Neural Network Surrogate Model Research Articles

A crack identification scheme based on neural network surrogate model and XFEM

Crack detection and identification is of great significance to the safety issues of engineering structures. In this paper, an intelligent crack identification scheme based on extended finite element and neural network surrogate model is proposed to realize the accurate identification of crack parameters. The method firstly employs extended finite element forward analysis to obtain the displacement data of measurement points on geometric models with different crack lengths, and inputs them as sample data to train the agent model, establishes a neural network-based inverse analysis model for crack identification, and automatically updates the threshold and weight of the neural network by using the Gray Wolf optimization algorithm to finally compute the globally optimal results. In the screening of the surrogate model, this paper verifies the advantages of the neural network surrogate model in data fitting and crack information extraction by comparing and analyzing the characteristics of neural network, support vector machine and other surrogate models, and optimizing the neural network surrogate model by adopting the Gray Wolf optimization algorithm. Finally, several numerical examples of different types of cracks are given to verify the validity of the proposed method, and the results show that the proposed method can accurately invert the geometric information of cracks.

Insights into Flexible Bioinspired Fins for Unmanned Underwater Vehicle Systems through Deep Learning.

The last few decades have led to the rise of research focused on propulsion and control systems for bio-inspired unmanned underwater vehicles (UUVs), which provide more maneuverable alternatives to traditional UUVs in underwater missions. Recent work has explored the use of time-series neural network surrogate models to predict thrust and power from vehicle design and fin kinematics. We expand upon this work, creating new forward neural network models that encapsulate the effects of the material stiffness of the fin on its kinematic performance, thrust, and power, and are able to interpolate to the full spectrum of kinematic gaits for each material. Notably, we demonstrate through testing of holdout data that our developed forward models capture the thrust and power associated with each set of parameters with high resolution, enabling highly accurate predictions of previously unseen gaits and thrust and FOM gains through proper materials and kinematics selection. As propulsive efficiency is of utmost importance for flapping-fin UUVs in order to extend their range and endurance for essential operations, a non-dimensional figure of merit (FOM), derived from measures of propulsive efficiency, is used to evaluate different fin designs and kinematics and allow for comparison with other bio-inspired platforms. We use the developed FOM to analyze optimal gaits and compare the performance between different fin materials. The forward model demonstrates the ability to capture the highest thrust and FOM with good precision, which enables us to improve thrust generation by 83.89% and efficiency by 137.58% with proper fin stiffness and kinematics selection, allowing us to improve material selection for bio-inspired fin design.

Neural networks for emergent behavior in biological microstructures

PurposeThis paper develops a neural network surrogate model based on a discrete lattice approach to investigate the influence of complex microstructures on the emergent behavior of biological networks.Design/methodology/approachThe adaptability of network-forming organisms, such as, slime molds, relies on fluid-to-solid state transitions and dynamic behaviors at the level of the discrete microstructure, which continuum modeling methods struggle to capture effectively. To address this challenge, we present an optimized approach that combines lattice spring modeling with machine learning to capture dynamic behavior and develop nonlinear constitutive relationships.FindingsThis integrated approach allows us to predict the dynamic response of biological materials with heterogeneous microstructures, overcoming the limitations of conventional trial-and-error lattice design. The study investigates the microstructural behavior of biological materials using a neural network-based surrogate model. The results indicate that our surrogate model is effective in capturing the behavior of discrete lattice microstructures in biological materials.Research limitations/implicationsThe combination of numerical simulations and machine learning endows simulations of the slime mold Physarum polycephalum with a more accurate description of its emergent behavior and offers a pathway for the development of more effective lattice structures across a wide range of applications.Originality/valueThe novelty of this research lies in integrating lattice spring modeling and machine learning to explore the dynamic behavior of biological materials. This combined approach surpasses conventional methods, providing a more holistic and accurate representation of emergent behaviors in organisms.

Learning experimental data to predict fluid-elastic instability and optimize configuration of tube arrays

Fluid-elastic instability (FEI) poses a particularly significant challenge to the safety of steam generators, given its potential to cause substantial damage in short periods. The tube bundle system model is inherently complex, resulting in expensive costs for experiments and numerical simulations during the design process. In this study, a substantial volume of FEI experimental data, conducted by various scholars, was compiled to create FEI surrogate model. The accuracy and feasibility of this surrogate model were demonstrated and compared to experimental results. This work represents the pioneering reference on the optimization of tube array configurations, achieved by combining FEI deep neural network (DNN) surrogate model with genetic algorithm (GA). The design variables for optimization encompassed tube array configurations and structural parameters. The optimization objective is to achieve a tubular array structure exhibiting a higher critical velocity, Ur, enabling safe operation of the heat exchanger tube bundle across a wider range of flow rates without any FEI accident. The developed approach holds promise for data-driven FEI analysis and optimized tube bundle design, resulting in substantial time and cost savings. We publicly share all code implementations, and we believe that our efforts open a door for the surrogate-model-assisted structural optimization of tube arrays.

Robust optimization of a novel ultraviolet (UV) photoreactor for water disinfection: A neural network approach

To optimize the ultraviolet (UV) water disinfection process, it is crucial to determine the ideal geometric dimensions of a corresponding model that enhance performance while minimizing the impact of uncertain photoreactor inputs. As water treatment directly affects people's lives, it is crucial to eliminate the risks associated with the non-ideal performance of disinfection photoreactors. Input uncertainties greatly affect photoreactor performance, making it essential to develop a robust optimization algorithm in advance to mitigate these effects and minimize the physical and financial resources required for constructing the photoreactors. In the suggested algorithm, a two-objective genetic algorithm is integrated with a non-intrusive polynomial chaos expansion (PCE) technique. Additionally, the Sobol sampling method is employed to select the necessary samples for understanding the system's behavior. An artificial neural network surrogate model is trained using sufficient data points derived from computational fluid dynamics (CFD) simulations. A novel type of UV photoreactors working based on exterior reflectors is chosen to optimize the process with three uncertain input parameters, including UV lamp power, UV transmittance of water, and diffusive fraction of the reflective surface. In addition, four geometrical design variables are considered to find the optimal configuration of the photoreactor. The standard deviation (SD) and the reciprocal of log reduction value (LRV) are set as the objective functions, calculated using PCE. The optimal design provides a LRV of 3.95 with SD of 0.2. The coefficient of variation (CoV) of the model significantly declines up to 7%, indicating the decreased sensitivity of the photoreactor to the input uncertainties. Additionally, it is discovered that the robust model exhibits minimal sensitivity to changes in reflectivity in various flow rates, and its output variability aligns with the SD obtained through robust optimization.

Model-predictive space heating control for energy flexibility – a case study using a long short-term memory neural network surrogate model and a genetic optimization algorithm

This paper presents a case study where a model predictive control (MPC) logic is developed for energy flexible operation of a space heating system in an educational building. A Long Short-Term Memory Neural Network (LSTM) surrogate model is trained on the output of an EnergyPlus building simulation model. This LSTM model is used within an MPC framework where a genetic algorithm is used to optimize setpoint sequences. The EnergyPlus model is used to validate the performance of the control logic. The MPC approach leads to a substantial reduction in energy consumption (7%) and energy costs (13%) with improved comfort performance. Additional energy costs savings are possible (7–16%) if a sacrifice in indoor thermal comfort is accepted. The presented method is useful for developing MPC systems in the design stages where measured data is typically not available. Additionally, this study illustrates that LSTM models are promising for MPC for buildings.

Damage detection of truss structures using meta-heuristic algorithms and optimized group method of data handling surrogate model

Optimization-based methods are increasingly being implemented for structural damage detection (SDD) problems through the minimization of the objective functions based on vibration data. Meanwhile, the challenge of high computational cost hinders the applied use of SDD methods, especially upon addressing large-scale structures. In this study, a two-stage damage detection approach is proposed for damage localization as well as extent quantification which reduces the computational time of SDD problems. In the first stage of the proposed framework, suspected locations of damage are identified by employing a residual force vector. Then in the second stage, damage extent of already identified elements will be assessed by model updating method through three different optimization algorithms, namely particle swarm optimization (PSO), differential evolution (DE), and enhanced colliding bodies optimization (ECBO). To spare time-consuming modal analysis in each iteration of optimization algorithms, a PSO-based group method of data handling (GMDH) polynomial neural network (PNN) surrogate model is proposed. Performance of GMDH PNN strongly depends on the input variables, the number of input variables, and the order of the polynomial. The PSO-based GMDH PNN, developed in this study, employs PSO for the optimum selection of these parameters. Furthermore, the efficiency of Elemental Energy Quotient Indicator (EEQI) as a new damage index proposed in this paper is compared with the approach based on the natural frequency-based index. This paper also deals with the incomplete modal data measured from limited number of sensors. This problem is addressed by employing the model reduction technique and solving the SDD problem with modal information obtained from the reduced model. Numerical examples show that the suggested PSO-based GMDH surrogate model is an accurate tool for various types of SDD problems, and it can provide promising results with a significant reduction in computational time.

[formula omitted][formula omitted]ave [formula omitted]earner: Predicting wave farms power output using effective meta-learner deep gradient boosting model: A case study from Australian coasts

Precise prediction of wave energy is indispensable and holds immense promise as ocean waves have a power capacity of 30–40 kW/m along the coast. Utilising this energy source does not generate harmful emissions, making it a superior substitute for fossil fuel-based energy. The computational expense associated with simulating and computing intricate hydrodynamic interactions in wave farms restricts optimisation methods to a few thousand evaluations and makes a challenging situation for training in deep neural prediction models. To address this issue, we propose a new solution: a Meta-learner gradient boosting method that employs four multi-layer convolutional dense neural network surrogate models combined with an optimised extreme gradient boosting. In order to train and validate the predictive model, we used four wave farm datasets, including the absorbed power outputs and 2D coordinates of wave energy converters (WECs) located along the southern coast of Australia, Adelaide, Sydney, Perth and Tasmania. Furthermore, the capability of the transfer learning strategy is evaluated. The WECs used in this study are of the fully submerged three-tether converter type, similar to the CETO prototype. The effectiveness of the proposed approach is assessed by comparing it with 15 well-established and effective machine learning (ML) methods. The experimental findings indicate that the proposed model is competitive with other ML and deep learning approaches, exhibiting considerable accuracy of 88.8%, 90.0%, 90.3%, and 84.4% in Adelaide, Perth, Sydney and Tasmania and improved robustness in predicting wave farm power output.

Computational Fluid Dynamics–Deep Neural Network (CFD-DNN) Surrogate Model with Graphical User Interface (GUI) for Predicting Hydrodynamic Parameters in Three-Phase Bubble Column Reactors

Computational Fluid Dynamics–Deep Neural Network (CFD-DNN) Surrogate Model with Graphical User Interface (GUI) for Predicting Hydrodynamic Parameters in Three-Phase Bubble Column Reactors

Framework for assessing the performance of overhead transmission lines under wind-temperature effects

This paper presents a framework for evaluating the wind-induced resistance of overhead transmission lines (OTLs), aiming to overcome the limitations associated with disregarding temperature effects and relying on a single metric for failure assessment. The proposed framework factors in temperature influence, carefully considers applied loads and structural characteristics, and accurately calculates failure probability based on a specific case study. Meteorological data is initially collected to create a probabilistic model of wind speed and temperature distribution, accounting for correlation. Components and systems are graded based on structural characteristics and damage modes. The neural network surrogate model is trained with finite element computations for load-structural response sample sets. The surrogate model, in conjunction with logistic regression, generates samples to determine fragility. Finally, the total structural failure probability and monthly failure likelihood are computed by integrating the joint probability model and fragility. Analyses highlight temperature's significant impact on conductor-ground line safety, validating the proposed framework's rationale.

Comprehensive Botnet Detection by Mitigating Adversarial Attacks, Navigating the Subtleties of Perturbation Distances and Fortifying Predictions with Conformal Layers

Botnets are computer networks controlled by malicious actors that present significant cybersecurity challenges. They autonomously infect, propagate, and coordinate to conduct cybercrimes, necessitating robust detection methods. This research addresses the sophisticated adversarial manipulations posed by attackers, aiming to undermine machine learning-based botnet detection systems. We introduce a flow-based detection approach, leveraging machine learning and deep learning algorithms trained on the ISCX and ISOT datasets. The detection algorithms are optimized using the Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) to obtain a baseline detection method. The Carlini & Wagner (C&W) and Generative Adversarial Network (GAN) attacks generate deceptive data with subtle perturbations, targeting each feature used for classification while preserving their semantic and syntactic relationships, which ensures that the adversarial samples retain meaningfulness and realism. An in-depth analysis of the required L2 distance from the original sample for the malware sample to misclassify is performed across various iteration checkpoints, showing different levels of misclassification at different L2 distances of the pertrub sample from the original sample. Our work delves into the vulnerability of various models, examining the transferability of adversarial examples from a Neural Network surrogate model to Tree-based algorithms. Subsequently, models that initially misclassified the perturbed samples are retrained, enhancing their resilience and detection capabilities. In the final phase, a conformal prediction layer is integrated, significantly rejecting incorrect predictions — 58.20% in the ISCX dataset and 98.94% in the ISOT dataset.

Stress-constrained optimization of multiscale structures with parameterized microarchitectures using machine learning

A multiscale topology optimization framework for stress-constrained design is presented. Spatially varying microstructures are distributed in the macroscale where their material properties are estimated using a neural network surrogate model for homogenized constitutive relations. Meanwhile, the local stress state of each microstructure is evaluated with another neural network trained to emulate second-order homogenization. This combination of two surrogate models — one for effective properties, one for local stress evaluation — is shown to accurately and efficiently predict relevant stress values in structures with spatially varying microstructures. An augmented lagrangian approach to stress-constrained optimization is then implemented to minimize the volume of multiscale structures subjected to stress constraints in each microstructure. Several examples show that the approach can produce designs with varied microarchitectures that respect local stress constraints. As expected, the distributed microstructures cannot surpass density-based topology optimization designs in canonical volume minimization problems. Despite this, the stress-constrained design of hierarchical structures remains an important component in the development of multiphysics and multifunctional design. This work presents an effective approach to multiscale optimization where a machine learning approach to local analysis has increased the information exchange between micro- and macroscales.

Study of Draft Tube Optimization Using a Neural Network Surrogate Model for Micro-Francis Turbines Utilized in the Water Supply System of High-Rise Buildings

Study of Draft Tube Optimization Using a Neural Network Surrogate Model for Micro-Francis Turbines Utilized in the Water Supply System of High-Rise Buildings

Real-time channel selection in multi-mode multiplexing optical interconnection implemented by hybrid algorithm and material system

Multi-mode multiplexing optical interconnection (MMOI) has been widely used as a new technology that can significantly expand communication bandwidth. However, the constant-on state of each channel in the existing MMOI systems leads to serious interference for receivers when extracting and processing information, necessitating introducing real-time selective-on function for each channel in MMOI systems. To achieve this goal, combining several practical requirements, we propose a real-time selective mode switch based on phase-change materials, which can individually tune the passing/blocking of different modes in the bus waveguide. We utilize our proposed particle swarm optimization algorithm with embedded neural network surrogate models (NN-in-PSO) to design this mode switch. The proposed NN-in-PSO significantly reduces the optimization cost, enabling multi-dimensional simultaneous optimization. The resulting mode switch offers several advantages, including ultra-compactness, rapid tuning, nonvolatility, and large extinction ratio. Then, we demonstrate the real-time channel selection function by integrating the mode switch into the MMOI system. Finally, we prove the fabricating robustness of the proposed mode switch, which paves the way for its large-scale application.

Artificial neural network-based sequential approximate optimization of metal sheet architecture and forming process

Abstract This paper introduces a sequential approximate optimization method that combines the finite element method (FEM), dynamic differential evolution (DDE), and artificial neural network (ANN) surrogate models. The developed method is applied to address two optimization problems. The first involves metamaterial design optimization for metal sheet architecture with binary design variables. The second pertains to optimizing process parameters in multi-stage metal forming, where the discrete nature arises owing to changing tool geometries across stages. This process is highly nonlinear, accumulating contact, geometric, and material nonlinear effects discretely through forming stages. The efficacy of the proposed optimization method, utilizing ANN surrogate models, is compared with traditionally used polynomial response surface (PRS) surrogate models, primarily based on low-order polynomials. Efficient learning of ANN surrogate models is facilitated through the FEM and Python integration framework. Initial data for surrogate model training is collected via Latin hypercube sampling and FEM simulations. DDE is employed for sequential approximate optimization, optimizing ANN or PRS surrogate models to determine optimal design variables. PRS surrogate models encounter challenges in dealing with nonlinear changes in sequential approximate optimization concerning discrete characteristics such as binary design variables and discrete nonlinear behavior found in multi-stage metal forming processes. Owing to the discrete nature, PRS surrogate models require more data and iterations for optimal design variables. In contrast, ANN surrogate models adeptly predict nonlinear behavior through the activation function's characteristics. In the optimization problem of Metal Sheet Architecture for design target C, the ANN surrogate model required an average of 4.6 times fewer iterations to satisfy stopping criteria compared to the PRS surrogate model. Furthermore, in the optimization of multi-stage deep drawing processes, the ANN surrogate model required an average of 6.1 times fewer iterations to satisfy stopping criteria compared to the PRS surrogate model. As a result, the sequential global optimization method utilizing ANN surrogate models achieves optimal design variables with fewer iterations than PRS surrogate models. Further confirmation of the method's efficiency is provided by comparing Pearson correlation coefficients and locus plots.

Surrogate-model-based optimization of a nonideal polymerization reactor for producing tailored molecular weight distribution

The optimization of polymerization reactors with nonideal mixing has been greatly hindered by a tremendous computation burden. In this study, a global optimization framework to integrate full MWD calculations and detailed reactor dynamics with nonideal mixing behaviors was developed by reducing the considerable computation time required for implementing full MWD. A combination of surrogate-model-based optimization and a computational fluid dynamics (CFD)-compartment reactor model was applied to produce the desired MWD in a nonideal reactor. A neural network (NN) surrogate model was trained using 3,000 MWD data produced from the mechanistic reactor model. The results showed that the model successfully described asymmetric MWD with bimodality resulting from the nonideal mixing behaviors in the reactor. The surrogate model was used to construct operation maps for the asymmetry and bimodality factors, with SHapley Additive exPlanations (SHAP) values showing that the flow rate of the chain transfer agent had the largest impact on the asymmetry factor, whereas temperature significantly influenced the bimodality factor. Finally, the optimization strategy based on the asymmetric and bimodality factors successfully determined operating conditions to produce four tailored MWDs, reported in actual polymerization reactors with R scores of 0.988–0.997.

Data assimilation method and application of shear stress transport turbulence model for complex separation of internal shock boundary layer flow

Data assimilation method and application of shear stress transport turbulence model for complex separation of internal shock boundary layer flow

Bayesian data-driven framework for structural health monitoring of composite structures under limited experimental data

Ultrasonic-guided waves can be used to monitor the health of thin-walled structures. However, the run of experimental damage tests on materials like carbon fiber-reinforced plastics can be impractical and costly. Instead, numerical models can be used to create hybrid datasets to train machine learning algorithms, integrating data from numerical and experimental tests. This paper presents a Bayesian-driven framework to compensate for limited experimental data regarding Lamb wave propagation in composite plates. Using Bayesian inference, the framework updates a numerical finite element model, considering observed uncertainties by sampling posterior probability density functions for input parameters using Markov–Chain Monte Carlo simulations with the Metropolis-Hastings algorithm. A neural network surrogate model speeds-up these simulations, leading to a model that replicates the uncertain experimental setup. This model then generates data to augment true experimental data. Finally, a one-dimensional convolutional neural network is trained on a three different datasets to analyze Lamb wave signals and assess damage. Comparing training strategies shows the hybrid dataset augmented by samples generated by the updated FE model gives the most accurate damage size predictions.

Analysis of slope stochastic fields using a novel deep learning model with attention mechanism

This paper proposes a novel deep learning model incorporating attention mechanisms for the analysis of slope stochastic fields. Initially, a deep learning model is designed to digitally image the stochastic field features of soil strength variability. This is achieved by discretizing the slope soil stochastic field using the Karhunen-Loeve expansion method and transforming the discrete results into digital images. These images are then used to establish a Convolutional Neural Network (CNN) surrogate model that maps the implicit relationship between stochastic field images and slope functional function values, thus calculating the probability of slope failure. The precision of the CNN surrogate model is enhanced through Bayesian optimization and five-fold cross-validation. Moreover, to overcome the limitations of existing data-driven landslide stability prediction models, this study also introduces a Spatial-Temporal Attention (STA) mechanism. By combining the CNN with Long Short-Term Memory (LSTM) networks, the model can accurately approximate the actual results of slope stability calculations in scenarios of high-dimensional representation imaging of stochastic fields and low-probability slope instability. Consequently, this significantly improves the computational efficiency of slope reliability analysis considering stochastic field simulations.

A comprehensive deep learning geometric shape optimization framework with field prediction surrogate and reinforcement learning

The optimization of aerodynamic components' geometric shapes demands a novel technical approach for adaptive and efficient exploration and decision-making within the design space. In this study, we introduce an innovative shape optimization framework that leverages deep reinforcement learning with neural network surrogate models. The field prediction surrogate, realized by two distinct U-net architectures, can efficiently generate holistic field solutions based on the transformed mesh coordinates. Subsequently, an inference engine dynamically calculates the key metric of the flow fields, serving as the objective function for the subsequent geometry-aware Deep Q network (DQN)-based optimization. The framework's efficacy is validated using a rocket nozzle as an illustrative example. During surrogate validation, under both friction and frictionless conditions, the l1 errors of the entire flow field of both the U-net vision transformer (ViT) and U-net convolutional neural network (CNN) architectures are less than 0.4%. The proposed U-net ViT consistently outperforms U-net CNN, and the superiority is particularly evident in complex flow areas, outlet sections, and vacuum thrust prediction. Following training, the DQN model is employed to explore the design variable space. The B-spline defining profile successfully is optimized to a final expanding segment shape with improved thrust. Under frictionless conditions, it closely approaches the theoretical optimum. In the practical condition considering friction, the optimized shape gains a 2.96% thrust improvement. The results demonstrate that the proposed framework, especially when coupled with U-net ViT, exhibits enhanced accuracy and adaptability for shape optimization tasks.