-Phase Flow and Transport Simulations

Summary Accurate numerical modeling of multiphase flow and transport mechanisms is essential to study varied, complex physical phenomena including flow in subsurface oil and gas reservoirs and subsurface aquifers subject to CO2 sequestration. State-of-the-art complete physics-based solvers suffer from many computational challenges. High-fidelity data-driven surrogate models that solve the governing partial differential equations (PDEs) have the potential to optimize the time to solution and increase confidence in critical business and engineering decisions through better quantification of solution statistics. We leverage the recently proposed Fourier neural operators (FNOs) with quasilinear time complexity to capture the spectral information from feature maps to solve the coupled porous flow and transport PDEs. Embedding Fourier layers within the residual blocks results in a highly effective structure that, while achieving competitive accuracy, also enables efficient training of deeper networks with a dramatically reduced number of trainable parameters. The resulting novel deep-learning (DL) architecture is coined as FResNet++. FResNet++ uses squeeze and excitation blocks, atrous spatial pyramid pooling (ASPP), and attention blocks to increase its sensitivity to the relevant features and capture multiscale information, and it is specifically tuned to operate optimally to learn from and predict numerically simulated flow (pressure and saturation) fields. We demonstrate the ability of FResNet++ to generalize over multiple high-dimensional input parameter spaces that describe subsurface permeability and porosity heterogeneity. The resulting DL architecture accurately captures the complex interplay between viscous forces and highly heterogeneous permeability and porosity fields. We investigate two-phase flow in porous media, which is the archetypal problem for reservoir simulation giving rise to a system of nonlinearly coupled PDEs with highly heterogeneous coefficients. We show in blind tests that FResNet++ predicts saturation fields more accurately compared to ResU-Net and original FNO with fully connected linear layers. We additionally investigate the effects of using alternative loss functions and an alternative way of utilizing FResNet++ to increase its effectiveness. For the first time in the literature, we show that the spatiotemporal evolution of pressure and saturation fields can be jointly predicted with good accuracy using a single FResNet++ network over long time horizons in response to previously unseen permeability and porosity fields. After a moderate training investment on graphics processing units (GPUs), FResNet++ yields a speedup of at least four orders of magnitude compared to a conventional numerical PDE solver and operates with notably fewer trainable parameters compared to the original FNO. Our numerical experiments validate that FNOs can be utilized in various convolutional neural network-based architectures and can effectively substitute for repetitive physics-based forward simulations for scenario testing.

Encoder–Decoder Convolutional Neural Networks for Flow Modeling in Unsaturated Porous Media: Forward and Inverse Approaches

The computational cost of approximating the Richards equation for water flow in unsaturated porous media is a major challenge, especially for tasks that require repetitive simulations. Data-driven modeling offers a faster and more efficient way to estimate soil moisture dynamics, significantly reducing computational costs. Typically, data-driven models use one-dimensional vectors to represent soil moisture at specific points or as a time series. However, an alternative approach is to use images that capture the distribution of porous media characteristics as input, allowing for the estimation of the two-dimensional soil moisture distribution using a single model. This approach, known as image-to-image regression, provides a more explicit consideration of heterogeneity in the porous domain but faces challenges due to increased input–output dimensionality. Deep neural networks (DNNs) provide a solution to tackle the challenge of high dimensionality. Particularly, encoder–decoder convolutional neural networks (ED-CNNs) are highly suitable for addressing this problem. In this study, we aim to assess the precision of ED-CNNs in predicting soil moisture distribution based on porous media characteristics and also investigate their effectiveness as an optimizer for inverse modeling. The study introduces several novelties, including the application of ED-CNNs to forward and inverse modeling of water flow in unsaturated porous media, performance evaluation using numerical model-generated and laboratory experimental data, and the incorporation of image stacking to account for transient moisture distribution. A drainage experiment conducted on a sandbox flow tank filled with monodisperse quartz sand was employed as the test case. Monte Carlo simulation with a numerical model was employed to generate data for training and validation of the ED-CNN. Additionally, the ED-CNN optimizer was validated using images obtained through non-intrusive photographic imaging. The results show that the developed ED-CNN model provides accurate approximations, addressing the high-dimensionality problem of image-to-image regression. The data-driven model predicted soil moisture with an R2 score of over 91%, while the ED-CNN optimizer achieved an R2 score of over 89%. The study highlights the potential of ED-CNNs as reliable and efficient tools for both forward and inverse modeling in the analysis of unsaturated flow.

Proof of concept: Predicting distress in cancer patients using back propagation neural network (BPNN)

BackgroundResearch findings suggest that a significant proportion of individuals diagnosed with cancer, ranging from 25% to 60%, experience distress and require access to psycho-oncological services. Until now, only contemporary approaches, such as logistic regression, have been used to determine predictors of distress in oncological patients. To improve individual prediction accuracy, novel approaches are required. We aimed to establish a prediction model for distress in cancer patients based on a back propagation neural network (BPNN). MethodsRetrospective data was gathered from a cohort of 3063 oncological patients who received diagnoses and treatment spanning the years 2011–2019. The distress thermometer (DT) has been used as screening instrument. Potential predictors of distress were identified using logistic regression. Subsequently, a prediction model for distress was developed using BPNN. ResultsLogistic regression identified 13 significant independent variables as predictors of distress, including emotional, physical and practical problems. Through repetitive data simulation processes, it was determined that a 3-layer BPNN with 8 neurons in the hidden layer demonstrates the highest level of accuracy as a prediction model. This model exhibits a sensitivity of 79.0%, specificity of 71.8%, positive predictive value of 78.9%, negative predictive value of 71.9%, and an overall coincidence rate of 75.9%. ConclusionThe final BPNN model serves as a compelling proof of concept for leveraging artificial intelligence in predicting distress and its associated risk factors in cancer patients. The final model exhibits a remarkable level of discrimination and feasibility, underscoring its potential for identifying patients vulnerable to distress.

Research on status‐driven hybrid evaluation method for mission success probability

AbstractMission Success Probability is a top‐level indicator that integrates reliability, maintainability, and spare parts security during the system's mission. Numerical simulation is used for mission success evaluation of complex system because it can consider any form of system structure and component failure distribution model. In actual engineering, system components usually deteriorate or fail with the extension of service time and need to be repaired or replaced, so there are dynamic changes in component status during the life cycle. To evaluate the mission success in such dynamic scenarios, it is necessary to repeat the simulation based on the whole system once the local parameters are adjusted. Despite the continuous improvement in computational performance, traditional numerical simulations are not well matched for rapid evaluation and timely decision making in dynamic scenarios. Therefore, an evaluation method that mixes proxy model and simulation sampling is proposed in this paper. The method uses the simulation to complete the data acquisition, and then analyzes the data by inverse function sampling and Copula function to obtain the model training samples, and finally realizes the approximate description of the local unit simulation process by proxy model. The experimental results show that the method can reduce the repetitive simulation scale from the whole system to local unit while ensuring high accuracy, avoiding many repeated calculations in variable scenarios.

Variable-Reluctance Permanent Magnet Resolver Design Guidelines

This article presents the design considerations and geometrical optimization of a variable-reluctance permanent magnet resolver (VRPM resolver). The above-mentioned resolver utilizes permanent magnets (PMs) and magnetic flux measurement units (MFMUs) instead of excitation windings and signal windings. The operating principle of the VRPM resolver is presented, and the geometric and magnetic properties of its various components that influence the accuracy are identified. To study the performance of the resolver, the magnetic equivalent circuit (MEC) model is utilized to ensure both calculation speed and accuracy, while taking saturation and nonlinear <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${B}$ </tex-math></inline-formula> – <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${H}$ </tex-math></inline-formula> curve effects into account. The developed analytical model is coupled with a multiobjective optimization solver to conduct repetitive simulations and attain an optimal design for the studied VRPM resolver. The objective of the optimization process is to achieve the minimum position error, while simultaneously improving the computational time. Time-stepping finite-element analysis (TSFEA) is used to validate and measure the performance in detail. Finally, experimental tests on a prototyped VRPM resolver are employed to verify the simulation results.

Multi-objective optimization of a sandwich rectangular-channel liquid cooling plate battery thermal management system: A deep-learning approach

Maintaining the battery's ideal temperature not only does boost its performance but also lengthens its life span. In this regard, a sandwich cold plate design was proposed and successfully optimized for a battery pack containing 96 cylindrical 4680 battery cells. To make the multi-objective optimization of the battery thermal management system computationally viable, a reduced-cost optimization strategy was proposed. Four design variables, i.e., channel height, number of channels, the distance between adjacent walls of two channels, and inlet velocity system, were considered in this study to reduce the temperature difference in the battery pack, reduce the system's power consumption, and increase its promptitude. The finite element method was used for the three-dimensional simulation of heat transfer and fluid dynamics of the system. The proposed combined steady-transient optimization approach employs neural networks to reduce the cost of repetitive computational fluid dynamics simulations for numerous cases considered by the genetic algorithm during the optimization process. According to the findings, utilizing this approach leads to a dramatic drop of about 95% in computational cost. The optimum design having two cold plates located on the top and bottom of the battery pack was found to have sixteen 1 cm-tall and 0.945 cm-wide channels (per plate) with an inlet velocity equal to 0.45 m/s. Consequently, the objectives faced roughly 20% reduction, 30% reduction, and 10% increase, respectively, as compared to the base design.