International Journal of Mechanics and Materials in Design最新文献

The development of electric vehicles (EVs) has increased demand for advanced Energy Management Systems (EMS) that optimizes energy distribution, intelligently handles multiple power sources. One of the primary issues is the intermittent, unpredictable nature of renewable sources which leads to fluctuating power availability, makes real-time energy balancing difficult. Additionally, EMS must handle nonlinear system behavior, rapid load variations requiring fast, accurate control strategies. Hence, an AI-enhanced EV EMS that uses intelligent algorithms, adaptive control techniques is presented in this work. The proposed solution integrates a solar Photovoltaic (PV) array with voltage boosted using an innovative High Gain Coupled Inductor Extended Boost (HiG-CI-X Boost) Converter before being sent onto a common high-voltage bus powering the Brushless DC (BLDC) motor. In order to ensure high performance, energy flexibility, the architecture also incorporates battery, supercapacitor energy storage devices, which are interconnected via bidirectional DC-DC converters. Deep Neural Network (DNN)-based State of Charge (SoC) estimator is incorporated to predict the battery state. A three-phase Voltage Source Inverter (VSI) converts the DC bus voltage to AC in order topower BLDC motor. Accurate control over torque, speed is ensured by a feedback loop. MATLAB simulation results show that the AI-integrated EMS significantly increases energy efficiency, reduces the load on individual storage units, maximizes transient response. Moreover, the proposed converter results with 97.63% efficiency with 1.46% increase from existing topologies, with NBO-ANFIS MPPT algorithm foreshowing tracking efficiency of 99.85% with settling time at 0.14 s, respectively. The proposed MPPT achieves approximately 9–10% higher tracking efficiency reduces the settling time by nearly 78–80%.The proposed system is a viable option for intelligent, ecologically friendly electric mobility because of its exceptional scalability, robustness.

The rapid development of ultrafast heating technology in the processing and fabrication of multi-layered piezoelectric composites has triggered numerous studies on the constitutive modeling and transient impact response of multi-physics coupling behavior for vibration control and intelligent thermal management. To accurately predict dynamic thermo-electromechanical responses in such extreme short-duration thermal conditions, this work aims to establish a novel piezoelectric thermoelasticity model by incorporating the two-parameter Mittag–Leffler fractional derivative into electron–phonon dual-phase-lag heat conduction law and strain-rate constitutive equation. The developed model is applied to investigate the transient thermal impact responses of multi-layer piezoelectric composites with non-idealized interfaces. The governing equations are solved by Laplace transformation technique associated with FFT-based numerical inversion algorithm. Dimensionless results indicate that the two-parameter Mittag–Leffler fractional derivative and material constants ratios remarkably accelerate heat propagation speed, enhance electric potential generation and mitigate excessive temperature or deformation.

Soft manipulators have gained increasing attention for their adaptability to unstructured environments, giving rise to various structural configurations. However, structural optimization typically requires case-specific parameter analysis, resulting in low optimization efficiency, while many designs still rely heavily on empirical parameter selection. This letter presents an intelligent structural optimization framework for soft manipulators. The key design parameters and performance indexes are first defined according to the considered manipulator to characterize its structural performance. A parametric finite element method (PFEM) is then used to generate performance datasets under varying parameter combinations. A genetic algorithm optimized backpropagation neural network (GABP) is subsequently trained to model the nonlinear parameter-performance relationships, forming the optimization objective function. Finally, a multi-objective particle swarm optimization (MOPSO) algorithm is applied to obtain the Pareto-optimal solutions from which optimal parameters are selected. The proposed optimization framework is validated through a case study of a cable-driven bellows soft manipulator. The GABP model accurately captures the complex mapping, achieving (R^2 > 0.998), and experiments further confirm the prediction accuracy with relative errors of 0.63% for stiffness k, 7.35% for bending stiffness EI, and 3.28% for natural frequency f. Structural optimization yields a 7.7% increase in relative compliance R, a 1.9% improvement in load capacity P, and a 34.8% increase in natural frequency f, demonstrating the effectiveness of the proposed optimization framework for enhancing the structural performance of the considered soft manipulator.

This study introduces a Jacobi–Ritz method to analyze the buckling, bending, free vibration, and damped vibration behaviors of functionally graded triply periodic minimal surface (FG-TPMS) microbeams supported by a viscoelastic foundation. The theoretical framework is based on higher-order shear deformation theory, with size-dependent effects accounted for by the modified couple stress theory. Two representative TPMS configurations—Primitive and Gyroid—alongside four distinct material gradation schemes (A1, A2, B1, and B2) are thoroughly investigated. Lagrange’s equation is used to achieve the governing equations, and the Ritz method with a Jacobi function is employed to solve the problem. The present method demonstrates significant flexibility in handling various boundary conditions, including simply-supported, clamped-free, and clamped-simply supported. The method’s accuracy and reliability are confirmed through comparisons with available results in existing literature. In addition, this research provides novel insights into the bending, buckling, and vibration characteristics of FG-TPMS microbeams resting on the viscoelastic foundation, offering valuable reference data for future studies and practical applications.

This study investigates the damped forced vibration response of bi-directional functionally graded (BDFG) nanobeams resting on a viscoelastic foundation, with emphasis on material gradation, nanoscale behaviour, foundation stiffness, and damping. To the best of the authors’ knowledge, this work establishes for the first time a canonical first-order ordinary differential equation formulation in the Laplace domain through conjugate momentums and their derivatives for the damped forced vibration of BDFG nanobeams resting on a viscoelastic foundation. The foundation is represented by visco-Winkler and visco-Pasternak components, allowing stiffness and damping effects to be considered simultaneously, while size dependency is incorporated through Eringen’s approach. The transfer matrix method is used to construct the element dynamic stiffness matrix, while Durbin’s technique recovers the transient response. Benchmark and convergence studies confirm the reliability of the procedure. Higher Winkler and Pasternak foundation stiffness coefficients reduce displacement amplitudes and shorten oscillation periods, while their damping components suppress the transient response and produce the strongest attenuation when acting simultaneously. Increasing the transverse gradient parameter increases displacement amplitudes and oscillation periods; however, the response is more sensitive to its initial increase, whereas further increases produce smaller changes. Increasing the axial gradient parameter reduces displacement amplitudes but generally lengthens oscillation periods. Most importantly, increasing nonlocality produces conventional stiffness-softening under simply supported conditions, whereas the opposite tendency is observed for the cantilever configuration, where displacement amplitudes decrease and oscillation periods shorten. This contrasting behaviour reveals the nonlocal stiffness-hardening paradox and demonstrates the strong dependence of the nonlocal response on boundary conditions.

This research outlines a dynamic aeroelastic framework, using the Carrera unified formulation (CUF), combined with the doublet lattice method for multilayer plates that appear in advanced sports equipment structural parts. The proposed sandwich is built like: on the upper side, a Lead Zirconate Titanate piezoelectric sheet, also labeled PZT-5H; then you have a periodic lattice cellular core that is filled with Polyvinyl Chloride (PVC) as a viscoelastic medium, and at the lower end, there is another Lead Zirconate Titanate layer identified as PZT-4. To get the effective mechanical properties of that lattice core, the work relies on the Voigt averaging technique, mainly to mirror the hierarchical viscoelastic behavior of the cellular architecture. Inside the CUF framework, a layerwise kinematic model is paired with Mixed Interpolation of tensorial components plate elements. This pairing helps in tracking transverse shear deformation plus interlaminar stress continuity, and it also supports the coupled aeroelastic response when the system is under dynamic loading, conditions that can easily disturb everything. On top of that, a parallel multimodal shunt damping strategy is attached to the piezoelectric layers. The idea is to curb structural vibrations and, at the same time, boost flutter stability, while also improving electrical energy harvesting capability. Numerical analyses show that the combined effects of the viscoelastic PVC-filled lattice core and the distributed piezoelectric shunt circuits actually reduce vibration amplitudes quite a bit, also delay the aeroelastic instability onset, and raise the harvested power efficiency. The proposed multiphysics formulation gives you an efficient computational approach for design and optimization of lightweight, vibration-controlled, and energy-aware sports equipment structures under a complex aerodynamic environment.

This work presents a numerical investigation of the thermomechanical behavior of a cylindrical auxetic lattice structure based on re-entrant unit cells arranged in the radial–circumferential plane. A computational framework for generating geometrically graded cylindrical auxetic lattices with controllable cell density and structural thickness was developed. Finite element simulations were performed to investigate the influence of lattice geometry and coefficient of thermal expansion (CTE) on the deformation response under thermal and coupled thermomechanical loading. The results demonstrate that the response is governed predominantly by topology-dependent kinematic mechanisms rather than solely by material properties. Increasing structural thickness suppresses rotational opening of the re-entrant cells and shifts the behavior toward that of a solid cylinder. Coupled thermomechanical simulations showed that internal pressure effectively suppresses thermally induced radial expansion, enabling deformation regimes approaching zero effective radial deformation. Comparison with different cylindrical lattice topologies demonstrated that structures with similar deformation magnitudes may exhibit substantially different stress redistribution behavior. The obtained results highlight the potential of cylindrical auxetic metamaterials with programmable thermal expansion and adaptive deformation response for advanced engineering applications.

Building Information Modeling (BIM) has transformed project delivery, yet contract design for BIM-enabled projects remains anchored in qualitative frameworks that implicitly assume uniform BIM capability across contracting parties. In practice, however, the designer and constructor often operate at substantially different levels of BIM capability maturity (BIM-CM), creating heterogeneous information environments that standard principal-agent models do not address. This study develops a normative analytical framework that treats BIM-CM as the central determinant of contract architecture. We construct a dual-agent principal-agent model in which the designer and constructor contribute complementary efforts through a collaborative production function, and we solve for optimal contract modes under three BIM-CM configurations: both parties at high maturity, both at low maturity, and–most distinctively–mixed maturity, where one party is mature and the other is not. Nine propositions are derived, and second-order conditions are verified to establish equilibrium uniqueness. The results yield three contract prescriptions. Under high BIM-CM, a compulsory (fixed-price) contract achieves the first-best: the risk-neutral owner bears all output risk. Under low BIM-CM, an incentive contract with risk-sharing is optimal. The mixed-maturity configuration produces a hybrid compulsory-plus-incentive contract absent from standard theory. A novel free-riding mechanism (Fig. 3) is identified: the high-maturity party’s observable effort, transmitted through the collaborative production function, raises the low-maturity party’s remuneration while depressing its own effort, such that the first-best remains unattainable even with one fully mature partner. Agency costs are decomposed into risk and incentive components and compared across configurations. The paper contributes a maturity-contingent contract typology and a decision heuristic that translate formal results into actionable guidance, replacing uniform qualitative prescriptions with differentiated contract recommendations grounded in BIM capability assessment.

Nanocomposites reinforced with multi-walled carbon nanotubes (MWCNTs) or graphene nanoplatelets (GNPs) have been widely studied due to their enhanced mechanical, thermal and electrical properties; however, hybrid systems combining both nanofillers remain relatively underexplored, particularly in sandwich structures with multifunctional materials. To address this gap, the vibration behaviors and multi-objective optimization of sandwich plate comprising auxetic honeycomb core, two hybrid nanocomposite intermediate layers reinforced with MWCNTs and GNPs and two magneto–electro–elastic (MEE) face sheets are investigated. Four different distribution patterns of CNTs and GNPs, namely O, X, U and V are considered. The governing equations are formulated based on Reddy’s higher order shear deformation plate theory, incorporating von Kármán nonlinear strain–displacement relationships. Various parametric studies are performed to examine the effects of the weight fractions and distribution patterns of CNTs and GNPs, MWCNT:GNP distribution ratio, GNP size, temperature increment, viscoelastic foundations as well as geometrical parameters. For the optimization problem, to reduce the computational cost of repeated analytical evaluations, six machine learning (ML) models are used to construct surrogate models integrated into the non-dominated sorting genetic algorithm II (NSGA-II) to simultaneously maximize the natural frequency and minimize the mass of the sandwich plate. The results demonstrate that the pure GNP reinforcement provides superior stiffness, whereas the balanced MWCNT:GNP ratio of 1:1 offers higher stiffness than the other hybrid reinforcement configurations and the pure CNT reinforcement exhibits the lowest stiffness. Furthermore, the integration of ML models into NSGA-II preserves the quality of the obtained solutions while reducing the computational cost by approximately 88.31–92.63%.

The main aim of this paper is to show the application of the intersection of transport phenomena in porous media and scientific machine learning approaches. Nowadays, machine learning is widely used to solve complex nonlinear fluid dynamics problems. In this study, we developed Reynolds-parameterized physics-informed neural networks (PINNs) under soft and hard frameworks to solve nonlinear turbine cooling disk model. The physics of our model involves solving momentum and energy equations for the laminar flow of a non-Newtonian fluid inside an axisymmetric porous channel. Hot gases heat the lower disk externally, while a cooling fluid is injected from the top disk. Using a stream function, we convert the governing nonlinear partial differential equations into coupled ordinary differential equations. We used a Reynolds-parameterized approach because it allows a single trained network to predict flow behavior across a continuous range of Reynolds numbers, eliminating the need to retrain the model for every new parameter value. Often, standard soft PINNs fail to satisfy boundary conditions because balancing the multi-objective loss function during training causes optimization issues. To fix this, we developed a hard PINNs approach for our model. By strictly encoding boundary conditions directly into the network architecture, the hard PINNs guarantee exact solutions at the boundaries. Validated against the Runge–Kutta method, the hard PINNs yield highly accurate results inside the domain, on the boundary, and strictly respect the underlying physics. Finally, polynomial surrogate correlations were estimated using polynomial fitting to generate interpretable models, validated by the coefficient of determination.