Reduction In Computation Time Research Articles

Enhancing computational efficiency in topology-optimized mode converters via dynamic update rate strategies

In the big data era, mode division multiplexing, as a technology for extended channel capacity, demonstrates potential in enhancing parallel data processing capability. Consequently, developing a compact, high-performance mode converter through efficient design methods is an urgent requirement. However, traditional design methodologies for these converters face significant computational complexities and inefficiencies. Addressing this challenge, this paper introduces a novel topology optimization design method for mode converters employing a Dynamic Adjustment of Update Rate (DAUR). This approach markedly reduces computational overhead, accelerating the design process while ensuring high performance and compactness. As a proof-of-concept, an ultra-compact dual-mode converter was designed. The DAUR method demonstrated an 80% reduction in computational time compared to traditional methods, while maintaining a compact design (only 1.4 μm × 1.4 μm) and an insertion loss under 0.68 dB across a wavelength range of 1525 nm to 1575 nm. Meanwhile, simulated inter-mode crosstalk remained below − 24 dB across a 40 nm bandwidth. A comprehensive comparison with traditional inverse design algorithms is presented, demonstrating our method’s superior efficiency and effectiveness. Our findings suggest that DAUR not only streamlines the design process but also facilitates exploration into more complex micro-nano photonic structures with reduced resource investment.

Optimized multi-stage constant current fast charging protocol suppressing lithium plating for lithium-ion batteries using reduced order electrochemical-thermal-life model

Multi-stage Constant Current (MCC) is a state-of-the-art fast-charging protocol considering battery aging. It divides the charging process into multiple stages, each with a different current amplitude based on specific transition criteria, significantly influencing battery performance, such as charging time, degradation rate, and thermal effects. A key challenge in designing MCC protocols is addressing the lithium plating (LiP), which can accelerate degradation and pose a severe risk of thermal runaway. Since the LiP onset conditions vary between fresh and aged cells, this paper proposes an optimized MCC (O-MCC) charging protocol suppressing LiP based on the battery's state of health. To accurately simulate LiP conditions, different platforms of reduced-order electrochemical-thermal-life models are designed, compared, and optimized for speed and accuracy using a genetic algorithm, resulting in a 36.4 % reduction in computational time while maintaining the accuracy of the Pseudo Two-Dimensional model. The Nonlinear Model Predictive Control algorithm is then used to optimize the MCC protocol, minimizing charging time while preventing LiP throughout life. Experimental results show that O-MCC reduces charging time by 11.7 % and capacity loss by 59.4 %, enhancing battery safety. Additionally, O-MCCs with varying constraints are developed to meet specific demands and simulated at the battery pack level.

Rapid Prediction and Parameter Evaluation of Process-Induced Deformation in L-Shape Structures Based on Feature Selection and Artificial Neural Networks

The process-induced deformation (PID) during the manufacturing of thermosetting composite materials can significantly compromise manufacturing precision. This paper introduces an innovative method that combines a finite element analysis (FEA), feature classification algorithms, and an Artificial Neural Network (ANN) framework to rapidly predict the PID of a typical L-shaped structure. Initially, a comprehensive range of parameters that influence PID are compiled in this research, followed by the generation of a dataset through FEA considering viscoelastic constitutive models, validated by experimental results. Influential parameters are classified using Random Forest and LASSO regression methods, with each parameter rated according to its impact on PID, delineating their varying degrees of importance. Subsequently, through a hyperparameter analysis, an ANN framework is developed to rapidly predict the PID, while also refining the assessment of the parameters’ significance. This innovative approach achieves a computational time reduction of 98% with less than a 5% loss in accuracy, and highlights that under limited computational conditions, considering only a subset or all of the parameters—the peak temperature, corner angle, coefficient of chemical shrinkage, coefficient of thermal expansion, curing pressure, and E1—minimizes accuracy loss. The study demonstrates that machine learning algorithms can effectively address the challenge of predicting composite material PID, providing valuable insights for practical manufacturing applications.

Dynamic modeling of post-combustion carbon capture process based on multi-gate mixture-of-experts incorporating dual-stage attention-based encoder-decoder network

Solvent-based post-combustion carbon capture (PCC) technology is a promising, near-term solution for decarbonizing power generation and industrial facilities. Model-based process simulation is crucial for the optimal design and operation of the PCC process. Recently, data-driven models have gained attention due to their adaptability, efficient computation and high accuracy. However, the nonlinearity, strong couplings and multi-time scale features of the PCC process pose significant challenges for model identification. To this end, this paper proposes a multi-gate mixture-of-experts incorporating dual-stage attention-based encoder-decoder (MMoE-DAED) network for dynamic modeling of the PCC process under wide operating conditions. An encoder-decoder composed of long short-term memory (LSTM) network is employed to extract features from the time-dependent input data and learn the complex dynamic interactions caused by the inertial and delay properties of the process. Dual-stage attention mechanism is incorporated into the encoder and decoder respectively to select the most relevant input features and their correlations within the time series data. To enhance multi-output prediction accuracy, multi-gate mixture-of-experts (MMoE) framework that considers correlations of multitask learning is implemented. Simulation results using operating data from a PCC experimental setup indicate that the proposed modeling approach accurately predicts the steady-state values and dynamic trends of the CO2 capture rate and stripper bottom temperature over a wide operating range. The RMSE, MAPE and R2 indices for the CO2 capture rate are 2.1592, 0.0295, 0.9641, respectively, and for the stripper bottom temperature are 0.1491, 0.0003, 0.9833, respectively. Validations on a PCC simulator further verify the accuracy and efficiency of the MMoE-DAED model, which enables an 80.87% reduction in computation time compared to the simulator. This paper points to a new direction for the data-driven dynamic modeling of complex energy conversion processes.

Integrated top-down process and voxel-based microstructure modeling for Ti-6Al-4 V in laser wire direct energy deposition process

Laser-wire metal additive manufacturing (AM) is one of the ideal direct energy deposition (DED) processes for creating large-scale parts with a medium level of complexity. However, the DED process involves complex thermal signatures and wide length scales making the fabrication of realistic AM components and part qualification often reliant on experimental trial-and-error optimization. While experimental measurements over the full volume of a part are valuable and necessary, measuring the entire area of a part is significantly laborious and practically infeasible, particularly for large parts in terms of cost and rapid qualification. Therefore, in this work, we developed an effective thermal and microstructure modeling framework based on the Johnson–Mehl-Avrami-Kolmogorov (JMAK) and Koistinen & Marburger (KM) models through a top-down approach that considers plate distortion-affected thermal profiles. A voxel-by-voxel simulation method is used to predict individual phase fractions of Ti-6Al-4 V. The predicted results were validated through detailed metallurgical measurements. A combined voxel-by-voxel approach with a sparse data reconstruction technique produced a near-perfect reconstruction of the original data. This approach anticipates a significant reduction in data points and computation time and resources. Lastly, we conclude with potential extensions of this work to other modeling efforts.

A cohesive fracture-enhanced phase-field approach for modeling the damage behavior of steel fiber-reinforced concrete

This study introduces a novel computational framework based on the phase-field fracture/damage model (PFM), providing rapid and accurate predictions of the damage behavior of Steel Fiber Reinforced Concrete (SFRC) material. The framework integrates the interfacial cohesive fracture model with the PFM to depict the interaction between the cementitious matrix and fibers, marking the first demonstration that this model adequately captures both local fracture phenomena between fibers and concrete (such as fiber bridging and interfacial debonding) and mesoscopic stress–strain behavior. Additionally, an enhanced geometric approximation is established based on the spatial distribution density function of fibers and particles, allowing for the transformation of the three-dimensions (3-D) fiber-reinforced material structure into an equivalent two-dimensions (2-D) material. This approach enables the reduction of computational time, a significant limitation of the phase-field method, thereby allowing an in-house code to perform various computations with complex material structures such as SFRC. The effectiveness of the approach is demonstrated through four examples involving variations in the microgeometry and material properties of SFRC structures, ranging from the simplest configurations to real experimental material structures. Extensive Monte Carlo simulations of the model are also employed to provide results closer to reality, which are then compared with recent experimental data and numerical models, demonstrating the efficacy of the proposed computational model.

Color Dominance-Based Polynomial Optimization Segmentation for Identifying Tomato Leaves and Fruits

Optimization processes or methods play an essential role in the continuous improvement of various human activities, particularly in agriculture, given its vital role in food production. In precision agriculture, which utilizes technology to optimize food production, a primary goal is to minimize the consumption of resources like water, fertilizers, and the detection of pests and diseases. In the fertilization process, it is essential to identify any deficiencies or excesses of chemical elements. Nutrient deficiencies, which are essential for plant development, are typically detected in the leaves of crops. This paper proposes a methodology for optimizing the color threshold dominance factors employed in the segmentation process for tomato crop leaves and fruits. The optimization is performed using an interpolation method to find the values that maximize the segmentation of leaves and fruits used by the color dominance segmentation method. A comparison of the interpolation method results with those obtained using a greedy algorithm, which iteratively finds the optimal segmentation values, shows nearly identical outcomes. Similarly, a UNetmodel is used for semantic segmentation, the results of which are inferior to those obtained by the proposed interpolation optimization method. The most significant contribution of the interpolation method is that it requires only a single iteration to generate the initial data, in contrast to the iterative search required by the greedy algorithm and the lengthy training process and video card dependency of the UNet model. This results in an 80% reduction in computation time.

A simulation study towards sparse-view spectral X-ray CT segmentation using spectrum-aware deep neural networks

ABSTRACT Spectral X-ray computed tomography (SXCT) is able to provide richer spectral information than conventional X-ray CT, thereby improving the capability of discriminating materials in segmentation tasks. However, segmentation algorithms for conventional CT images often lose their performance in material segmentation with SXCT images due to the spectrum-dependent attenuation properties of X-rays. Recently, deep neural networks (DNNs) have demonstrated their effectiveness in segmentation tasks. However, processing SXCT with DNNs requires significant memory resources and may take significant computation time. To address these issues, this paper introduces a simulation study towards realising deep-learning-based SXCT image segmentation. The proposed method employs a convolutional neural network (CNN) that selectively utilises the significant spectrum channels for accurate material segmentation. The proposed method builds upon a prior study on adaptive CT reconstruction, but to further enhance the performance, we introduce a self-attention-based spectrum-selection module. During training, the selection module assigns importance weights to spectrum channels based on the attention scores. Once the network has been trained, the proposed method can obtain satisfactory segmentation results from projection images at one or a combination of several spectrum channels recommended by the selection module. The experimental results demonstrate that the proposed method achieves a significant reduction in computation time while maintaining the segmentation quality.

Robust global optimization of plasmonic filter based on Q factor analysis using graded index

Abstract To reduce the impact of uncertainties in various manufacturing processes while maintaining satisfactory performance level, we propose a robust global optimization approach for plasmonic filters based on extraordinary optical transmission by use of Q factor analysis with reasonable computational cost. The gradient index is employed as a metric for robustness, and the multiobjective genetic algorithm is selected as a global optimization tool. The figures of merit and gradient index are chosen as two objective functions with the design variables being the slit width, slit height, and period of the slit array, respectively. The finite element method is employed to investigate a variety of optical properties rigorously. The numerical optimization results demonstrate that the proposed approach provides improved effectiveness and efficiency in designing plasmonic filters under fabrication uncertainties by using Q factor analysis. The method proposed in this study enables systematic robust optimization while accounting for manufacturing uncertainties, which would be infeasible under typical time constraints. Specifically, within the proposed optimization framework, Q factor analysis can be employed due to a 20.60-times reduction in computational time. To the best of our knowledge, this represents the first robust optimization design for enhanced filter performance based on Q factor analysis in EOT, necessitating comprehensive spectrum calculations alongside fabrication tolerance considerations.

Acceleration of Numerical Modeling of Uranium In Situ Leaching: Application of IDW Interpolation and Neural Networks for Solving the Hydraulic Head Equation

The application of In Situ Leaching (ISL) has significantly boosted uranium production in countries like Kazakhstan. Given that hydrodynamic and chemical processes occur underground, mining enterprises worldwide have developed models of reactive transport. However, modeling these complex processes demands considerable computational resources. This issue is particularly significant in the context of numerical analyses of mining processes or when modeling production scenarios in uranium mining by the ISL technique, given that a substantial portion of computational resources is allocated to solving the hydraulic head equation. This work aims to explore the applicability of PINNs to accelerate hydrodynamic simulations of the ISL process. The solution of the Poisson equation is accelerated by generating an initial approximation for the iterative method through the application of the Inverse Distance Weighting (IDW) interpolation and PINNs. The impact of various factors, including the computational grid and the spacing between wells, on both the accuracy and efficiency of initial approximation and the overall solution of the elliptic equation are explored. Employing the hydraulic head distribution obtained through PINNs as the initial approximation led to a significant reduction in computation time and a decrease in the number of iterations by a factor of 2.8 to 7.10.

An efficient conditional GAN-based framework for high-resolution prediction of tyre-pavement contact stresses – a contribution towards a digital twin of the road system

ABSTRACT The high-resolution prediction of tyre-pavement contact stresses is crucial for advancing pavement design and vehicle safety. A significant knowledge gap exists in accurately and quickly predicting tyre-pavement contact stresses, especially under varying conditions. Addressing this challenge, this study is part of the concerted effort within the project CRC/TRR 339, contributing to the series ‘A Contribution towards a Digital Twin of the Road System’. It aims to bridge this gap by introducing a novel Multimodal Conditional Deep Convolutional Generative Adversarial Network (MC-DCGAN) model for forecasting tyre-pavement contact stresses, offering an efficient and accurate method. The incorporation of a multimodal conditional module within the GAN-based framework allows for controllable outputs tailored to specific tyre types, inflation pressures, and loads. Leveraging the DCGAN architecture significantly enhances prediction accuracy. Experimental validation confirms the method’s high accuracy, strong generalisation capabilities, and a substantial reduction in computational time – nearly two orders of magnitude lower compared to traditional Finite Element simulations. This research not only fills a critical void in the high-precision prediction of tyre-pavement contact stresses, but also holds significant implications for optimising pavement design and enhancing road traffic safety, marking a significant step towards realising a digital twin of the road system.

A fast quadrupole-based analytical model for solving transient heat transfer in ventilated double-skin walls and assessing their energy saving potential

This research proposes a fast and accurate method for modeling transient heat transfer in ventilated double-skin façades (VDFs) with forced ventilation to predict their energy-saving potential. Using the thermal quadrupole formalism, the method solves the VDF problem in the Laplace domain while considering the full transient nature of the heat transfer; the only approximation is in space. The solution involves obtaining a general transfer function that predicts heat transfer rates or air temperatures at ventilated cavity's exit. The quadrupole method is validated against computational fluid dynamic numerical simulations conducted under similar meteorological, design and boundary conditions. A very good agreement was found (less than 3 % average deviation) with a significant reduction in computation time (less than 3s against 6h for CFD calculations). The model is computationally efficient and can consider important factors such as the opacity of the walls, construction materials, and air cavity design parameters. Finally, the model allows the assessment of the energy-saving potential of VDFs under various scenarios compared to conventional systems, which helps contributing to more sustainable building design practices. The energy efficiency of a VDF configuration was compared against a conventional wall system. Energy savings of 8.4 and 5.2 % were obtained, in cold and hot climates respectively.

Fast neural network inverse model to maximize throughput in ultra-wideband WDM systems

Ultra-wideband systems expand the optical bandwidth in wavelength-division multiplexed (WDM) systems to provide increased capacity using the existing fiber infrastructure. In ultra-wideband transmission, power is transferred from shorter-wavelength WDM channels to longer-wavelength WDM channels due to inelastic inter-channel stimulated Raman scattering. Thus, managing launch power is necessary to improve the overall data throughput. While the launch power optimization problem can be solved by the particle swarm optimization method it is sensitive to the objective value and requires intensive objective calculations. Hence, we first propose a fast and accurate data-driven deep neural network-based physical layer in this paper which can achieve 99%−100% throughput compared to the semi-analytical approach with more than 2 orders of magnitude improvement in computational time. To further reduce the computational time, we propose an iterative greedy algorithm enabled by the inverse model to well approximate a sub-optimal solution with less than 6% performance degradation but almost 3 orders of magnitude reduction in computational time.

Efficient storage and regression computation for population-scale genome sequencing studies.

In the era of big data in human genetics, large-scale biobanks aggregating genetic data from diverse populations have emerged as important for advancing our understanding of human health and disease. However, the computational and storage demands of whole genome sequencing (WGS) studies pose significant challenges, especially for researchers from underfunded institutions or developing countries, creating a disparity in research capabilities. We introduce new approaches that significantly enhance computational efficiency and reduce data storage requirements for WGS studies. By developing algorithms for compressed storage of genetic data, focusing particularly on optimising the representation of rare variants, and designing regression methods tailored for the scale and complexity of WGS data, we significantly lower computational and storage costs. We integrate our approach into PLINK 2.0. The implementation demonstrates considerable reductions in storage space and computational time without compromising analytical accuracy, as evidenced by the application to the AllofUs project data. We optimized the runtime of an exome-wide association analysis involving 19.4 million variants and the body mass index phenotype of 125,077 individuals, reducing it from 695.35 minutes (approximately 11.5 hours) on a single machine to just 1.57 minutes using 30 GB of memory and 50 threads (or 8.67 minutes with 4 threads). Additionally, we extended this approach to support multi-phenotype analyses. We anticipate that our approach will enable researchers across the globe to unlock the potential of population biobanks, accelerating the pace of discoveries that can improve our understanding of human health and disease.

Energy-Preserving Perturbation Method for Thermally Induced Dynamics in Aerospace Tensegrity Structures

Tensegrity structures are emerging as pivotal components in the assembly of ultralarge modular spacecraft in orbit, primarily due to their exceptional volume-to-mass ratio and robust controllability. During on-orbit operations, given the low damping and high flexibility of tensegrity, its vibration-induced deformations affect the magnitude of the heat flux, presenting a typical force–shape coupling problem. This paper introduces an energy-preserving matrix perturbation solution designed to efficiently and accurately address the thermally induced vibration challenges in tensegrity. This method effectively mitigates the energy dissipation and numerical instability issues. Furthermore, the analysis incorporates the impact of the axial component of heat flux on structural vibration, uncovering the influence mechanism of this often-overlooked slow variable on the dynamic behavior of tensegrity. The proposed method demonstrates a deviation of less than 1% between the thermal response and the results obtained from finite element analysis, while achieving a 70% reduction in computation time. The varying axial thermal load leads to a rapid decrease in system frequency, resulting in divergent thermal vibrations. Additionally, the significant dynamic stress can potentially surpass the structural stress limits. These findings underscore the critical importance of considering such effects in the design and calculation of tensegrity structures for aerospace applications.

A non-iterative compensation method for machining errors of thin-walled parts considering coupling effect of tool-workpiece deformation

Thin-walled parts have significant application value in the aerospace industry due to their high strength-to-weight ratio. However, their low rigidity makes them susceptible to cutting force-induced error, which may seriously affect their machining accuracy. Error compensation is an effective method for addressing this issue. However, the commonly used mirror compensation method can cause residual errors due to the coupling effect of the tool-workpiece deformation. The influence mechanism of this coupling effect on error compensation is analyzed through an iterative compensation method. This method efficiently reduces residual errors. However, its computation efficiency is insufficient to meet the requirements of real-time compensation. Therefore, a non-iterative compensation method is proposed to directly calculate the compensation values considering the coupling effect of the tool-workpiece deformation. Through the approximate invariance of the overall cutting coefficient matrix and the pre-given system parameter, the proposed method avoids the repeated calculation of cutting forces and improves the computation efficiency. Experiment results of milling thin-walled blades show that after compensation using the proposed method, the machining accuracy of the thin-walled blade has seen a further increase of 18.1% in comparison to the mirror compensation method. Moreover, the proposed method achieves comparable compensation accuracy to the iterative method with a 66% reduction in computation time. The proposed method has significant potential for real-time compensation in the machining of complex 5-axis thin-walled parts.

A machine-learning-based column generation heuristic for electric bus scheduling

Bus scheduling in public transit consists in determining a set of bus schedules to cover a set of timetabled trips at minimum cost. This planning process has evolved recently with the advent of electric buses that introduce constraints related to vehicle autonomy and battery charging process. In particular, column-generation algorithms have regained popularity for solving problems similar to the one considered in this paper, namely, the multi-depot electric vehicle scheduling problem (MDEVSP) with a piecewise linear charging function and capacitated charging stations. To tackle large-scale MDEVSP instances, we design a column generation (CG) heuristic that relies on reduced-sized networks to generate the bus schedules. The reduction is achieved by selecting a priori a subset of the arcs. Multiple selection techniques are studied: some are based on a greedy heuristic and others exploit a supervised learning algorithm relying on a graph neural network. It turns out that combining both selection types yields the best computational results. On 405 artificial instances involving between 568 and 1474 trips and generated from real bus lines in Montreal, the network reduction technique produced an average computational time reduction of 71.6% (compared to the same CG heuristic but without network reduction), while deteriorating solution cost by an average of 2.2%. On 8 larger instances containing more than 2500 trips on average, the proposed solution method also provided an average time saving of 52.5% with an average gap of 4.2% thanks to a transfer learning approach.

RDAG U-Net: An Advanced AI Model for Efficient and Accurate CT Scan Analysis of SARS-CoV-2 Pneumonia Lesions.

Background/Objective: This study aims to utilize advanced artificial intelligence (AI) image recog-nition technologies to establish a robust system for identifying features in lung computed tomog-raphy (CT) scans, thereby detecting respiratory infections such as SARS-CoV-2 pneumonia. Spe-cifically, the research focuses on developing a new model called Residual-Dense-Attention Gates U-Net (RDAG U-Net) to improve accuracy and efficiency in identification. Methods: This study employed Attention U-Net, Attention Res U-Net, and the newly developed RDAG U-Net model. RDAG U-Net extends the U-Net architecture by incorporating ResBlock and DenseBlock modules in the encoder to retain training parameters and reduce computation time. The training dataset in-cludes 3,520 CT scans from an open database, augmented to 10,560 samples through data en-hancement techniques. The research also focused on optimizing convolutional architectures, image preprocessing, interpolation methods, data management, and extensive fine-tuning of training parameters and neural network modules. Result: The RDAG U-Net model achieved an outstanding accuracy of 93.29% in identifying pulmonary lesions, with a 45% reduction in computation time compared to other models. The study demonstrated that RDAG U-Net performed stably during training and exhibited good generalization capability by evaluating loss values, model-predicted lesion annotations, and validation-epoch curves. Furthermore, using ITK-Snap to convert 2D pre-dictions into 3D lung and lesion segmentation models, the results delineated lesion contours, en-hancing interpretability. Conclusion: The RDAG U-Net model showed significant improvements in accuracy and efficiency in the analysis of CT images for SARS-CoV-2 pneumonia, achieving a 93.29% recognition accuracy and reducing computation time by 45% compared to other models. These results indicate the potential of the RDAG U-Net model in clinical applications, as it can accelerate the detection of pulmonary lesions and effectively enhance diagnostic accuracy. Additionally, the 2D and 3D visualization results allow physicians to understand lesions' morphology and distribution better, strengthening decision support capabilities and providing valuable medical diagnosis and treatment planning tools.

Backward induction-based deep image search.

Conditional image retrieval (CIR), which involves retrieving images by a query image along with user-specified conditions, is essential in computer vision research for efficient image search and automated image analysis. The existing approaches, such as composed image retrieval (CoIR) methods, have been actively studied. However, these methods face challenges as they require either a triplet dataset or richly annotated image-text pairs, which are expensive to obtain. In this work, we demonstrate that CIR at the image-level concept can be achieved using an inverse mapping approach that explores the model's inductive knowledge. Our proposed CIR method, called Backward Search, updates the query embedding to conform to the condition. Specifically, the embedding of the query image is updated by predicting the probability of the label and minimizing the difference from the condition label. This enables CIR with image-level concepts while preserving the context of the query. In this paper, we introduce the Backward Search method that enables single and multi-conditional image retrieval. Moreover, we efficiently reduce the computation time by distilling the knowledge. We conduct experiments using the WikiArt, aPY, and CUB benchmark datasets. The proposed method achieves an average mAP@10 of 0.541 on the datasets, demonstrating a marked improvement compared to the CoIR methods in our comparative experiments. Furthermore, by employing knowledge distillation with the Backward Search model as the teacher, the student model achieves a significant reduction in computation time, up to 160 times faster with only a slight decrease in performance. The implementation of our method is available at the following URL: https://github.com/dhlee-work/BackwardSearch.

Uncertain seismic response and robustness analysis of isolated continuous girder bridges

When calculating the seismic response of structures, conventional methods often neglect uncertainty and suffer from excessive computation. Although robustness metrics have been effective in evaluating structural safety, they are not suitable for isolated continuous girder bridges (ICGB). To address this issue, this paper introduces an uncertain seismic response and robustness assessment method tailored to ICGB: Firstly, the method considers uncertainty in material parameters and ground motion by constructing a data set using Monte Carlo simulation. Secondly, a genetic algorithm for parallel optimisation of radial basis function neural network (GAPO-RBFNN) is trained to predict the seismic responses of the bridges. Thirdly, a robustness metric specifically designed for bridges is derived and applied to bridge. Finally, the proposed robustness metric is validated on lead rubber bearing (LRB) and shape memory alloy-lead rubber bearing (SMA-LRB) bridges. Results indicate that the GAPO-RBFNN model accurately approximates the mapping relationship between structural parameters and seismic response, with an average error of only 0.32% and a 70% reduction in computation time compared with traditional methods. Moreover, the new robustness metric overcomes the limitations of the dismantled member method and is applicable to bridge. The robustness of the SMA-LRB-ICGB, strengthened by SMA, is improved, providing evidence of the effectiveness of the proposed robustness metric.