Pub Date : 2026-03-06DOI: 10.1016/j.ijmecsci.2026.111487
Yuquan Chen, Zhaoyong Sun, Yifan Zhang, Shuihai Dou, Yanping Du, Tuo Liu
This paper proposes a design method for arbitrarily shaped flexural waveguides based on quasi-conformal transformation. By virtue of this numerical approach, prescribed complex waveguide geometries are mapped onto a graded distribution of the effective refractive index, which is physically realized through direct and precise thickness modulation of an isotropic thin plate. This method effectively overcomes the critical limitation of geometric inflexibility inherent in existing flexural waveguide designs, which have been confined to regular shapes such as circular arcs. The feasibility of the proposed approach is demonstrated through the design, simulation, fabrication, and experimental validation of three distinct functional arbitrarily shaped flexural waveguides, namely: a vibration isolation waveguide, a wave bending waveguide, and an energy harvesting waveguide. These prototypes exhibit high-efficiency vibration isolation within 100–120 kHz, controlled 47° wave bending over 60–85 kHz with low phase distortion, and significant energy concentration within 100–120 kHz—more than doubling the local vibration intensity and enhancing piezoelectric output by over 100%. This work provides a general approach for custom flexural wave manipulation, it holds potential application value in fields such as structural vibration control, non-destructive testing, and vibration energy harvesting.
{"title":"Arbitrarily shaped broadband flexural waveguides based on quasi-conformal transformation","authors":"Yuquan Chen, Zhaoyong Sun, Yifan Zhang, Shuihai Dou, Yanping Du, Tuo Liu","doi":"10.1016/j.ijmecsci.2026.111487","DOIUrl":"https://doi.org/10.1016/j.ijmecsci.2026.111487","url":null,"abstract":"This paper proposes a design method for arbitrarily shaped flexural waveguides based on quasi-conformal transformation. By virtue of this numerical approach, prescribed complex waveguide geometries are mapped onto a graded distribution of the effective refractive index, which is physically realized through direct and precise thickness modulation of an isotropic thin plate. This method effectively overcomes the critical limitation of geometric inflexibility inherent in existing flexural waveguide designs, which have been confined to regular shapes such as circular arcs. The feasibility of the proposed approach is demonstrated through the design, simulation, fabrication, and experimental validation of three distinct functional arbitrarily shaped flexural waveguides, namely: a vibration isolation waveguide, a wave bending waveguide, and an energy harvesting waveguide. These prototypes exhibit high-efficiency vibration isolation within 100–120 kHz, controlled 47° wave bending over 60–85 kHz with low phase distortion, and significant energy concentration within 100–120 kHz—more than doubling the local vibration intensity and enhancing piezoelectric output by over 100%. This work provides a general approach for custom flexural wave manipulation, it holds potential application value in fields such as structural vibration control, non-destructive testing, and vibration energy harvesting.","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"198 1","pages":""},"PeriodicalIF":7.3,"publicationDate":"2026-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147392678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-22DOI: 10.1016/j.ijmecsci.2026.111299
Ying Yu , Jiayao Feng , Xiaolin Duan , Ke Liu , Yuxiang Cai
This paper presents a novel approach for programming motion paths of zero-energy modes along arbitrary trajectories in mechanical metamaterials. Moving beyond previous studies confined to linear propagation paths, we engineer chain-like linkages with independently tunable horizontal and vertical spacing parameters to actively encode and control soliton propagation along user-defined 2D paths. Our findings reveal that geometric adjustments not only enable new motion modes but also control over energy transmission efficiency. The soliton dynamics, simulated using the Finite Particle Method, validate robust motion and energy transfer processes within the structure, demonstrating successful propagation along programmed spiral and sinusoidal paths. The flipper phase achieving optimal efficiency while paths like spirals incur significant energy trade-offs. Our work provides advancements in designing topologically protected, reconfigurable mechanical systems, offering new methods for efficient energy and motion transfer in applications ranging from soft robotics to adaptive structural systems.
{"title":"Programming zero-energy mode in curved one dimensional metamaterial","authors":"Ying Yu , Jiayao Feng , Xiaolin Duan , Ke Liu , Yuxiang Cai","doi":"10.1016/j.ijmecsci.2026.111299","DOIUrl":"10.1016/j.ijmecsci.2026.111299","url":null,"abstract":"<div><div>This paper presents a novel approach for programming motion paths of zero-energy modes along arbitrary trajectories in mechanical metamaterials. Moving beyond previous studies confined to linear propagation paths, we engineer chain-like linkages with independently tunable horizontal and vertical spacing parameters to actively encode and control soliton propagation along user-defined 2D paths. Our findings reveal that geometric adjustments not only enable new motion modes but also control over energy transmission efficiency. The soliton dynamics, simulated using the Finite Particle Method, validate robust motion and energy transfer processes within the structure, demonstrating successful propagation along programmed spiral and sinusoidal paths. The flipper phase achieving optimal efficiency while paths like spirals incur significant energy trade-offs. Our work provides advancements in designing topologically protected, reconfigurable mechanical systems, offering new methods for efficient energy and motion transfer in applications ranging from soft robotics to adaptive structural systems.</div></div>","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"313 ","pages":"Article 111299"},"PeriodicalIF":9.4,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146033915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-20DOI: 10.1016/j.ijmecsci.2026.111283
Lizhuo Chen, Ning Chen, Jian Liu, Baizhan Xia
As space exploration advances, conventional circuits may fail in harsh magnetic environments, creating a demand for alternative computing systems. Aiming at the challenge that the structure of the mechanical neural network realizes the multi fully connected layers, a mechanical neural network made of cord and mechanical metamaterial is designed in this paper. This mechanical neural network information transmission is not restricted to a single direction, thereby exhibiting high spatial adaptability. It offers clear advantages in confined or embedded settings and supports compact, fully connected architectures for future miniaturization. In addition, modular design enables flexible assembly and system integration. In preliminary tests, the developed mechanical neural network can solve the problem of handwritten digit multi-classification based on multi-layer connection, and it is proved to have high spatial adaptability. The mechanical neural network can transmit signals between layers flexibly. This work paves the way for compact multi-layer mechanical neural networks with a wide range of applications in the complex environment like future outer space fields.
{"title":"Spatially adaptable mechanical neural networks with multi fully connected layers","authors":"Lizhuo Chen, Ning Chen, Jian Liu, Baizhan Xia","doi":"10.1016/j.ijmecsci.2026.111283","DOIUrl":"10.1016/j.ijmecsci.2026.111283","url":null,"abstract":"<div><div>As space exploration advances, conventional circuits may fail in harsh magnetic environments, creating a demand for alternative computing systems. Aiming at the challenge that the structure of the mechanical neural network realizes the multi fully connected layers, a mechanical neural network made of cord and mechanical metamaterial is designed in this paper. This mechanical neural network information transmission is not restricted to a single direction, thereby exhibiting high spatial adaptability. It offers clear advantages in confined or embedded settings and supports compact, fully connected architectures for future miniaturization. In addition, modular design enables flexible assembly and system integration. In preliminary tests, the developed mechanical neural network can solve the problem of handwritten digit multi-classification based on multi-layer connection, and it is proved to have high spatial adaptability. The mechanical neural network can transmit signals between layers flexibly. This work paves the way for compact multi-layer mechanical neural networks with a wide range of applications in the complex environment like future outer space fields.</div></div>","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"313 ","pages":"Article 111283"},"PeriodicalIF":9.4,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-28DOI: 10.1016/j.ijmecsci.2026.111323
Shuai Chen , Shouning Deng , Jie Xu , Yanan Liu , Yunlong Chen , Xue Wan , Bing Wang , Linzhi Wu
Triply periodic minimal surface (TPMS) porous structures exhibit superior specific surface area, smooth geometry, and interconnected pore networks, which make them promising candidates for multifunctional applications such as lightweight energy absorption, thermal protection, and biomedical implants. However, systematic understanding of their coupled mechanical–thermal behavior and gradient design strategies remains limited. In this work, a comprehensive study on the mechanical and thermal performance of TPMS porous structures was conducted through parametric modeling, additive manufacturing (AM), finite element simulations, and experimental validation. Three representative TPMS configurations (Schwarz, Gyroid, and Diamond) were first constructed based on implicit equations, and Ti-6Al-4 V samples were fabricated using selective laser melting (SLM). Quasi-static compression tests and heat conduction experiments were performed to evaluate structural behavior, and the results were validated by finite element analysis. The effects of cell size, volume fraction, and structural type on compressive modulus, yield strength, and equivalent thermal conductivity were systematically revealed. Furthermore, functionally graded TPMS structures were proposed using one-dimensional gradients and fusion-transition strategies based on linear, sinusoidal, and power functions. Simulation results demonstrated that gradient configurations enable smooth variation in stiffness, strength, and thermal conductivity, providing superior tunability compared to uniform structures. The comparative analysis highlighted the potential of fusion-transition designs to overcome strength–thermal trade-offs and achieve customized multifunctional performance. This study establishes a theoretical and experimental foundation for the design of TPMS gradient porous structures, offering valuable guidance for the development of next-generation lightweight, load-bearing, and thermally efficient composite systems.
{"title":"Mechanical and thermal performance of functionally graded TPMS porous structures","authors":"Shuai Chen , Shouning Deng , Jie Xu , Yanan Liu , Yunlong Chen , Xue Wan , Bing Wang , Linzhi Wu","doi":"10.1016/j.ijmecsci.2026.111323","DOIUrl":"10.1016/j.ijmecsci.2026.111323","url":null,"abstract":"<div><div>Triply periodic minimal surface (TPMS) porous structures exhibit superior specific surface area, smooth geometry, and interconnected pore networks, which make them promising candidates for multifunctional applications such as lightweight energy absorption, thermal protection, and biomedical implants. However, systematic understanding of their coupled mechanical–thermal behavior and gradient design strategies remains limited. In this work, a comprehensive study on the mechanical and thermal performance of TPMS porous structures was conducted through parametric modeling, additive manufacturing (AM), finite element simulations, and experimental validation. Three representative TPMS configurations (Schwarz, Gyroid, and Diamond) were first constructed based on implicit equations, and Ti-6Al-4 V samples were fabricated using selective laser melting (SLM). Quasi-static compression tests and heat conduction experiments were performed to evaluate structural behavior, and the results were validated by finite element analysis. The effects of cell size, volume fraction, and structural type on compressive modulus, yield strength, and equivalent thermal conductivity were systematically revealed. Furthermore, functionally graded TPMS structures were proposed using one-dimensional gradients and fusion-transition strategies based on linear, sinusoidal, and power functions. Simulation results demonstrated that gradient configurations enable smooth variation in stiffness, strength, and thermal conductivity, providing superior tunability compared to uniform structures. The comparative analysis highlighted the potential of fusion-transition designs to overcome strength–thermal trade-offs and achieve customized multifunctional performance. This study establishes a theoretical and experimental foundation for the design of TPMS gradient porous structures, offering valuable guidance for the development of next-generation lightweight, load-bearing, and thermally efficient composite systems.</div></div>","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"313 ","pages":"Article 111323"},"PeriodicalIF":9.4,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-27DOI: 10.1016/j.ijmecsci.2026.111318
Zhiyong Qiu , Huan He , Linfeng Qu , Huyin Wang
High-temperature random vibration fatigue is a critical cause of aerospace structural failure, while obtaining high-temperature fatigue curves (S-N curves) remains time-consuming and costly. Thus, a novel and engineering-oriented estimation approach is proposed to predict fatigue strength and high-cycle S-N curves of metallic materials over a wide temperature range. The method requires only room-temperature S-N curve data and limited tensile and yield strengths at multiple temperatures, to establish a direct quantitative relationship between mechanical property degradation and fatigue behavior. The approach was validated through literature data and high-temperature random vibration fatigue tests on TA15 titanium alloy. The results confirm its accuracy and generality, demonstrating that fatigue strength decreases non-linearly with temperature and is strongly correlated with mechanical properties. The predicted high-temperature S-N curves of TA15, applied to fatigue life prediction, showed good agreement with experimental data, confirming the method’s predictive reliability. Further investigations reveal that, under high-temperature random vibration, both stress and velocity response power spectral densities shift toward lower frequencies while maintaining their overall spectral shapes. The combined effects of temperature-dependent stiffness degradation, modal damping, and excitation spectrum distribution lead to a non-monotonic variation in fatigue life with temperature. A moderate temperature rise improves fatigue life owing to higher damping, whereas further heating reduces it as stiffness degradation dominates. This paper presents an efficient, experimentally validated framework for estimating temperature-dependent S-N curves that markedly reduces high-temperature fatigue testing costs. It provides theoretical and engineering guidance for fatigue design and durability assessment of aerospace structures under thermal-vibrational coupling conditions.
{"title":"High-temperature fatigue curve estimation and random vibration fatigue failure","authors":"Zhiyong Qiu , Huan He , Linfeng Qu , Huyin Wang","doi":"10.1016/j.ijmecsci.2026.111318","DOIUrl":"10.1016/j.ijmecsci.2026.111318","url":null,"abstract":"<div><div>High-temperature random vibration fatigue is a critical cause of aerospace structural failure, while obtaining high-temperature fatigue curves (S-N curves) remains time-consuming and costly. Thus, a novel and engineering-oriented estimation approach is proposed to predict fatigue strength and high-cycle S-N curves of metallic materials over a wide temperature range. The method requires only room-temperature S-N curve data and limited tensile and yield strengths at multiple temperatures, to establish a direct quantitative relationship between mechanical property degradation and fatigue behavior. The approach was validated through literature data and high-temperature random vibration fatigue tests on TA15 titanium alloy. The results confirm its accuracy and generality, demonstrating that fatigue strength decreases non-linearly with temperature and is strongly correlated with mechanical properties. The predicted high-temperature S-N curves of TA15, applied to fatigue life prediction, showed good agreement with experimental data, confirming the method’s predictive reliability. Further investigations reveal that, under high-temperature random vibration, both stress and velocity response power spectral densities shift toward lower frequencies while maintaining their overall spectral shapes. The combined effects of temperature-dependent stiffness degradation, modal damping, and excitation spectrum distribution lead to a non-monotonic variation in fatigue life with temperature. A moderate temperature rise improves fatigue life owing to higher damping, whereas further heating reduces it as stiffness degradation dominates. This paper presents an efficient, experimentally validated framework for estimating temperature-dependent S-N curves that markedly reduces high-temperature fatigue testing costs. It provides theoretical and engineering guidance for fatigue design and durability assessment of aerospace structures under thermal-vibrational coupling conditions.</div></div>","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"313 ","pages":"Article 111318"},"PeriodicalIF":9.4,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-27DOI: 10.1016/j.ijmecsci.2026.111315
Heng Feng , Christopher DiGiovanni , Cael Johnston , Jidong Kang , Hassan Ghassemi-Armaki , Tingting Zhang , Kaan Inal
Resistance spot welds of advanced high strength steel (AHSSs) are susceptible to various fracture modes associated with distinct fracture paths leading to different loading capacity. To better predict the potential fracture modes and loading capacity of AHSS welds, a new force-based fracture criterion was developed based on the material strength of different regions in the welds together with identification of potential fracture paths. The established fracture criterion is then implemented into finite element (FE) models with the nugget simplification as a beam element in LS-Dyna. To be specific, the dissimilar spot weld stack-ups of Gen3 1180 steel and LCE 1000 steel sheets are investigated. The proposed model is validated by tests of coach peel, tensile shear, and KSII with different loading orientations. The model also predicts the transition of the fracture modes well, particularly when altering the welding procedure to vary the nugget size or microstructure of the heat affected zone (HAZ). Further, it is revealed that using the average material properties across the entire HAZ instead of the distinct HAZ regions, the model still accurately predicts the fracture modes and load capacity of the spot welds.
{"title":"New fracture criterion for spot welds of advanced high-strength steels","authors":"Heng Feng , Christopher DiGiovanni , Cael Johnston , Jidong Kang , Hassan Ghassemi-Armaki , Tingting Zhang , Kaan Inal","doi":"10.1016/j.ijmecsci.2026.111315","DOIUrl":"10.1016/j.ijmecsci.2026.111315","url":null,"abstract":"<div><div>Resistance spot welds of advanced high strength steel (AHSSs) are susceptible to various fracture modes associated with distinct fracture paths leading to different loading capacity. To better predict the potential fracture modes and loading capacity of AHSS welds, a new force-based fracture criterion was developed based on the material strength of different regions in the welds together with identification of potential fracture paths. The established fracture criterion is then implemented into finite element (FE) models with the nugget simplification as a beam element in LS-Dyna. To be specific, the dissimilar spot weld stack-ups of Gen3 1180 steel and LCE 1000 steel sheets are investigated. The proposed model is validated by tests of coach peel, tensile shear, and KSII with different loading orientations. The model also predicts the transition of the fracture modes well, particularly when altering the welding procedure to vary the nugget size or microstructure of the heat affected zone (HAZ). Further, it is revealed that using the average material properties across the entire HAZ instead of the distinct HAZ regions, the model still accurately predicts the fracture modes and load capacity of the spot welds.</div></div>","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"313 ","pages":"Article 111315"},"PeriodicalIF":9.4,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-27DOI: 10.1016/j.ijmecsci.2026.111312
Sibo Chai , Yan Chen , Zhong You , Jiayao Ma
Curved-crease origami utilizes coupled folding of creases and bending of panels to realize morphable shapes and programmable mechanical properties. Actuation based on stimuli-responsive materials can achieve effective folding/unfolding of origami structures and metamaterials without bulky mechanical loading systems, but the traditional strategy for straight-crease origami that relies on folding of active creases cannot accommodate the panel-bending-dominated deformation of curved-crease origami. In this work, a novel panel-driven actuation method is proposed, which enables active folding of curved-crease origami with large folding ratios and allows for programmable final configurations through the control of generators. Specifically, the geometric model and folding kinematics for general curved-crease origami are first established. Actuators are then incorporated into the panels, and it is demonstrated that actuation efficiency is maximized when the actuators are aligned perpendicular to the generators. Subsequently, the structure is discretized into finite crease-generator elements, and the equilibrium relationship between actuation strain and folding angle is derived using the minimum energy principle. Moreover, by continuously varying actuator width or strain along the elements, two strategies of programmable-width actuation and programmable-strain actuation are proposed to adjust curvature and program 3D morphing shapes, including planar arcs, planar spirals, and spatial spirals that deviate from natural energy-minimizing paths, which are validated through experiments using thermally responsive bimetal and numerical simulation. Compared with the crease-driven method, the panel-driven one can achieve large-scale folding of curved-crease origami structures with relatively small actuation strains, making it suitable for most existing active materials. Therefore, this work provides a theoretical framework for active folding of curved-crease origami.
{"title":"Panel-driven actuation framework for curved-crease origami","authors":"Sibo Chai , Yan Chen , Zhong You , Jiayao Ma","doi":"10.1016/j.ijmecsci.2026.111312","DOIUrl":"10.1016/j.ijmecsci.2026.111312","url":null,"abstract":"<div><div>Curved-crease origami utilizes coupled folding of creases and bending of panels to realize morphable shapes and programmable mechanical properties. Actuation based on stimuli-responsive materials can achieve effective folding/unfolding of origami structures and metamaterials without bulky mechanical loading systems, but the traditional strategy for straight-crease origami that relies on folding of active creases cannot accommodate the panel-bending-dominated deformation of curved-crease origami. In this work, a novel panel-driven actuation method is proposed, which enables active folding of curved-crease origami with large folding ratios and allows for programmable final configurations through the control of generators. Specifically, the geometric model and folding kinematics for general curved-crease origami are first established. Actuators are then incorporated into the panels, and it is demonstrated that actuation efficiency is maximized when the actuators are aligned perpendicular to the generators. Subsequently, the structure is discretized into finite crease-generator elements, and the equilibrium relationship between actuation strain and folding angle is derived using the minimum energy principle. Moreover, by continuously varying actuator width or strain along the elements, two strategies of programmable-width actuation and programmable-strain actuation are proposed to adjust curvature and program 3D morphing shapes, including planar arcs, planar spirals, and spatial spirals that deviate from natural energy-minimizing paths, which are validated through experiments using thermally responsive bimetal and numerical simulation. Compared with the crease-driven method, the panel-driven one can achieve large-scale folding of curved-crease origami structures with relatively small actuation strains, making it suitable for most existing active materials. Therefore, this work provides a theoretical framework for active folding of curved-crease origami.</div></div>","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"313 ","pages":"Article 111312"},"PeriodicalIF":9.4,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-30DOI: 10.1016/j.ijmecsci.2026.111325
Tianhao Wang , Tiangui Ye , Yukun Chen , Yukun Li , Guoyong Jin , Xinyu Jia
This study develops a semi-analytical reduced-order modeling framework for the underwater vibro-acoustic analysis of acoustic black hole-piezoelectric shunt damping (ABH-PSD) composite plates submerged in a semi-infinite heavy fluid. A variable-fidelity projection-based model order reduction (MOR) strategy is proposed by exploiting the intrinsic difference between the highly localized structural response induced by the ABH and the spatially smooth acoustic field, enabling efficient and accurate fluid-structure coupling. The partition collocation points method (PCPM) is incorporated to avoid the direct evaluation of frequency-dependent singular quadruple integrals arising from the Rayleigh radiation formulation. The model is validated against finite element method (FEM) simulations and available experimental data, demonstrating excellent accuracy while achieving an approximate 90% reduction in the computational cost of the acoustic subproblem. Based on this framework, systematic underwater parametric investigations of ABH plates are conducted, revealing the dominant role of the ABH indentation radius in modal redistribution and the associated low-frequency mode clustering under heavy fluid loading. Furthermore, the integration of PSD, particularly through parallel and series negative-capacitance resistive-inductive (PNCRL and SNCRL) circuits, provides substantial low-frequency vibro-acoustic suppression, with maximum reductions of 18.4 dB in mean-square velocity level (MVL) and 18.1 dB in sound pressure level (SPL) at the fundamental mode. Comparison with equal-areal-density plates reveals a clear synergistic vibro-acoustic suppression mechanism, in which the ABH concentrates vibrational energy into the piezoelectric region, while the shunt circuit efficiently dissipates the concentrated energy, offering an effective strategy for lightweight underwater structures with enhanced low-frequency vibro-acoustic stealth performance.
{"title":"Semi-analytical model and underwater vibro-acoustic analysis of ABH-PSD composite plates","authors":"Tianhao Wang , Tiangui Ye , Yukun Chen , Yukun Li , Guoyong Jin , Xinyu Jia","doi":"10.1016/j.ijmecsci.2026.111325","DOIUrl":"10.1016/j.ijmecsci.2026.111325","url":null,"abstract":"<div><div>This study develops a semi-analytical reduced-order modeling framework for the underwater vibro-acoustic analysis of acoustic black hole-piezoelectric shunt damping (ABH-PSD) composite plates submerged in a semi-infinite heavy fluid. A variable-fidelity projection-based model order reduction (MOR) strategy is proposed by exploiting the intrinsic difference between the highly localized structural response induced by the ABH and the spatially smooth acoustic field, enabling efficient and accurate fluid-structure coupling. The partition collocation points method (PCPM) is incorporated to avoid the direct evaluation of frequency-dependent singular quadruple integrals arising from the Rayleigh radiation formulation. The model is validated against finite element method (FEM) simulations and available experimental data, demonstrating excellent accuracy while achieving an approximate 90% reduction in the computational cost of the acoustic subproblem. Based on this framework, systematic underwater parametric investigations of ABH plates are conducted, revealing the dominant role of the ABH indentation radius in modal redistribution and the associated low-frequency mode clustering under heavy fluid loading. Furthermore, the integration of PSD, particularly through parallel and series negative-capacitance resistive-inductive (PNCRL and SNCRL) circuits, provides substantial low-frequency vibro-acoustic suppression, with maximum reductions of 18.4 dB in mean-square velocity level (MVL) and 18.1 dB in sound pressure level (SPL) at the fundamental mode. Comparison with equal-areal-density plates reveals a clear synergistic vibro-acoustic suppression mechanism, in which the ABH concentrates vibrational energy into the piezoelectric region, while the shunt circuit efficiently dissipates the concentrated energy, offering an effective strategy for lightweight underwater structures with enhanced low-frequency vibro-acoustic stealth performance.</div></div>","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"313 ","pages":"Article 111325"},"PeriodicalIF":9.4,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146089487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-16DOI: 10.1016/j.ijmecsci.2026.111278
Xi Wang , Jidong Zhao , Zhen-Yu Yin , Xiaoying Zhuang
The deep energy/Ritz method (DEM/DRM) offers advantages over physics-informed neural networks (PINNs), including reduced derivative orders and accelerated training. However, DEM encounters critical failure modes in both forward and inverse analyses, with underlying mechanisms and robust remedies remaining underexplored. To our knowledge, this work presents the first formal analysis that systematically identifies two distinct DEM failure modes, forward divergence and inverse collapse, and establishes their root causes along with sound countermeasures. In forward analysis, DEM training may diverge due to artificial energy minimization, where abrupt loss reductions below the physically admissible minimum occur with catastrophic errors, which are thermodynamically infeasible but remain unclarified. We prove that this stems from numerical integration inaccuracies in neural network representations, inducing pathological overfitting with escalating complexity. In inverse problems involving unknown material parameters or Neumann boundary conditions, we reveal that DEM fails because its variational formulation with respect to such unknown parameters is not well defined. To overcome these limitations, we propose a novel Energy-Informed Neural Operator Network (EINO), integrating a new regularization technique. Our framework incorporates: (1) a finite-element-informed regularization that lower-bounds the loss by the ground-truth FEM energy to ensure stability, and (2) a deep operator architecture with two-stage training that reconstructs unknown parameters/boundary conditions by embedding inverse constraints. Comprehensive benchmarks on 2D/3D linear/nonlinear solid mechanics and diffusion problems confirm EINO’s superiority over DEM. EINO resolves forward divergence even on very coarse meshes and achieves substantially lower parameter errors in inverse discovery (e.g., <2% relative error under 200% Gaussian noise). The elucidated failure mechanisms and the EINO framework collectively promote physics-constrained learning for surrogate modeling and inverse uncertainty quantification, minimizing the reliance on labeled data.
{"title":"Failure mechanisms and resolution in deep energy method","authors":"Xi Wang , Jidong Zhao , Zhen-Yu Yin , Xiaoying Zhuang","doi":"10.1016/j.ijmecsci.2026.111278","DOIUrl":"10.1016/j.ijmecsci.2026.111278","url":null,"abstract":"<div><div>The deep energy/Ritz method (DEM/DRM) offers advantages over physics-informed neural networks (PINNs), including reduced derivative orders and accelerated training. However, DEM encounters critical failure modes in both forward and inverse analyses, with underlying mechanisms and robust remedies remaining underexplored. To our knowledge, this work presents the first formal analysis that systematically identifies two distinct DEM failure modes, forward divergence and inverse collapse, and establishes their root causes along with sound countermeasures. In forward analysis, DEM training may diverge due to artificial energy minimization, where abrupt loss reductions below the physically admissible minimum occur with catastrophic errors, which are thermodynamically infeasible but remain unclarified. We prove that this stems from numerical integration inaccuracies in neural network representations, inducing pathological overfitting with escalating complexity. In inverse problems involving unknown material parameters or Neumann boundary conditions, we reveal that DEM fails because its variational formulation with respect to such unknown parameters is not well defined. To overcome these limitations, we propose a novel Energy-Informed Neural Operator Network (EINO), integrating a new regularization technique. Our framework incorporates: (1) a finite-element-informed regularization that lower-bounds the loss by the ground-truth FEM energy to ensure stability, and (2) a deep operator architecture with two-stage training that reconstructs unknown parameters/boundary conditions by embedding inverse constraints. Comprehensive benchmarks on 2D/3D linear/nonlinear solid mechanics and diffusion problems confirm EINO’s superiority over DEM. EINO resolves forward divergence even on very coarse meshes and achieves substantially lower parameter errors in inverse discovery (e.g., <2% relative error under 200% Gaussian noise). The elucidated failure mechanisms and the EINO framework collectively promote physics-constrained learning for surrogate modeling and inverse uncertainty quantification, minimizing the reliance on labeled data.</div></div>","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"313 ","pages":"Article 111278"},"PeriodicalIF":9.4,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145995481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：