Computational Effort Research Articles

Data‐driven forecasting of ship motions in waves using machine learning and dynamic mode decomposition

SummaryData‐driven forecasting of ship motions in waves is investigated through feedforward and recurrent neural networks as well as dynamic mode decomposition. The goal is to predict future ship motion variables based on past data collected on the field, using equation‐free approaches. Numerical results in two case studies involving the course‐keeping of a naval destroyer in a high sea state using simulation data at model scale are presented. The proposed methods reveal successful in predicting ship motions both in short‐term and medium‐term perspectives with accuracy and reduced computational effort, thus enabling further advances in the identification, control, and optimization of ships operating in waves.

Robust Gaussian Process Regression Method for Efficient Tunneling Pathway Optimization: Application to Surface Processes.

Simulation of surface processes is a key part of computational chemistry that offers atomic-scale insights into mechanisms of heterogeneous catalysis, diffusion dynamics, and quantum tunneling phenomena. The most common theoretical approaches involve optimization of reaction pathways, including semiclassical tunneling pathways (called instantons). The computational effort can be demanding, especially for instanton optimizations with an ab initio electronic structure. Recently, machine learning has been applied to accelerate reaction-pathway optimization, showing great potential for a wide range of applications. However, previous methods still suffer from numerical and efficiency issues and were not designed for condensed-phase reactions. We propose an improved framework based on Gaussian process regression for general transformed coordinates, which has improved efficiency and numerical stability, and we propose a descriptor that combines internal and Cartesian coordinates suitable for modeling surface processes. We demonstrate with 11 instanton optimizations in three representative systems that the improved approach makes ab initio instanton optimization significantly cheaper, such that it becomes not much more expensive than a classical transition-state theory rate calculation.

Machine Learning Algorithms That Emulate Controllers Based on Particle Swarm Optimization—An Application to a Photobioreactor for Algal Growth

Particle Swarm Optimization (PSO) algorithms within control structures are a realistic approach; their task is often to predict the optimal control values working with a process model (PM). Owing to numerous numerical integrations of the PM, there is a big computational effort that leads to a large controller execution time. The main motivation of this work is to decrease the computational effort and, consequently, the controller execution time. This paper proposes to replace the PSO predictor with a machine learning model that has “learned” the quasi-optimal behavior of the couple (PSO and PM); the training data are obtained through closed-loop simulations over the control horizon. The new controller should preserve the process’s quasi-optimal control. In identical conditions, the process evolutions must also be quasi-optimal. The multiple linear regression and the regression neural networks were considered the predicting models. This paper first proposes algorithms for collecting and aggregating data sets for the learning process. Algorithms for constructing the machine learning models and implementing the controllers and closed-loop simulations are also proposed. The simulations prove that the two machine learning predictors have learned the PSO predictor’s behavior, such that the process evolves almost identically. The resulting controllers’ execution time have decreased hundreds of times while keeping their optimality; the performance index has even slightly increased.

Efficient Detection of Exchangeable Factors in Factor Graphs

To allow for tractable probabilistic inference with respect to domain sizes, lifted probabilistic inference exploits symmetries in probabilistic graphical models. However, checking whether two factors encode equivalent semantics and hence are exchangeable is computationally expensive. In this paper, we efficiently solve the problem of detecting exchangeable factors in a factor graph. In particular, we introduce the detection of exchangeable factors (DEFT) algorithm, which allows us to drastically reduce the computational effort for checking whether two factors are exchangeable in practice. While previous approaches iterate all O(n!) permutations of a factor's argument list in the worst case (where n is the number of arguments of the factor), we prove that DEFT efficiently identifies restrictions to drastically reduce the number of permutations and validate the efficiency of DEFT in our empirical evaluation.

Notch corrosion fatigue behaviour and simplified strain energy density-based notch fatigue life assessment of Grade-A shipbuilding steel

Corrosion fatigue is a key challenge in lifecycle management of ship hull structures. Although sea wave fatigue stresses are linear-elastic, highly-damaging elastic–plastic conditions occur at geometrical discontinuities or notches. Estimation of notch stress-strains to characterize notch fatigue behaviour conventionally requires large computational efforts. Also, constant strain energy density (SED), as a driving force, is far more effective in realistic applications versus constant stress and strain fatigue life approaches. Identified research gaps were the development of a simplified SED approach for fatigue life assessment of hole-notched structures. Fatigue studies were conducted on hole-notched specimens of Grade-A shipbuilding steel in air and corrosion mediums. Strong synergistic interactions of corrosion and fatigue damage mechanisms were studied by scanning electron microscope (SEM) fractography. Notch corrosion fatigue life was reduced by 70% from notch air fatigue life due to environmental effect. Two new parameters are proposed to improve the efficiency of notch fatigue life assessment, namely “fatigue notch strain energy density concentration factor (K_SED) and “fatigue notch toughness”. This improved fatigue life assessment method was demonstrated as equally reliable but much simpler to implement, eliminating the complex computational analysis. This improved method has promising applications in real-time structural health monitoring (SHM) of ship hull structures.

Thermo-magnetic radiative flow in porous enclosure with deep-learning parameter estimation

This study numerically investigates convective heat transfer, fluid flow, and entropy generation within an inverted T-shaped porous enclosure filled with a hybrid nanofluid. The interplay between thermal radiation and a magnetic field is considered, as these factors significantly influence thermal acquisition processes in renewable and thermal engineering applications, particularly solar power collectors with the selected physical domain. By analyzing various combinations of flow parameters, this work aims to identify favorable conditions for maximizing heat transfer efficiency in such systems. A Darcy–Forchheimer–Brinkman-based mathematical model has been incorporated and numerically simulated using the penalty finite element method (FEM). Moreover, the variation of fluid flow, isotherms, and entropy generation is thoroughly analyzed for the variation of non-dimensional permeability parameter (Darcy number, Da), buoyancy force (Rayleigh number, Ra), magnetic field (Hartmann number, Ha), porosity value (ϵ), and thermal radiation parameter (Rd). The comprehensive result demonstrates that increasing values of Ra, Da, ϵ, and Rd intensify entropy generation, convective heat, and fluid flow phenomena, whereas the increasing value of Ha substantially weakens these phenomena. The thermal efficiency of the model attends at the lower buoyancy forces (Ra≤105) for each flow parameter (except for Ha). However, the increasing magnetic forces (Ha) substantially improve the thermal efficiency of the developed model. Additionally, this study employs a feed-forward neural network (FNN) based deep learning (DL) approach to predict the parametric influence on convective heat transport rate (Num), total entropy generation rate (Stot), and total Bejan number due to heat transport irreversibility (Beθ,tot) based on the minimally supplied experimental data for FNN training process. The trained FNN model predicts the results very quickly with 98.6% accuracy compared to FEM results, demonstrating its efficiency in approximating the parametric influence of complex mathematical model results. Moreover, this approach significantly reduces computational costs and efforts compared to traditional numerical methods, making it a valuable tool for forecasting the influence of an unknown set of flow parameters in complex mathematical models.

Transient surrogate modeling of modally reduced structures with discontinuous loads and damping

In this work, a surrogate model for structural, transient and discontinuously excited finite element method simulations is developed. This allows to reduce the computational effort of repeated calculations of identical models under different load cases. The architecture of the surrogate combines fully connected neural network layers with long short-term memory layers. For the reproduction of different damping ratios, a categorical variable is added to the continuous input data. Based on a recursive flow of the predicted data back to the input layer, long-term dependencies do not vanish due to short-input sequences. The system dimension is reduced by applying the model-order reduction technique for modal decomposition. The high accuracy of the surrogate and the reduction of computational costs are shown on an academic example of a cantilever beam and a real-world example of a robot. The advantages of our approach are illustrated in comparison with state-of-the-art surrogates for transient finite element analysis. By using the surrogate proposed in this study, oscillations due to discontinuous excitation of mechanical structures can be reproduced. For this purpose, only short-input sequences are necessary since the excitation of the oscillations does not have to be part of the input sequence during the whole duration of the oscillations. Due to the categorical variable for the damping ratio, the surrogate can account for the influence of different damping in parameter studies.

Experimental‐Assisted Approach to Develop a Numerical Model for Simulating the Reaction Propagation in Reactive Multilayers

One outstanding feature of self‐propagating reactions is their ability to release heat of reaction over both temporal and spatial scales, enabling the sustained progression of the reaction after a local ignition. They propagate in the form of a continuous reaction front through the mixture of the starting materials. Previous research on reactive materials has predominantly focused on unraveling the microstructure property relationships influencing released energy in reacting multilayers. This involved considering coupled differential equations, including the heat conduction equation and Fick's law. In this study, the introduction of a purely thermal numerical macroscale model is made, incorporating two states of material properties that differentiate between the thermal characteristics before and after phase formation. The homogenization of material properties before the phase formation is accomplished through the consideration of directional‐temperature‐dependent thermal conductivity and temperature‐dependent‐specific heat capacity. The energy‐release function is derived using experimental data for the reaction velocity depending on bilayer thickness. This model allows for the exploration of reaction motion and temperature profiles, achieving qualitative conformity with experimental measurements for freestanding foil, and necessitating reasonable computational effort.

Optimal integration and planning of PV and wind renewable energy sources into distribution networks using the hybrid model of analytical techniques and metaheuristic algorithms: A deep learning-based approach

The optimal integration of Renewable Energy Sources (RESs) into distribution networks is crucial for sustainable energy generation. However, existing approaches utilized for optimal RES integration, including metaheuristic and analytical methods, exhibit limitations. Analytical methods are simple to implement, computationally efficient, and able to model power system mechanisms, yet they face scalability issues, making them inefficient for large-scale distribution systems. On the other hand, metaheuristic approaches, while flexible for complex problems, suffer from premature convergence problems, limiting their application in real distribution networks. To address these issues, this paper introduces a novel hybrid model that combines the strengths of analytical and metaheuristic techniques. The proposed model employs a deep learning-based method called bi-directional long short-term memory (B-LSTM) network is utilized to handle the uncertainties associated with the Photovoltaic (PV) and Wind Turbine (WT) generation and load demand. Subsequently, a distinctive analytical multi-objective index-based procedure is developed to minimize the real and reactive energy loss and voltage deviation by analyzing the branch current change caused by the RES penetration in distribution networks. The optimization problem is then solved using the binary particle swarm optimization (BPSO) technique. Importantly, the presented method considers the long-term nature of optimal RES integration and reflects the seasonal impacts of the load and RES output powers. The proposed approach is applied to the IEEE 33-bus test system. The simulation results show that the introduced hybrid model, by addressing the limitations of existing approaches, offers a promising solution for efficient and effective RES integration in distribution systems while requiring low computational effort, thereby advancing the transition towards sustainable energy infrastructure.

Graph Neural Networks to Simulate Flexible Pavement Responses Using Three-Dimensional Finite Element Analysis Data

The significant computational costs and efforts required for accurate three-dimensional (3D) finite element (FE) pavement response calculations necessitate an expedited approach. This study proposes a graph neural network (GNN)-based simulator for the modeling of 3D pavement structural responses under tire loading. The GNN model was trained using 240 simulations of 3D pavement FE data of flexible pavement structures. The simulator represented the state of pavement structure meshes in FE analysis at any given timestep as a graph, with FE nodes encoded as graph nodes and mesh edges as graph edges. The dynamic behaviors of pavement FEs were computed via learned message-passing between two graphs within two continuous timesteps. The one-step mean squared error (MSE) and rollout MSE were used as evaluation metrics for the GNN model. The results showed that, given an initial state of FE responses, the model could perform accurate one-step predictions, extending to trajectory predictions with one-step MSE as low as [Formula: see text] and rollout MSE around [Formula: see text]. The prediction framework is efficient; it requires a week of model training but only a mere 5 min of prediction for each single case. This a contrast to traditional 3D FE analyses that can span hours to weeks for a single case. The hyperparameters, including the number of message-passing steps M and the number of historical timesteps C, were established as 10 and 1, respectively, based on model performance and computation time. It was also observed that data normalization before training could significantly reduce model simulation noise.

Robust Automated Tumour Segmentation Network Using 3D Direction-Wise Convolution and Transformer

AbstractSemantic segmentation of tumours plays a crucial role in fundamental medical image analysis and has a significant impact on cancer diagnosis and treatment planning. UNet and its variants have achieved state-of-the-art results on various 2D and 3D medical image segmentation tasks involving different imaging modalities. Recently, researchers have tried to merge the multi-head self-attention mechanism, as introduced by the Transformer, into U-shaped network structures to enhance the segmentation performance. However, both suffer from limitations that make networks under-perform on voxel-level classification tasks, the Transformer is unable to encode positional information and translation equivariance, while the Convolutional Neural Network lacks global features and dynamic attention. In this work, a new architecture named TCTNet Tumour Segmentation with 3D Direction-Wise Convolution and Transformer) is introduced, which comprises an encoder utilising a hybrid Transformer-Convolutional Neural Network (CNN) structure and a decoder that incorporates 3D Direction-Wise Convolution. Experimental results show that the proposed hybrid Transformer-CNN network structure obtains better performance than other 3D segmentation networks on the Brain Tumour Segmentation 2021 (BraTS21) dataset. Two more tumour datasets from Medical Segmentation Decathlon are also utilised to test the generalisation ability of the proposed network architecture. In addition, an ablation study was conducted to verify the effectiveness of the designed decoder for the tumour segmentation tasks. The proposed method maintains a competitive segmentation performance while reducing computational effort by 10% in terms of floating-point operations.

This is SpArta: Rigorous Optimization of Regionally Resolved Energy Systems by Spatial Aggregation and Decomposition

Energy systems with high shares of renewable energy are characterized by local variability and grid limitations. The synthesis of such energy systems, therefore, requires models with high spatial resolution. However, high spatial resolution increases the computational effort. Here, we present the SpArta method for rigorous optimization of regionally resolved energy systems by Spatial Aggregation and decomposition. SpArta significantly reduces computational effort while providing a near-optimal solution at the full spatial resolution of sector-coupled energy systems.SpArta first reduces problem size by spatially aggregating the energy system using clustering. The aggregated problem is then relaxed and restricted to obtain a lower and an upper bound. The spatial resolution is iteratively increased until the difference between upper and lower bound satisfies a predefined optimality gap.Finally, each cluster of the aggregated problem is redesigned at full resolution. For this purpose, SpArta decomposes the original synthesis problem into subproblems for each cluster. Combining the redesigned cluster solutions yields an optimal feasible solution of the full-scale problem within a predefined optimality gap. SpArta thus optimizes large-scale energy systems rigorously with significant reductions in computational effort. We apply SpArta to a case study of the sector-coupled German energy system, reducing the computational time by a factor of 7.5, compared to the optimization of the same problem at full spatial resolution. As SpArta shows a linear increase in computational time with problem size, SpArta enables computing larger problems allowing to resolve energy system designs with improved accuracy.

Mapping flagellated swimmers to surface-slip driven swimmers

Flagellated microswimmers are ubiquitous in natural habitats. Understanding the hydrodynamic behavior of these cells is of paramount interest, owing to their applications in bio-medical engineering and disease spreading. Since the last two decades, computational efforts have been continuously improved to accurately capture the complex hydrodynamic behavior of these model systems. However, modeling the dynamics of such swimmers with fine details is computationally expensive due to the large number of unknowns and the small time-steps required to solve the equations. In this work we propose a method to map fully resolved flagellated microswimmers to coarse, active slip driven swimmers which can be simulated at a reduced computational cost. Using the new method, the slip driven swimmers move with the same velocity, to machine precision, as the flagellated swimmers and generate a similar flow field with a controlled accuracy. The method is validated for swimming patterns near a no-slip boundary, interactions between swimmers and scattering with large obstacles.

A novel lightweight dynamic focusing convolutional neural network LAND-FCNN for EEG emotion recognition

The inefficiency of model inference and the limited sample data are significant issues in electroencephalogram (EEG) emotion recognition models based on neural networks. Therefore, we propose a novel lightweight adaptive dynamic focusing convolutional neural network (LAND-FCNN). We use Partial Convolution (PConv) and Batch interactive attention in constructing the Backbone network. PConv reduces the computational cost by selectively extracting features from a subset of input channels, utilizing feature graph redundancy. Batch interactive attention establishes connections between various types of samples to address sample scarcity without relying on data augmentation. Then Backbone serves as the encoder for the Glance and Focus Dynamic Inference Networks (GFnet). The patch proposal network in GFnet adaptively captures task-relevant areas of each sample, allowing encoders to process only those areas. This increases inference speed and ensures reliable results with minimal computational effort. Extensive validation of diverse datasets with varying distributions confirms the efficacy and precision of LAND-FCNN.

Improving analytic approximations of TSP tour lengths with adjustment factors

Optimizing Traveling Salesman Problem (TSP) tours requires substantial computational effort, leading researchers to develop approximations relating tour length to the number of visited points, n. Existing models, such as the √nA predictor, effectively approximate tour lengths for large-capacity vehicles but sacrifice accuracy for small n values relevant for most practical applications. Consequently, this study addresses this gap by proposing models with uniform node distributions, which incorporate realistic factors, such as central vs. random starting points and various service zone shapes. These factors are then integrated into a single equation, enhancing applicability. Furthermore, the exponent of n is statistically estimated to be significantly different from 0.5, challenging previous studies. Our proposed model estimates TSP tour lengths more accurately, particularly for small n values, and maintains accuracy for large n values, with errors below 3.11% for up to 600 points. This model offers a more precise and versatile alternative to current models.

Modeling gas flow in low-permeability formations: An efficient combination of mixed finite elements and high order time integration schemes

Numerical simulation of gas flow in low permeability formations is a challenging task due to the high nonlinearity induced by (i) the compressibility of the gas, (ii) the Klinkenberg slippage effect and (iii) the Langmuir adsorption of the gas on the pore surface. Because of these nonlinearities, modeling gas flow in low permeability formations requires a great deal of computational effort. In this work, we develop an efficient numerical model using advanced spatial and temporal discretization methods for a simultaneous solution of the coupled equations of gas flow, cubic Peng-Robinson equation of state, slippage effect and Langmuir adsorption. The spatial discretization is performed with the lumped hybrid formulation of the mixed finite element method which is well adapted for fluid flow in heterogeneous porous media. The time integration is performed with high-order methods via the method of lines (MOL) which allows large time steps and efficient solution of the nonlinear system of equations. Numerical experiments, performed for gas extraction in a heterogeneous domain, point out the high efficiency of the new model since it can be until 10 times more efficient than the classical first-order time discretization method. Results of global sensitivity analysis show that the intrinsic gas-phase permeability and the Klinkenberg factor are the most influential parameters controlling the pumping rate in the case of gas extraction in a homogeneous domain.

Detection of tuberculosis using customized MobileNet and transfer learning from chest X-ray image

One of the most contagious diseases in the world, tuberculosis (TB) is brought on by the bacteria Mycobacterium tuberculosis. This hazardous scenario can cause life losses and requires expert doctors and several hours to detect the disease. Using the MobileNet transfer learning model, a computationally lightweight model has been proposed in this study. The optimal model for the diagnosis of tuberculosis has been determined after testing numerous variants on the base model with pre-trained weights. A computationally light transfer learning model is proposed to obtain the maximum overall accuracy of 98.66%. The improvement over the best existing model is quite significant. The transfer learning model (DenseNet) utilized in this existing model is based on a very complex convolutional neural network (CNN), and as a result, the model requires greater amounts of time. The performance of the other existing models is relatively less in comparison to our proposed model, and the methods have a number of other drawbacks. Our goal in this work is to create a more accurate model that requires less computational effort. When compared to previous models, our model has a very less number of trainable parameters, which causes the model to converge more quickly and predict more accurately. Our approach also has the benefit of being simply able to modify its weights when the system is further updated with new datasets. Additionally, because of its lightweight architecture, it can be installed on mobile devices as well as used in web-based applications with ease. To analyze and validate the proposed method, we have collected data from Kaggle and used MC, CHN and NIH datasets.

3DFRINet: A Framework for the Detection and Diagnosis of Fracture Related Infection in Low Extremities Based on 18F-FDG PET/CT 3D Images

Fracture related infection (FRI) is one of the most devastating complications after fracture surgery in the lower extremities, which can lead to extremely high morbidity and medical costs. Therefore, early comprehensive evaluation and accurate diagnosis of patients are critical for appropriate treatment, prevention of complications, and good prognosis. 18Fluoro-deoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) is one of the most commonly used medical imaging modalities for diagnosing FRI. With the development of deep learning, more neural networks have been proposed and become powerful computer-aided diagnosis tools in medical imaging. Therefore, a fully automated two-stage framework for FRI detection and diagnosis, 3DFRINet (Three Dimension FRI Network), is proposed for 18F-FDG PET/CT 3D imaging. The first stage can effectively extract and fuse the features of both modalities to accurately locate the lesion by the dual-branch design and attention module. The second stage reduces the dimensionality of the image by using the maximum intensity projection, which retains the effective features while reducing the computational effort and achieving excellent diagnostic performance. The diagnostic performance of lesions reached 91.55% accuracy, 0.9331 AUC, and 0.9250 F1 score. 3DFRINet has an advantage over six nuclear medicine experts in each classification metric. The statistical analysis shows that 3DFRINet is equivalent or superior to the primary nuclear medicine physicians and comparable to the senior nuclear medicine physicians. In conclusion, this study first proposed a method based on 18F-FDG PET/CT three-dimensional imaging for FRI location and diagnosis. This method shows superior lesion detection rate and diagnostic efficiency and therefore has good prospects for clinical application.

Multi-sensor data fusion-enabled lightweight convolutional double regularization contrast transformer for aerospace bearing small samples fault diagnosis

Aiming at the problems of low information utilization and lack of feature mining capability in multi-sensor fusion networks, this study presents a multi-sensor data fusion-enabled lightweight convolutional double regularization contrast transformer for aerospace bearing small samples fault diagnosis. Firstly, a metric termed integrated cliff entropy is devised to assign weights to vibration signals from diverse sensor channels. It aims to enhance the cyclic impulse characteristics within the fused signals, thereby facilitating more precise fault identification. Secondly, a lightweight Diwaveformer architecture is constructed as the backbone of contrast learning. It enables the global and local features of faulty signals to be comprehensively extracted with less computational effort. Finally, a double contrast loss is constructed to optimize the distribution of intra-class and inter-class features to improve the fault identification ability of the network with small samples. Additionally, a discard regularization method is designed to remove the projection head during the contrast learning process, further advancing the model lightweight. Our method achieved accuracies of 95.54% and 92.56% on two aerospace bearing datasets with extremely sparse training samples, which proved its superior performance.

EnTdecker - A Machine Learning-Based Platform for Guiding Substrate Discovery in Energy Transfer Catalysis.

Due to the magnitude of chemical space, the discovery of novel substrates in energy transfer (EnT) catalysis remains a daunting task. Experimental and computational strategies to identify compounds that successfully undergo EnT-mediated reactions are limited by their time and cost efficiency. To accelerate the discovery process in EnT catalysis, we herein present the EnTdecker platform, which facilitates the large-scale virtual screening of potential substrates using machine-learning (ML) based predictions of their excited state properties. To achieve this, a data set is created containing more than 34,000 molecules aiming to cover a vast fraction of synthetically relevant compound space for EnT catalysis. Using this data predictive models are trained, and their aptitude for an in-lab application is demonstrated by rediscovering successful substrates from literature as well as experimental validation through luminescence-based screening. By reducing the computational effort needed to obtain excited state properties, the EnTdecker platform represents a tool to efficiently guide substrate selection and increase the experimental success rate for EnT catalysis. Moreover, through an easy-to-use web application, EnTdecker is made publicly accessible under entdecker.uni-muenster.de.