High-fidelity Physics-based Model Research Articles

Global sensitivity analysis of ultrasonic testing simulations of slot-like defects with multifidelity modeling

Abstract In this paper, an efficient global sensitivity analysis (GSA) method for simulation-based ultrasonic testing (UT) of slot-like defects using multifidelity modeling with novel termination criterion is proposed. GSA quantifies the effect of quantities of interest with variability (e.g., position, height, and angle) on the output (e.g., amplitude). GSA with Sobol' indices requires the use of Monte Carlo simulations (MCS) when dealing with nonlinear problems having many parameters. It is impractical to perform GSA directly on high-fidelity physics-based models due to their long evaluation times and the large number of required samples. Multifidelity methods construct surrogate models based on data from an accurate high-fidelity model (HFM) and fast low-fidelity models (LFMs). The multifidelity surrogates evaluate quickly and can be used in lieu of the HFM to accelerate the GSA. Conventional multifidelity methods construct the surrogate to meet a prespecified error metric before using it within an analysis. This requires a separate set of testing data and an often arbitrary error metric threshold. To avoid these, a novel multifidelity modeling termination criterion for GSA is proposed that is based on the absolute relative change of the Sobol' indices. The proposed approach is demonstrated on a simulated UT case inspecting a slot-like defect with three uncertainty variables. The results show a potential for significant reduction in computational cost compared with conventional approaches.

Federated Learning and Unlearning as Enablers of Wind Turbine Digital Twins

Real-time capable models are paramount for the successful adaptation of digital twin technology into industries such as wind energy, but high-fidelity physics-based models cannot achieve the required speed, while data-driven and hybrid methods require large amounts of training data which is typically confidential. In this work, the combination of federated learning with hybrid modeling is proposed to train fast and reliable models across multiple confidential data sets owned by different stakeholders. The approach is demonstrated on physics-guided neural networks to estimate the lift and drag of wind turbine airfoils. A scenario is devised where multiple confidential data subsets are confined to different client devices. It is shown that the physics-guided neural networks can be trained through federated learning across those devices and data subsets and that the resulting models can recover and even surpass the accuracy of a model that is trained conventionally by merging the data subsets on a single device. The presented approach is highly scalable and can be easily adapted to many other applications. This work also discusses federated unlearning methods, which allow data owners to remove all traces of a data subset used in training if they decide to revoke their contribution.

Unveiling the plating-stripping mechanism in aluminum batteries with imidazolium-based electrolytes: A hierarchical model based on experiments and ab initio simulations

Aluminum batteries with imidazolium-based electrolytes present a promising avenue toward the post-lithium-ion battery era. A critical bottleneck is the development of reversible aluminum metal anodes, which is hindered by sluggish battery charge–discharge characteristics due to the reversible/irreversible side reactions on the anodic and cathodic sides. The indispensable discernment of the stripping-plating mechanisms at the electrode–electrolyte interface is not well explored due to the complexity of the various reactions occurring at the surface of the aluminum anode. Herein, a high-fidelity physics-based model is coupled with density functional theory to explain the stripping-plating mechanisms that occur on the surface of the aluminum anode at different current densities. Sensitivity analysis is performed on the experimentally validated physics-based model using a machine-learning Gaussian process regression model to identify the most significant parameters for the plating-stripping mechanism of aluminum. The electrodeposition of aluminum is controlled by both diffusion and kinetics and is limited by the kinetics of the electrochemical reactions at a high current density. This work highlights the assurance of combining models at different scales, machine learning algorithms, and experiments to analyze the behavior of complex electrochemical systems.

Systemic risk mitigation strategy for power grid cascade failures using constrained spectral clustering

Power grids are prone to damage induced by natural or anthropogenic hazard events that might disrupt the functionality of key/multiple grid components concurrently, resulting in a chain of cascade failures spreading throughout the grid. Through integrating grid operation-guided with structure-driven modeling strategies, the current study proposes an approach to manage the risks of such cascade failure (known as systemic-risks) to minimize the possibility of large-scale catastrophic blackouts. The operation-guided modeling strategy is implemented through dispatch and load shedding to rebalance power demand and supply after disruptive events. On the other hand, the grid structure-driven modeling strategy adopted intentional controlled islanding approach through employing a constrained spectral clustering algorithm. Introducing the latter algorithm within the integrated (operation + structure) cascade failure model facilitated identifying the optimal cut-set lines to separate the grid into a group of functioning sub-grids following initial failure and prior to cascade propagation. To demonstrate the utility of the developed systemic risk management strategy, an actual power grid was simulated using a high-fidelity physics-based model under different disruption scenarios to compare the cascade failure size with and without strategy implementation, considering different numbers of sub-grids. The simulations demonstrate that the integrated (dispatch & load shedding-controlled islanding) strategy can effectively boost the overall grid robustness, and subsequently its resilience, and effectively manage catastrophic blackout systemic risks.

Machine learning-based model predictive controller design for cell culture processes.

The biopharmaceutical industry continuously seeks to optimize the critical quality attributes to maintain the reliability and cost-effectiveness of its products. Such optimization demands a scalable and optimal control strategy to meet the process constraints and objectives. This work uses a model predictive controller (MPC) to compute an optimal feeding strategy leading to maximized cell growth and metabolite production in fed-batch cell culture processes. The lack of high-fidelity physics-based models and the high complexity of cell culture processes motivated us to use machine learning algorithms in the forecast model to aid our development. We took advantage of linear regression, the Gaussian process and neural network models in the MPC design to maximize the daily protein production for each batch. The control scheme of the cell culture process solves an optimization problem while maintaining all metabolites and cell culture process variables within the specification. The linear and nonlinear models are developed based on real cell culture process data, and the performance of the designed controllers is evaluated by running several real-time experiments.

A surrogate-assisted uncertainty quantification and sensitivity analysis on a coupled electrochemical–thermal battery aging model

High-fidelity physics-based models are required to comprehend battery behavior at various operating conditions. This paper proposes an uncertainty quantification analysis on a coupled electrochemical–thermal aging model to improve the reliability of a battery model, while also investigating the impact of parametric model uncertainties on battery voltage, temperature, and aging. The coupled model’s high computing cost, however, is a significant barrier to perform uncertainty quantification (UQ) and sensitivity analysis (SA). To address this problem, a surrogate model – i.e, by simulating the outcome of a quantity of interest that cannot be easily computed or measured – based on the Gaussian process regression (GPR) theory and principle component analysis (PCA) is built, using a small collection of finite element simulation results as synthetic training data. In total, 43 variable electrochemical–thermal parameters as well as 13 variable aging parameters are studied and estimated. Moreover, the trained surrogate model is also used in the parameterization of the electrochemical and thermal models. The results show that the uncertainties in the input parameters significantly affect the estimations of battery voltage, temperature, and aging. Based on this sensitivity analysis, the most influential parameters affecting the above mentioned battery outputs are reported. This approach is thereby helpful for developing robust and reliable high-fidelity battery aging models with potential applications in digital twins as well as for synthetic data generation.

Intelligent Risk Prognosis and Control of Foundation Pit Excavation Based on Digital Twin

Timely risk information acquisition and diagnosis during foundation pit excavation (FPE) processes are vital for ensuring the safe and effective construction of underground urban infrastructures. Unfortunately, diverse geological and hydrogeological conditions and complex shapes of the foundation pit create barriers for reliable FPE risk prognosis and control. Furthermore, typical support systems during FPE use temporary measures, which have limited capacity to confront excessive loads, large deformations, and seepage. This study aims to establish an intelligent risk prognosis and control framework based on digital twin (DT) for ensuring safe and effective FPE processes. Previous studies have conducted extensive experimental and numerical analyses for examining unsafe conditions during FPE. How to enable intelligent risk prognosis and control of tedious FPE processes by integrating physics-based models and sensory data collected in the field is still challenging. DT could help to establish the interaction and feedback mechanisms between the physical and virtual space. In this study, the authors have established a DT model that consists of a physical space model and a high-fidelity physics-based model of a foundation pit in virtual space. As a result, a mechanism for effective acquisition and fusion of heterogeneous information from both physical and virtual space is established. Then, the authors proposed an integrated model and data-driven approach for examining safety risks during FPE. In the end, the authors have validated the proposed method through a case study of the FPE of the Wuhan Metro Line. The results show that the proposed method could provide theoretical and practical support for future intelligent FPE.

Physics-informed machine learning with differentiable programming for heterogeneous underground reservoir pressure management

Avoiding over-pressurization in subsurface reservoirs is critical for applications like CO_2 sequestration and wastewater injection. Managing the pressures by controlling injection/extraction are challenging because of complex heterogeneity in the subsurface. The heterogeneity typically requires high-fidelity physics-based models to make predictions on CO_2 fate. Furthermore, characterizing the heterogeneity accurately is fraught with parametric uncertainty. Accounting for both, heterogeneity and uncertainty, makes this a computationally-intensive problem challenging for current reservoir simulators. To tackle this, we use differentiable programming with a full-physics model and machine learning to determine the fluid extraction rates that prevent over-pressurization at critical reservoir locations. We use DPFEHM framework, which has trustworthy physics based on the standard two-point flux finite volume discretization and is also automatically differentiable like machine learning models. Our physics-informed machine learning framework uses convolutional neural networks to learn an appropriate extraction rate based on the permeability field. We also perform a hyperparameter search to improve the model’s accuracy. Training and testing scenarios are executed to evaluate the feasibility of using physics-informed machine learning to manage reservoir pressures. We constructed and tested a sufficiently accurate simulator that is 400 000 times faster than the underlying physics-based simulator, allowing for near real-time analysis and robust uncertainty quantification.

Deep learning-based solvability of underdetermined inverse problems in nonlinear ultrasonic characterization of micro damages

Detection and evaluation of micro-damages in the early stages of engineering failure are crucial for various industrial structures to ensure their safety and prevent further catastrophic accidents. The nonlinear ultrasonic technique (NUT) has gained increasing popularity and recognition for breaking through the detection sensitivity limit upon micro-damages that usually are invisible to conventional linear techniques. However, it remains an ongoing challenge to quantitatively characterize micro-damages using NUT due to great difficulties in fully modeling the complicated interaction mechanism between the nonlinear ultrasonic waves and micro-damages. This work presents a data-driven perspective for solving multiparameter underdetermined inverse problems that are at the core of NUT, while allowing by-passing the creation of high-fidelity physics-based models. Nonlinear Lamb wave measurements with group-velocity mismatching are conducted to introduce both size and localization information of damages to the assembled dataset. A nonlinearity-aware discrete wavelet transform-bidirectional long short-term memory network is proposed to directly process nonlinear ultrasonic responses to automatically model latent nonlinear dynamics, thus establishing the complex mapping between the nonlinear ultrasonic signals and the multi-dimensional damage features. In particular, an attempt is made to augment the physical explainability of the proposed deep learning approach through a frequency component importance analysis. The trained network enables accurate and explainable predictions of length and localization of closed cracks and robustness against varying degrees of noise. Our work paves a promising and practical way to promote the transformation of NUT from the qualitative analysis for accurate and efficient quantitative prediction.

A Novel Digital Twin Architecture with Similarity-Based Hybrid Modeling for Supporting Dependable Disaster Management Systems.

Disaster management systems require accurate disaster monitoring and prediction services to reduce damages caused by natural disasters. Digital twins of natural environments can provide the services for the systems with physics-based and data-driven disaster models. However, the digital twins might generate erroneous disaster prediction due to the impracticability of defining high-fidelity physics-based models for complex natural disaster behavior and the dependency of data-driven models on the training dataset. This causes disaster management systems to inappropriately use disaster response resources, including medical personnel, rescue equipment and relief supplies, to ensure that it may increase the damages from the natural disasters. This study proposes a digital twin architecture to provide accurate disaster prediction services with a similarity-based hybrid modeling scheme. The hybrid modeling scheme creates a hybrid disaster model that compensates for the errors of physics-based prediction results with a data-driven error correction model to enhance the prediction accuracy. The similarity-based hybrid modeling scheme reduces errors from the data dependency of the hybrid model by constructing a training dataset using similarity assessments between the target disaster and the historical disasters. Evaluations in wildfire scenarios show that the digital twin decreases prediction errors by approximately 50% compared with those of the existing schemes.

Control-oriented Model of HVAC and Battery Cooling Systems in Electric Vehicles

Energy-efficient thermal management of heating, ventilation, and air-conditioning (HVAC) and battery cooling systems is essential for an enhanced driving range of electric vehicles. With recent advances in vehicle connectivity technologies and vehicle computation units, a predictive model has been emphasized in control design. However, the system complexity due to refrigerant circuits and coupling with the battery cooling system makes it challenging to develop a computationally inexpensive model for potential model predictive control applications. This paper presents a novel control-oriented model to capture the behavior of a combined HVAC and battery cooling system for electric vehicles. To reduce the complexity due to the nonlinear two-phase flow heat transfer, the refrigerant circuit is modeled based on the assumption of a quasi-steady ideal vapor-compression cycle. The proposed model is calibrated with a three-step optimization process and then validated against a high-fidelity physics-based model with open-loop simulations. The simulation results show the effectiveness of the proposed model with four states and five actuators: the root mean square errors (RMSE) of the battery temperature and the coolant temperature are 0.41°C and 0.35°C, respectively, and the normalized RMSEs of the heat transfer rate at the heat exchangers are within the range of 2.8% to 10.6%.

Hybrid Approach for Health Monitoring of Mud Motor Fleet

A mud motor is a kind of positive displacement motor (PDM) that is used to transform the hydraulic energy of drilling fluids (mud) into mechanical energy. This mechanical energy enables the drill bit to cut the rock and drill a well. It is one of the key parts of downhole assembly that is placed in the drillstring to provide additional power to the bit while drilling as its power downhole output is still unmatched. Mud motor failure is a common and costly issue in drilling operations. A proper prediction of the failure as well as an estimation of the remaining useful life (RUL) are essential for timely downhole mud motor maintenance and drilling optimization.
 Until now, the oil and gas industry has lacked reliable procedures to monitor and maintain the health of mud motors, resulting in unnecessary maintenance costs as well as unpredictable and costly drilling failures. Recently, Schlumberger has addressed this problem with an industry-first prognostics and health management (PHM) solution, which not only estimates the health of the mud motor and tracks RUL, but also creates a new service for clients and provides a competitive advantage. Timely mud motor retirement and maintenance will ultimately reduce failures and NPT.
 The proposed PHM solution is suitable for real-time implementation and combines two different sterling algorithms for reliable prediction of possible problems with the mud motors. It enables the estimation of the mud motor health both on the system level with the entire mud motor (system level PHM model) and on the subcomponent level (power section PHM model) – the most critical component of the mud motor.
 The system-level algorithm model leverages both surface and downhole drilling data as well as mud motor characteristic curves to compute the severity of mud motor degradation. A special mud motor degradation indicator is defined. The indicator is calculated to evaluate the degree of power section decay at each time recorded from thousands of field jobs. The trends of the degradation with respect to drilling time and drilling distance are extracted for each motor job. Based on the study of large datasets, good correlation was observed between the mud motor degradation indicator and mud motor failures.
 The power section PHM model uses downhole measurements to estimate the RUL of the elastomer – the life-limiting component inside the power section. It is based on a high-fidelity model and uses a hybrid approach by combining a high-fidelity physics-based model of a power section and data-driven approaches with machine learning techniques. Machine learning methods were applied to derive a reduced order surrogate model (ROM) of power sections from the original physics-based models for real-time applications. This ROM outputs the estimation of performance and fatigue characteristics of the considered power section depending on the considered drilling conditions such as differential pressure, downhole temperature, flow rate, and mud compatibility. As the result, the model analyzes accumulative risk of fatigue failure and produces real-time health information for the power section as a percentage of the remaining lifespan.
 The new solution for mud motor PHM was successfully verified and tested in the field. Comparison of the predicted mud motor fatigue life with the actual observed postjob conditions and job failures demonstrated good results of the developed models. The PHM enables optimization of mud motor selection, drilling configuration, and maintenance operations by minimizing RUL uncertainties while facilitating rerun decisions and avoiding overmaintenance and premature retirements. The whole solution is currently being integrated into a drilling platform including the maintenance system, the well construction planning, and the execution. It maximizes the equipment usage with increased drilling performance without sacrificing reliability and enables optimal fleet management of a drilling process for revenue maximization.

Uncertainty Quantification of Multiple Gas Transport and Sorption in Porous Polymers

A high-fidelity physics-based model of mixed-gas transport coupled with kinetic and equilibrium adsorption is derived, and experiments were performed in order to calibrate and exercise the model. In the literature, a continuum-scale model that couples Fickian diffusion with Henry’s law absorption, and kinetic Langmuir adsorption was previously developed to describe the diffusion and sorption of moisture in porous materials. Here, we expand the model to gases, rather than moisture, derive, and implement a competitive adsorption mechanism into the model to enable mixed-gas sorption. This model facilitates a mechanistic-based understanding of the sorption and diffusion processes of mixed gases in polymeric materials. Diffusion and sorption experiments were conducted for a range of partial pressures; model validation and calibration were carried out by comparing modeled mass uptake and experimental data considering the uncertainties of conceptualized (or assumed) physical processes and system parameters. Uncertainty quantification and sensitivity analysis methods are described and exercised here to demonstrate the capability of this predictive model.

Issue: Selected paper of OIPE 2016 on modeling and optimal design of electromagnetic and electronic devices

Issue: Selected paper of OIPE 2016 on modeling and optimal design of electromagnetic and electronic devices

Data-Driven Model Predictive Control with Regression Trees—An Application to Building Energy Management

Model Predictive Control (MPC) plays an important role in optimizing operations of complex cyber-physical systems because of its ability to forecast system’s behavior and act under system level constraints. However, MPC requires reasonably accurate underlying models of the system. In many applications, such as building control for energy management, Demand Response, or peak power reduction, obtaining a high-fidelity physics-based model is cost and time prohibitive, thus limiting the widespread adoption of MPC. To this end, we propose a data-driven control algorithm for MPC that relies only on the historical data. We use multi-output regression trees to represent the system’s dynamics over multiple future time steps and formulate a finite receding horizon control problem that can be solved in real-time in closed-loop with the physical plant. We apply this algorithm to peak power reduction in buildings to optimally trade-off peak power reduction against thermal comfort without having to learn white/grey box models of the systems dynamics.

Methodology for Path Planning with Dynamic Data-Driven Flight Capability Estimation

This paper presents methodology to enable path planning for an unmanned aerial vehicle that uses dynamic data-driven flight capability estimation. The main contribution of the work is a general mathematical approach that leverages offline high-fidelity physics-based modeling together with onboard sensor measurements to achieve dynamic path planning. The mathematical framework, expressed as a constrained partially observable Markov decision process, accounts for vehicle capability constraints and is robust to modeling error and disturbances in both the vehicle process and measurement models. Vehicle capability constraints are incorporated using probabilistic support vector machine surrogates of high-fidelity physics-based models that adequately capture the richness of the vehicle dynamics. Sensor measurements are treated in a general manner and can include combinations of multiple modalities such as that through global positioning system signals, inertial mass unit outputs, and structural strain data of the airframe. Results are presented for a simulated three-dimensional environment and point-mass airplane model. The vehicle can dynamically adjust its trajectory according to the observations it receives about its current state of health, thereby retaining a high probability of survival and mission success.

An energetically consistent concurrent multiscale method for heterogeneous heat transfer and phase transition applications

A concurrent multiscale method is developed to model time-dependent heat transfer and phase transitions in heterogeneous media and is formulated in a way such that the energy being exchanged between scales is conserved. Ensuring this energetic consistency among scales enables the implementation of high fidelity physics-based models at critical locations within the coarse-scale to temporally and spatially resolve highly complex and localized phenomena. To achieve this, only Neumann boundary conditions are applied over the fine scale domain, ensuring a conservative formulation. The coarse-scale solution is used to reconstruct these Neumann boundary conditions on the fine scale, which are then used to evolve a separate system of governing equations. The results on the fine scale are then sent back to the coarse scale through an energy-based homogenization scheme. Transient simulations for the heat equation are implemented with the proposed method to demonstrate its accuracy in energy conservation and effectiveness, including the coupling of a phase field model at the fine scale to a coarse-scale heat equation.

An effective model for analysis of reinforced concrete members and structures under blast loading

The traditional fiber beam model has been widely used in the seismic analysis of reinforced concrete members and structures. However, the inability to capture shear failure restricts its application to blast loadings. In this article, a numerical model that considers both rate-dependent shear behavior and damage effect is proposed based on the traditional fiber beam element. This is achieved using the modified compression-field theory with a concrete damage model and bilinear steel model in the principal directions. Meanwhile, a condensed three-dimensional stress–strain relation from the isotropic hardening plasticity model is implemented to simulate longitudinal reinforcement bars, as large shear strain would be produced under severe blast loads. The proposed model is validated by comparing the numerical and test results. The high-fidelity physics-based finite element model, validated by the same experiment, is also used in the study to prove the efficiency of the proposed model. Case studies of a reinforced concrete beam and a six-story reinforced concrete frame structure subjected to blast loads are then carried out. The results indicate that the proposed model is reliable compared with the high-fidelity physics-based model. In addition to the accuracy, comparisons of the computational time show an excellent performance with respect to the efficiency of the proposed model.

Modeling of the Vacuum Arc Remelting Process for Estimation and Control of the Liquid Pool Profile

Vacuum arc remelting (VAR) is an industrial metallurgical process widely used throughout the specialty metals industry to cast large alloy ingots. The final ingot grain structure is strongly influenced by the molten metal pool profile, which in turn depends on the temperature distribution in the ingot. A reduced-order model of the solidifying ingot was developed specifically for dynamic control and estimation of the depth of molten liquid pool atop the ingot in a VAR process. This model accounts only for the thermal aspects of the system ignoring other physical domains such as fluid flow and electromagnetic effects. Spectral methods were used to obtain a set of nonlinear dynamic equations which capture the transient characteristics of liquid pool profile variations around a quasi-steady operating condition. These nonlinear equations are then linearized and further simplified by suppressing fast modes. The resulting system was used to construct a linear-quadratic-gaussian (LQG) controller which was tested in a laboratory-scale furnace showing a good performance. A high-fidelity physics-based model is used in real-time to provide information about the solidifying ingot and potential solidification defects.

Modeling by numerical reduction of modes for multivariable control of an optical-fiber draw process

Motivated by a need for a method to derive practical and physical-based dynamic models that capture the essential characteristics of an optical-fiber draw process for precision control of diameter uniformity, we extend the Karhunen-Loeve decomposition technique with a Galerkin procedure to derive a reduced-order model (ROM) for a multivariable distributed-parameter system. We validated the ROM derived from a high-fidelity physics-based model by simulating a modern optical-fiber draw process, the numerical solutions for which have been experimentally verified in our earlier studies. Perturbation studies demonstrated that the 24th-order ROM agrees remarkably well with the original nonlinear semi-two-dimensional and quasi-one-dimensional distributed models. We further examine the efficiency of the ROM in the context of a model-based H/sub /spl infin///LQG fiber drawing control system for the regulation of the fiber diameter and tension. The results show that variations in fiber diameter can be reduced significantly by appropriately distributing the number of retained eigenmodes among the physical state variables in the ROM. We also demonstrate that controlling the surrounding air temperature in addition to the draw speed is very effective in regulating both the fiber diameter and tension while simultaneously keeping the draw speed and temperature fluctuations to a minimum. Note to Practitioners-Because of the stringent production requirements (on draw speed, tension, and temperature), diameter uniformity is a challenging distributed control problem in modern fiber production where progressively larger diameter preforms are drawn at higher speeds. The reduced-order model offers an effective way to observe physical variables in a multivariable distributed-parameter system, which may not be physically measured. While developed in the context of fiber diameter control, the modeling techniques presented in this paper are applicable to other material processing systems, such as deposition thickness control in semiconductor wafer manufacturing.