Search Results (62)

Evacuated-tube solar collectors charged with hybrid nanofluids can raise thermal output, but their coupled thermal and hydraulic response depends on many interacting variables, making the design an optimisation rather than a prediction problem. This study recasts it as a constrained, mixed-variable, three-objective task that maximises thermal efficiency and the Nusselt number while minimising pumping power over the hybrid pair, base fluid, weight fraction, component-one share and flow rate, and develops a memetic metaheuristic: the Adaptive Memetic Hybrid (AMH). A histogram gradient-boosted surrogate trained on 54,432 reduced-order runs, with held-out coefficients of determination of at least 0.9999, provides a fast screen, while a continuous reduced-order model validated to within 0.02 percent serves as the objective; the surrogate is accurate off-grid for efficiency but not for pumping power or the Nusselt number. Nine optimisers, comprising four baselines, three recent metaheuristics, and two AMH variants, were validated on twelve ZDT, DTLZ, and constrained problems over thirty trials using the hypervolume, generational distances, and spacing, and analysed with Friedman, Nemenyi, and Holm-corrected Wilcoxon tests. AMH attained the best mean Friedman rank of 3.08 (chi-square 65.7, p = 3.6 × 10⁻¹¹), significantly outperforming the recent methods and NSGA-III and remaining competitive with the strongest classical algorithms. On the collector, the reduced-order front recovers the 1512-design brute-force maximum efficiency to within 0.02 percent and improves the trade-off through continuous flow rates. The study is a deterministic, model-based optimisation process: the surrogate serves as a tool for fast screening and diagnostics, while the reconstructed reduced-order model is the objective for the final continuous optimisation. The collector application has a low effective design dimension, being governed mainly by the base fluid and the loop flow rate, so the decisive separation of the algorithms is established on the benchmark suite rather than on the collector. Experimental validation remains a task for future work. Full article

Applied engineering studies that use evolutionary multi-objective optimisation (MOO) rarely report confidence bounds on the Pareto-optimal solutions they recommend. This paper presents a four-stage sensitivity-analysis protocol that supplies them: cross-algorithm benchmarking, Friedman and Bonferroni-corrected Wilcoxon testing, hyperparameter sensitivity analysis, and ISO/IEC GUM-compliant uncertainty propagation. The protocol is first validated on the ZDT1, ZDT3 and DTLZ2 benchmarks (n = 2–12 decision variables), then demonstrated on an enclosed screw-conveyor design using Discrete Element Method (DEM) surrogates built by Response Surface Methodology (RSM). On the benchmarks, it correctly identified both algorithmic equivalence (MOGA and NSGA-II indistinguishable on four of five instances) and MOEA/D’s characteristic weakness on the disconnected ZDT3 front, whose hypervolume degraded most with dimensionality. For the engineering case, MOGA matched NSGA-II (p = 0.47–0.79) and remained within 0.5% hypervolume of SMS-EMOA over 12 runs, while hyperparameter variation stayed below 1% (max CV = 0.962%). DEM achieved 8.2% mean absolute percentage error for mass flow rate across 42 CCD operating conditions. The MOGA-optimised 100 mm pitch (4.86°, 138.86 rpm) delivered 0.416 kg/s at 6.95 N·m, a specific energy consumption of 0.0675 kWh/tonne and a 63.0% reduction against the 75 mm baseline. Full article

This study presents an integrated computational framework combining Computational Fluid Dynamics (CFD), experimental validation, and surrogate modelling to analyse and optimise the thermal performance, thermodynamic stability, and energy efficiency of a refrigerated truck compartment. CFD simulations were conducted to investigate airflow distribution and temperature uniformity under operating temperatures ranging from 0 to 5 °C. The results showed that airflow circulation was primarily governed by the evaporator outlet, while recirculation zones enhanced air mixing but were insufficient to completely eliminate localised hotspots. Increasing the operating temperature from 0 to 5 °C resulted in a rise in the maximum compartment temperature from 6.26 to 10.01 °C. Thermodynamic stability analysis revealed that operation within the 3–5 °C range provided more stable thermal conditions due to reduced refrigeration load and improved temperature uniformity. Experimental measurements of airflow velocity and evaporator surface temperature demonstrated good agreement with CFD predictions, confirming the reliability of the numerical model. Furthermore, CFD-based optimisation reduced the electrical energy consumption of the eTRU system by 10.0%, decreasing the energy intensity from 0.117 to 0.105 kWh km⁻¹, while maintaining improved thermal stability throughout the refrigerated compartment. The surrogate model achieved excellent predictive performance with R² = 0.9576 and RMSE = 0.0386, demonstrating its suitability for rapid optimisation of refrigerated transport systems. The proposed framework offers an effective tool for improving thermal management, enhancing energy efficiency, and supporting the development of sustainable refrigerated transport systems. Full article

Hybrid nanofluids raise the thermal output of evacuated-tube solar collectors, but the heat-transfer gain comes at the cost of friction and pumping power, so the working fluid must be chosen in line with a constrained thermo-hydraulic criterion. Its design is posed here as a constrained single-objective optimisation over the hybrid pair, base fluid, weight fraction, component share and flow rate. A variance-based screening of a 54,432-run full-factorial dataset reduces the design space from eight variables to five, justified by the invariance of the optimal hybrid pair across thirty-six operating points. Gradient-boosted surrogates for the four state and constraint responses (efficiency, pumping power, Reynolds number and outlet temperature) reproduce the simulator to held-out coefficients of determination of 0.9998–1.0000; the performance-criterion surrogate has a lower global coefficient of determination (0.46) and is assessed by top-region ranking accuracy. The reduced problem is solved with a selection hyper-heuristic over twelve operators, pairing an upper-confidence-bound selector with late-acceptance hill-climbing. Under Friedman-, Nemenyi- and Holm-corrected Wilcoxon testing, it ranks within the leading statistically indistinguishable group, but is not separable from a random-selection ablation, locating its value in robustness rather than adaptivity; furthermore, it attains the grid-reference optimum within 0.40 percent. The criterion optimum is an Al₂O₃-Cu suspension in ethylene-glycol and water at three percent loading; the constrained-efficiency optimum reaches a surrogate thermal efficiency of 0.760, cross-checked against a grid-reference value of 0.762 and a reconstructed-model value of 0.760, conditional on the supplied reduced-order model and its single-tube dataset convention. The constrained-efficiency objective is treated as the primary design objective, while the performance criterion is reported as a secondary screening index that characterises the additive rather than selecting the operating fluid. Full article

Mechanistic simulation and machine learning are powerful but complementary tools: physics-based simulation is interpretable yet computationally expensive and blind to real-world context, whereas machine learning is fast but data-hungry and opaque. Biological systems resolve this tension elegantly, coupling physically grounded mechanoreceptor sensing with higher-level neural interpretation that places those signals in context. Inspired by this layered architecture, we present Mechanics-AI, an open-source framework that mirrors the same sensing-then-interpretation logic computationally. A first learning layer (ML1) emulates expensive finite-element, computational fluid dynamics and multiphysics simulations to produce interpretable physical metrics such as stress, strain, shear, and thermal and moisture fields, while a second layer (ML2) fuses these metrics with heterogeneous real-world metadata to predict categorical outcomes and design recommendations. Eight algorithms are benchmarked automatically, the most accurate is selected for each task, and Shapley additive explanations expose the dominant physical drivers to preserve interpretability. The framework is demonstrated across three independent domains using a single unchanged pipeline: forensic traumatic brain injury prediction, optimisation of a bio-inspired humanoid bioreactor for tissue engineering, and a zero-emission building (ZEBAI) framework that couples thermo-hygro-mechanical simulation with Sobol-sampled surrogate modelling to design sustainable, low-carbon envelopes from recycled aggregate concrete by balancing structural safety, energy and embodied carbon. Despite entirely different physics, data and objectives, the same architecture generalises across all three, showing that bio-inspired, layered coupling of mechanistic simulation and contextual learning offers a reusable, interpretable route to cross-domain engineering prediction and sustainable design. Full article

This study optimised the thermal performance of an air-cooled weapon turret control computer used in military missions by integrating experimental, numerical, and machine learning methods. To address thermal localisation and heat accumulation in high-power-density electronic components, two heat pipes with high effective thermal conductivity were embedded in the heat sink block. Based on numerical predictions, this configuration yielded an 8% thermal enhancement; however, its experimental verification remains a subject for future work. The accuracy of the three-dimensional Computational Fluid Dynamics model was validated within an acceptable error tolerance using experimental data from a physical prototype. To reduce the high computational cost of the parametric design space, Gaussian Process Regression, notable for its probabilistic nature, was employed as a surrogate model instead of traditional artificial neural networks, which tend to overfit small-scale deterministic data. The GPR model successfully mapped the system variance, demonstrating an extremely high coefficient of determination and a minimal margin of error in predicting the maximum chip temperature and system pressure drop. Cross-validation analyses conclusively demonstrated that the model has high generalisation capability without overfitting the dataset. Multi-objective Pareto optimisation was conducted by scanning a dense design grid generated over the trained continuous surrogate model. The optimal balanced design configuration identified along the Pareto front maintained the chip temperature within the safe zone, below the critical operating limit, while significantly reducing aerodynamic resistance on the fan, energy consumption, and noise issues, without compromising the system’s thermal performance. The developed GPR-based methodology offers a stable and reliable optimisation framework that minimises trial-and-error costs in the design of military thermal management systems. Full article

With increasing energy demand and environmental emission concerns, biodiesels have emerged as a promising alternative to traditional fossil diesel. This study investigates the effects of soybean (Glycine max) biodiesel–diesel blends containing 0–80% biodiesel on the performance, combustion, and emission characteristics of a diesel engine. The performance parameters evaluated include brake specific fuel consumption (BSFC), brake mean effective pressure (BMEP), brake power (BP), and brake thermal efficiency (BTE). Combustion characteristics are assessed through in-cylinder pressure, heat release rate (HRR), and cumulative heat release fractions (HRFs), while emission characteristics are analysed in terms of smoke opacity and nitrogen oxides (NOx) emissions. The primary objective of this study is to evaluate the impact of high-percentage biodiesel and fuel injection timing on overall engine characteristics. A one-dimensional (1D) model was developed using Diesel-RK (Version: 4.30.189) and GT-Suite software (Version: GT-ISE 2026). The models were calibrated and validated through model to model and further validated against experimental data. Using the simulation generated dataset, a machine learning-based surrogate prediction model was developed to rapidly estimate engine performance and emission characteristics directly from engine operating conditions and biodiesel blend ratios, eliminating the need for repeated Diesel-RK simulations. Multiple regression algorithms were evaluated, with the XGBoost tree ensemble model providing the highest prediction accuracy across the investigated performance parameters. Moreover, the Diesel-RK data were validated with experimental data. The results demonstrated model-to-model data and between model to experimental data, confirming the reliability of the developed models. The proposed Diesel-RK simulation and machine learning framework enables rapid and accurate prediction of engine performance and emissions, providing an efficient tool for evaluating biodiesel blends and supporting future engine design and optimisation. Full article

This study addresses the inverse design of carbon fibre-reinforced polymer laminated plates with prescribed natural frequencies. The problem is difficult because stacking sequences are discrete, the design space is large, and multiple layups may produce nearly identical frequency spectra. This study does not seek to introduce a new tandem-network architecture. Rather, it adapts the established tandem inverse-design strategy to the discrete and non-unique vibration design of carbon fibre-reinforced polymer laminated plates. In the proposed framework, a trainable inverse network is coupled to a pre-trained forward frequency surrogate, allowing the inverse model to be optimised through frequency reconstruction instead of direct ply-angle supervision. A dataset of 50,000 symmetric CFRP laminates is generated using Classical Laminate Theory and a Rayleigh–Ritz vibration solver, covering four boundary conditions and a range of plate geometries. The forward model achieves R² values above 0.99 and mean absolute percentage errors below 3% for the first five natural frequencies. Compared with a genetic algorithm, the proposed inverse model provides stacking sequences about 7000 times faster while producing multiple feasible designs for each target. Permutation sensitivity analysis shows that plate geometry has the strongest influence on the frequency response, followed by boundary condition and ply orientation. Four engineering cases confirm the method’s usefulness for vibration isolation, frequency-gap control, and multi-mode frequency prescription. The principal contribution is the integration of multi-boundary-condition vibration modelling, discrete stacking-sequence inverse design, response-based treatment of non-uniqueness, speed/diversity benchmarking, and sensitivity-based physical interpretation within a single composite-laminate design framework. Full article

Background: Breast cancer remains a leading cause of cancer-related mortality, and reliable computational decision support is increasingly viewed as a complement to expert pathological assessment rather than a replacement for it. Variational quantum classifiers (VQCs) and Quantum Support Vector Machines (QSVMs) have recently been promoted as candidate models for medical classification, yet most published comparisons rely on internal hold-out validation alone and report only a single point estimate of discrimination, omitting calibration, decision-analytic value, and explainability—three ingredients that any clinically credible model must furnish. Methods: We assembled a complete quantum–classical machine learning pipeline and evaluated it under a deliberately stringent protocol designed to expose, rather than conceal, the limitations of current Noisy Intermediate-Scale Quantum (NISQ)-era models. The analytical hypothesis was conservative and stated in advance; in light of saturated classical baselines on this benchmark, we did not anticipate a quantum advantage in raw discrimination, and we framed the study as a methodological probe rather than as a competition. Using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset (n = 569) for development and an independent Wisconsin Original (WBC) cohort (n = 683) for external validation, we benchmarked five classical learners (XGBoost, LightGBM, CatBoost, RandomForest, RBF-SVM), two quantum models (an eight-qubit VQC implemented in PennyLane and a ZZ-feature-map QSVM implemented in Qiskit), and a stacked hybrid ensemble. The evaluation framework combined Optuna-driven hyperparameter optimisation, internal–external cross-validation, and external validation on the independent WBC cohort. Robustness and interpretability were then probed through circuit depth and embedding rotation ablation, depolarising noise stress tests, learning curve and feature stability analysis, decision curve analysis, and dual SHAP-based explanations covering both a direct tree-based explanation and a quantum surrogate. Reporting followed the TRIPOD + AI guideline. Results: On the internal test partition, RBF-SVM achieved the highest discrimination (AUC = 0.998), with XGBoost, LightGBM, CatBoost, the hybrid ensemble, and the VQC clustering between 0.992 and 0.996; the QSVM with a ZZ-fidelity kernel underperformed substantially (AUC = 0.727). Pairwise tests for correlated ROC curves indicated that most differences among top models were not statistically significant. On the external WBC cohort, model rankings reorganised, as RBF-SVM (AUC = 0.986, 95% CI 0.946–0.997), RandomForest (0.985, 95% CI 0.945–0.996), VQC (0.983, 95% CI 0.942–0.995), and the hybrid ensemble (0.982, 95% CI 0.941–0.995) all retained near-ceiling discrimination with extensively overlapping confidence intervals. Ablation analysis demonstrated that the choice of embedding rotation is decisive—Z-rotation embeddings collapsed VQC performance to chance levels (AUC ≈ 0.50), whereas X- and Y-rotations preserved it. Depolarising noise up to p = 0.10 had a negligible effect on the VQC, and SHAP analyses converged on worst concave points, mean concave points, and worst area as the dominant predictors across both classical and quantum models. Decision curve analysis showed positive net benefit for both classical and hybrid models across the clinically meaningful threshold range, exceeding both the treat-all and treat-none reference strategies throughout. Conclusions: In the present regime, the principal contribution of QML is not raw discrimination—modern classical learners are already at the data ceiling—but the construction of a rigorous, reproducible, externally validated, and interpretable benchmarking framework in which quantum models can be fairly compared with their classical counterparts. Because evaluation was confined to curated benchmark datasets rather than real-world clinical populations, the interpretability and net benefit findings reported here should be read as benchmark-level evidence and not as a demonstration of readiness for clinical deployment. Full article

Frequency instability caused by reduced system inertia in inverter-dominated islanded microgrids represents a critical challenge in renewable-integrated power systems. Conventional fixed-parameter controllers exhibit limited adaptability to uncertain and time-varying low-inertia conditions. This paper proposes an LSTM–MPC + VIC framework that embeds a Long Short-Term Memory (LSTM) surrogate predictor directly within a Model Predictive Control (MPC) optimisation loop, coordinated with a Virtual Inertia Controller (VIC) for immediate transient support. The LSTM provides data-driven frequency predictions without requiring precise analytical system modelling, while the VIC supplies reactive inertial damping within the same control cycle. The proposed controller is evaluated against Proportional–Integral–Derivative (PID), PSO-optimised PID, and standard MPC baselines on a 50 Hz islanded microgrid. Results demonstrate the lowest maximum frequency deviation of 0.009748 Hz, fastest settling time of 36.34 s, and minimum integral absolute error of 0.12283 Hz·s among all controllers. A Lyapunov-based Input-to-State Stability (ISS) analysis, incorporating the load disturbance term via Young’s inequality, confirms an ISS ultimate bound of 0.057866 Hz and an effective decay rate of 1.2952 s⁻¹. Robustness is further validated through multi-scenario testing, parametric sensitivity analysis, component ablation, and computational feasibility assessment, confirming suitability for real-time deployment in low-inertia microgrid systems. Full article

In structural engineering, the iterative optimisation of complex timber wall components is fundamentally limited by the prohibitive computational costs of non-linear Finite Element Analysis (FEA). To overcome this bottleneck, this study introduces a highly efficient machine learning-based surrogate model utilising a convolutional encoder–decoder architecture to predict the global buckling load factor (BLF) directly from structural topology. We evaluate three distinct modeling strategies: a Direct prediction model, a Sequential model, and a multitask Dual-Loss model designed to simultaneously predict the BLF and reconstruct spatial tensile stress fields. Experimental results demonstrate that both the Direct and Dual-Loss approaches achieve near-perfect predictive accuracy on in-distribution data, yielding a coefficient of determination (R²) of approximately 0.996. Crucially, these surrogates accelerate inference times by a factor of roughly 12,800 compared to traditional iterative solvers, reducing evaluation times from hours to mere seconds. Furthermore, the models exhibit exceptional robustness and extrapolative capability under rigorous out-of-distribution testing. The models maintain high predictive fidelity when subjected to cross-dataset distributional shifts (R² $\approx 0.94$) and when evaluating intentionally low-performing, highly vulnerable configurations (R²$\approx 0.967$). Extensive validation on structurally disjoint, hold-out wall geometries confirms the models’ ability to generalise to entirely unseen topologies without introducing systematic bias (R²$>0.95$). By successfully internalising the underlying physical principles of load redistribution, this surrogate framework provides a reliable, computationally inexpensive foundation for enabling real-time, autonomous generative design and structural optimisation. Full article

Phenomenological crystal plasticity (CP) models are widely used in Integrated Computational Materials Engineering (ICME) to bridge microstructural features with engineering-scale mechanical behaviour. However, their practical application is hindered by two major challenges: high computational costs of physics-based simulations, and the labour-intensive, trial-and-error nature of parameter calibration. These challenges are amplified in additively manufactured (AM) materials, where location-dependent properties require calibration to be repeated at multiple points to produce a detailed property map. Additionally, a limited understanding of how individual parameters of the CP models influence stress–strain predictions across the strain spectrum compounds these issues, making it challenging to utilise CP models for efficient materials design. To address these limitations, we developed an integrated framework that combines deep neural network (DNN) surrogates, optimisation algorithms (OAs), and explainable AI (XAI) techniques. We also utilised experimental tensile data from AM AlSi10Mg alloy as ground truth since AM materials are expected to benefit the most from our investigation. We demonstrate that, by using OAs such as a Natural Evolutionary Strategy or a Genetic Algorithm, the calibration process can be made more accurate and significantly accelerated. We also investigated the utility of employing deep neural network (DNN) surrogates of CP simulations in the calibration process. The fast-solving DNN surrogates achieved substantial time savings in the absence of OAs, i.e., during exhaustive parameter searches mandated by trial-and-error strategies. However, their effectiveness in parameter discovery was context-dependent when used in conjunction with OAs, since OAs can sometimes converge with fewer simulations than required for DNN training. Furthermore, we applied Shapley Additive exPlanations (SHAP), an XAI method, which revealed intricate interactions among some CP parameters, offering insight into why conventional trial-and-error calibration approaches often prove challenging. Our study contributes to strengthening the practical relevance of CP models for modelling-informed materials engineering and optimisation applications. Finally, our integrated framework offers broad applicability beyond materials modelling, enabling accelerated discovery of tuneable parameters in phenomenological models and providing deeper insight into their contributions to predictions. Full article

Water-soil gushing caused by tunnel leakage can induce severe ground disturbance and threaten the safety of shield tunnels, yet rapid prediction remains difficult because high-fidelity numerical simulations are computationally expensive. This study develops an interpretable machine-learning framework for predicting gushing-induced ground disturbance around shield tunnels based on a validated two-phase Material Point Method database. Six governing variables are considered, including the tunnel depth ratio, gushing location, soil friction angle, Young’s modulus, intrinsic permeability, and soil gushing mass. Three representative response variables were selected, namely the maximum ground settlement, flow-zone width, and flow-zone centroid angle. Five algorithms, including MLP, RF, XGBoost, SVR, and Ridge, were established and compared, with hyperparameters optimised using Optuna. The results show that nonlinear models consistently outperform the linear baseline, among which MLP, RF, and XGBoost achieve the best overall accuracy and robustness. Error-distribution analysis further indicates that MLP and RF yield the highest proportion of low-error predictions. SHAP interpretation shows that SGM is the dominant factor governing maximum settlement and flow-zone width, whereas gushing location primarily controls the flow-zone centroid angle. The proposed framework provides an efficient and physically interpretable surrogate for rapid hazard assessment of gushing-induced ground disturbance in shield tunnelling. Full article

This paper presents a novel, efficient computational framework for the optimisation of the fundamental frequency of multi-layered composite slabs with consideration of uncertainties. The approach is based on Finite Element Method (FEM) data generation, Deep Neural Network (DNN) surrogate modelling, deterministic optimisation using the genetic algorithm (GA), Morris Sensitivity Analysis (SA), and quantile-based optimisation, including uncertainties and using the GA. Different boundary condition configurations are considered. The surrogate model is trained on FEM-generated samples and subsequently used to replace expensive modal analyses during optimisation, significantly reducing the optimisation evaluation cost for one boundary condition variant. The proposed method achieves near-identical optimal non-dimensional parameter $\mathsf{\Omega }$ values to those reported in the literature for Bayesian Optimisation (BO), with discrepancies of less than 0.5%. To improve robustness to manufacturing tolerances, an additional uncertainty-aware optimisation is performed, in which model parameters are perturbed with normally distributed noise. By maximising the 5% quantile of the non-dimensional parameter $\mathsf{\Omega }$, robust optimal solutions are obtained with minimal loss in performance. Overall, the DNN-GA framework enables fast and accurate optimisation of composite laminates and provides both deterministic and robust design recommendations at a fraction of the computational cost of traditional FEM-based optimisation workflows. Full article

The optimisation design of agricultural machinery is shifting from offline, experience-driven engineering towards adaptive, data-driven, and closed-loop intelligent optimisation. Conventional approaches based on computer-aided engineering (CAE), empirical testing, mathematical modelling, and static multi-objective optimisation have provided an important engineering foundation, but they remain limited under unstructured field conditions involving soil heterogeneity, crop variability, climatic disturbance, and nonlinear machinery–environment interactions. This review systematically examines the evolution of intelligent optimisation design for agricultural machinery from conventional simulation-based methods to artificial intelligence (AI)- and digital twin (DT)-enabled paradigms. First, mathematical modelling, response surface methodology, discrete element method (DEM), computational fluid dynamics (CFD), multi-body dynamics (MBD), heuristic algorithms, and early AI-assisted surrogate optimisation are reviewed to clarify their contributions and limitations. Second, frontier enabling technologies are analysed, including agriculture-specific large models, generative AI, lightweight edge intelligence, deep reinforcement learning (DRL), embodied AI, federated learning (FL), and privacy-preserving computing. Third, system-level applications integrating DT and AI are discussed, with emphasis on full-lifecycle machinery optimisation, device–edge–cloud collaborative control, multi-agent fleet coordination, predictive maintenance, and Agriculture 5.0-oriented intelligent equipment systems. Key deployment bottlenecks are further identified, including sim-to-real inconsistency, virtual–physical mismatch in DTs, edge-side trade-offs among accuracy, latency, energy consumption, and cost, insufficient validation standards, and economic adoption barriers. Finally, a 2025–2030 roadmap is proposed, highlighting large-model–DT closed loops, control biomimetics, green low-carbon optimisation, and trustworthy human–machine symbiosis for sustainable Agriculture 5.0. Full article