Modeling complex, non-stationary dynamics remains challenging for deterministic neural networks. We present the Chaos-Integrated Synaptic-Memory Network (CISMN), which embeds controlled chaos across four modules—Chaotic Memory Cells, Chaotic Plasticity Layers, Chaotic Synapse Layers, and a Chaotic Attention Mechanism—supplemented by a logistic-map learning-rate schedule. Rigorous stability analyses (Lyapunov exponents, boundedness proofs) and gradient-preservation guarantees underpin our design. In experiments, CISMN-1 on a synthetic acoustical regression dataset (541 samples, 22 features) achieved R ² = 0.791 and RMSE = 0.059, outpacing physics-informed and attention-augmented baselines. CISMN-4 on the PMLB sonar benchmark (208 samples, 60 bands) attained R ² = 0.424 and RMSE = 0.380, surpassing LSTM, memristive, and reservoir models. Across seven standard regression tasks with 5-fold cross-validation, CISMN led on diabetes (R ² = 0.483 ± 0.073) and excelled in high-dimensional, low-sample regimes. Ablations reveal a scalability–efficiency trade-off: lightweight variants train in <10 s with >95% peak accuracy, while deeper configurations yield marginal gains. CISMN sustains gradient norms (~2300) versus LSTM collapse (<3), and fixed-seed protocols ensure <1.2% MAE variation. Interpretability remains challenging (feature-attribution entropy ≈ 2.58 bits), motivating future hybrid explanation methods. CISMN recasts chaos as a computational asset for robust, generalizable modeling across scientific, financial, and engineering domains.

The buckling mode in piezoelectric materials offers advantages such as an increased measurable strain range, ease of installation, and extended service life. This paper investigates the potential of piezoelectric sensors operating in buckling mode for structural strain measurement by evaluating key factors including boundary conditions, sensor response linearity under dynamic loading, and impedance engineering to optimize the voltage–strain relationship. A structural extension was developed to facilitate sensor integration and to enable the application of different buckling boundary conditions. Results show that the clamped–clamped configuration generated at least 1.65 times higher output voltage, and three times greater peak strain compared to other boundary conditions. An experimentally validated analytical model was employed to assess and improve the performance of buckled piezoelectric sensors in dynamic environments. The findings highlight that introducing initial buckling reduces signal perturbations, enhances voltage linearity across loading frequencies, and extends the effective strain measurement range. Furthermore, impedance engineering was used to successfully mitigate the nonlinear effects of transient response, thereby improving signal stability and accuracy in dynamic strain monitoring applications.

The wide application of curved composite profiles across various industries raises questions about transferring findings from standardised tests to curved structures. Particularly for low-velocity impacts, understanding the deformation and damage behaviour of curved structures is crucial to achieve their lightweight potential. Carbon fibre reinforced polymer (CFRP) composites often show no surface visible damage after low-energy impacts, but barely visible impact damage (BVID) can compromise load-bearing capacity under continuous loads. This study aims to evaluate the effectiveness of self-reporting thin-ply hybrid composites as coating layers for structural health monitoring in curved panels, focusing specifically on enhancing the visual detection of BVID. Quasi-isotropic curved composite panels made from IM7 carbon/8552 epoxy were first manufactured, and their mechanical response under quasi-static indentation was analysed and compared to flat panels. Next, a hybrid composite—comprising a layer of unidirectional S-glass/epoxy and thin-ply YS-90A carbon/epoxy—was applied to the outer surfaces of the curved panels. To simulate a real-world application, the curved panels were designed with dimensions similar to those of composite hydrogen storage tanks. The results indicate that the hybrid composite sensors functioned satisfactorily and provided direct correlations between visible surface damage on the surface and BVID. The visual inspection results are connected to data obtained from the load-displacement graphs, enabling a thorough analysis of sensor performance in visualising different stages of BVID.

This paper reports the development, optimization, and real-world application of an innovative wobbling triboelectric nanogenerator to harvest energy from wind or vibrations. The harvester features a spring-supported structure, purposely designed to become unbalanced and reversibly shift from a steady to a non-steady state in response to minimal wind or vibration stimuli. Unlike conventional wind turbines, this approach transforms wind energy into contact-separation events via a wobbling structure. The harvester’s mechanisms are engineered to enhance power generation efficiency, optimizing parameters like electrode dimensions and contact-separation quality. Experimental findings showcase a maximum output power density of 1.6 W/m2 under optimal conditions, employing a fixed suspending mechanism for heightened impact energy during contact. Moreover, the harvester efficiently charges a 3.7 V lithium-ion battery with over 4.5 μA, showcased in a self-powered light mast as a practical demonstration. The harvester provides cost-effectiveness by utilizing inexpensive, easily accessible materials without complex fabrication. This research paves the way for future exploration in integrating triboelectric nanogenerators with wobbling mechanisms, offering a promising pathway for sustainable power generation across various applications, including lighting and IoT sensor nodes.

High energy consumption in residential buildings poses significant challenges, prompting governments to regulate this sector through comprehensive energy assessments and classification strategies. This study introduces a multi-layer perceptron artificial neural network (ANN) model to grade and predict energy consumption levels in residential buildings in Tabriz, Iran, based on their geometric and functional characteristics. This study uses the K-Nearest Neighbors (KNN) algorithm to classify energy consumption grades based on energy ratio (R-value). Six sample buildings were modeled using Rhinoceros 3D version 7 and Grasshopper version 1.0.0007 software to extract key energy-influencing factors. A parametric geometric model was developed for rapid data generation and validated against reference buildings to ensure reliability. Building classifications spanned areas of 40 to 300 square meters and heights of up to six stories, with energy evaluations conducted using EnergyPlus. The collected data informed the ANN model, enabling accurate predictions for existing and future constructions. The results demonstrate that the model achieves a remarkable prediction error of just 0.001, facilitating efficient energy assessments without requiring extensive modeling expertise. This research emphasizes the role of geometric features and natural lighting in energy consumption prediction, highlighting the model’s practicality for early design evaluations and architectural validations.

Favourable pseudo-ductile behaviour under compressive loading with a knee-point was achieved for unidirectional (UD) interlayer hybrids made of thin-ply high modulus carbon/epoxy (CF/EP) layers sandwiched between standard thickness glass/epoxy (GF/EP). The UD thin-ply hybrids were tested under two loading scenarios: 1. Direct compressive loading, 2. Four-point bending loading. In both cases, the damage mechanisms responsible for the pseudo-ductile behaviour are fragmentation of the carbon layer and localised delamination, which later propagates unstably. The final failure of the UD thin-ply hybrid composites examined in four-point bending loading occurs at a higher strain than that under direct compressive loading. This is due to the strain gradient in bending, which results in a lower energy release rate than in direct compression. An increasing carbon layer thickness reduces the final delamination failure strain of the UD thin-ply hybrid composites in compression, but the knee-point strain is not affected.

Accurate and reliable strain measurement is essential for effective condition monitoring of engineering structures. This study presents an analytical and experimental investigation into the performance of piezoelectric sensors for structural strain measurements, evaluating the effect of attachment strategy and the properties of the substrate and the sensor. Lead zirconate titanate (PZT) and polyvinylidene fluoride (PVDF) sensors were evaluated in two attachment configurations: Fully Attached (FA) and Two-End Attached (TEA). A voltage-strain relationship was developed based on principles of piezoelectricity, electrical circuit modelling, and solid mechanics. Results indicate that sensor performance is significantly influenced by the attachment method. Specifically, the TEA configuration reduced the impact of substrate properties and improved uniaxial strain measurement accuracy by up to 32 % compared to the FA configuration. The FA configuration exhibited sensitivity to the substrate's Poisson ratio, leading to a nonlinear voltage-strain response. In contrast, the TEA configuration provided pure uniaxial strain measurements by reducing the effects of shear lag and substrate elasticity. These findings provide a comprehensive approach to using piezoelectric sensors for structural strain measurement, allowing for the placement of sensors on various substrates without the need for calibration by effectively utilizing sensor and substrate properties along with the attachment strategy. The study provides a novel analytical–experimental comparison of sensor attachment methods, showing how TEA significantly improves uniaxial strain accuracy and reduces substrate dependency in piezoelectric strain measurements.

Structural fatigue can lead to catastrophic failures in various engineering applications and must be properly monitored and effectively managed. This paper provides a state-of-the-art review of recent developments in structural fatigue monitoring using piezoelectric-based sensors. Compared to alternative sensing technologies, piezoelectric sensors offer distinct advantages, including compact size, lightweight design, low cost, flexible formats, and high sensitivity to dynamic loads. The paper reviews the working principles and recent advancements in passive piezoelectric-based sensors, such as acoustic emission wave and strain measurements, and active piezoelectric-based sensors, including ultrasonic wave and dynamic characteristic measurements. These measurements, captured under in-service dynamic strain, can be correlated to the remaining structural fatigue life. Case studies are presented, highlighting applications of fatigue life monitoring in metals, polymeric composites, and reinforced concrete structures. The paper concludes by identifying challenges and opportunities for advancing piezoelectric-based sensors for fatigue life monitoring in engineering structures.

This study evaluates the performance of damaged concrete beams retrofitted with a purpose-designed mechanochromic composite, which provides structural reinforcement and visual feedback for structural health monitoring (SHM). The retrofitting process utilizes externally bonded reinforcement (EBR) on pre-damaged concrete prisms. The mechanochromic composite, a thin-ply hybrid material made of unidirectional ultra-high modulus (UHM) carbon/epoxy and S-glass/epoxy layers, changes color to indicate structural overload when the UHM carbon layer fractures due to excessive strain. Eighteen concrete specimens were prepared and subjected to four-point bending tests, assessing various combinations of damaged, undamaged, retrofitted, and non-retrofitted configurations. Results showed that the mechanochromic composite functions effectively as both a passive visual sensor and reinforcement. For instance, a 5 % crack depth reduced load-bearing capacity by 30 %, however, retrofitting with the mechanochromic composite improved load-bearing capacity by up to 208 % compared to undamaged beams. The study further discusses the effects of different damage levels on load-bearing capacity through flexural strength, load-displacement curves, and failure modes.

This paper investigates the relationship between nanomaterials concentration, scaffold topologies, and mechanical and vibrational performance of additive manufactured Polylactic Acid (PLA) scaffolds reinforced with Graphene Oxide (GO) and Carbon Nanotube (CNT). Three different scaffold topologies namely Cube, Diamond, and Lattice Diamond were designed. Every scaffold comprised PLA reinforced with GO and CNT at different concentration levels of 0 wt%, 0.2 wt%, 0.4 wt%, and 0.6 wt%. The mechanical performance of scaffolds was evaluated via compressive testing. A representative volume element (RVE) model, evaluated using periodic boundary conditions, was developed and successfully validated against experimental results, with a deviation between 13.5 % to 23 %. The novelty of the work lies in incorporating GO and SWCNT agglomeration within the RVE models, enabling a more accurate comparison with 3D-printed composite test samples. Additionally, a comprehensive evaluation of scaffold shapes, geometry, vibrational response, and reinforcement concentration, provides valuable insights into their performance. Experimental and RVE results indicated that the PLAs reinforced with 0.6 wt% GO and 0.6 wt% CNT experience agglomerations. Finite element simulations using the Mooney-Rivlin hyperelastic model showed that the compressive strength of the structures followed this order: Diamond > Cube > Lattice Diamond. The results showed that increasing GO and CNT content from 0 wt% to 0.4 wt% led the Diamond scaffold to exhibit the greatest improvement in compressive strength and elastic modulus, with increases of 79.3 %, 26 %, 165.9 %, and 129 %, respectively. It was found that the additive manufactured modified PLA Diamond scaffold displays the best mechanical performance among the other scaffold topologies. The finite element analysis using the Lanczos Eigensolver revealed that mechanical enhancements and structural topology directly impact natural frequency variations, making them major parameters to consider for specific applications.

Engineering structures, such as bridges, wind turbines, airplanes, ships, buildings, and offshore platforms, often experience uncertain dynamic loadings due to environmental factors and operational conditions. The lack of knowledge about the load spectrum for these structures poses challenges in terms of design and can lead to either over-engineering or catastrophic failure. This research introduces a robust and innovative device, analogous to a "Fitbit" for structures, capable of measuring complex loading conditions throughout the structure's lifespan. The proposed approach involves developing a middleware, referred to as an "extension," which facilitates the transfer of mechanical deformation to a piezoelectric sensor. This approach overcomes challenges associated with directly attaching piezoelectric sensors to the structure's surface such as rupture possibility in higher strain and attaching on rough surfaces. The feasibility study primarily focuses on validating the performance of the extension and monitoring variation trends. The ultimate objective is to develop an Internet of Things (IoT) sensor node capable of measuring applied cyclic loads. To achieve this goal, an electronic system and embedded software will be developed to capture the complex load spectrum and convert it into a fatigue damage index for predicting the structure's fatigue life. The collected data will be transmitted to the user through a wireless communication platform. The proposed sensor design is versatile, allowing for both attachment and embedding and is demonstrated here for monitoring fatigue in engineering structures.

A sensor for visualizing the fatigue load cycles was designed, fabricated, and tested. The sensor is made of a glass/carbon hybrid composite and utilizes the delamination length at the glass/carbon interface as an indicator for fatigue cycles. Appropriate design parameters were obtained by performing finite element analysis on the delamination development at the interface between the glass and carbon layers. Hybrid sensors with different carbon layer thicknesses were manufactured, attached to glass/epoxy substrates, and tested under fatigue loading. The predicted results based on the Paris law for crack extensions in one configuration are compared with the experiments for a different configuration to illustrate the efficacy of the approach.

The present paper reports an overview of mechanochromic self-reporting thin-ply hybrid composite sensors, which are designed to visually indicate overload in structures. These sensors, made from combinations of high-strain and low-strain materials, change appearance earlier than the final fracture, providing a clear visual cue of damage like delamination or strain overload. They can be used for various applications, for example, overload monitoring, and to detect barely visible impact damage in composites.

The ability of fibre reinforced composites to deform with a non-linear stress–strain response and gradual, rather than sudden, catastrophic failure is reviewed. The principal mechanisms by which this behaviour can be achieved are discussed, including ductile fibres, progressive fibre fracture and fragmentation, fibre reorientation, and slip between discontinuous elements. It is shown that all these mechanisms allow additional strain to be achieved, enabling a yield-like behaviour to be generated. In some cases, the response is ductile and in others pseudo-ductile. Mechanisms can also be combined, and composites which give significant pseudo-ductile strain can be produced. Notch sensitivity is reduced, and there is the prospect of increasing design strains whilst also improving damage tolerance. The change in stiffness or visual indications of damage can be exploited to give warning that strain limits have been exceeded. Load carrying capacity is still maintained, allowing continued operation until repairs can be made. Areas for further work are identified which can contribute to creating structures made from high performance ductile or pseudo-ductile composites that fail gradually.

Traditional inspection methods often fall short in detecting defects or damage in fibre-reinforced polymer (FRP) composite structures, which can compromise their performance and safety over time. A prime example is barely visible impact damage (BVID) caused by out-of-plane loadings such as indentation and low-velocity impact that can considerably reduce the residual strength. Therefore, developing advanced visual inspection techniques is essential for early detection of defects, enabling proactive maintenance and extending the lifespan of composite structures. This study explores the viability of using novel bio-inspired hybrid composite sensors for detecting BVID in laminated FRP composite structures. Drawing inspiration from the colour-changing mechanisms found in nature, hybrid composite sensors composed of thin-ply glass and carbon layers are designed and attached to the surface of laminated FRP composites exposed to transverse loading. A comprehensive experimental characterisation, including quasi-static indentation and low-velocity impact tests alongside non-destructive evaluations such as ultrasonic C-scan and visual inspection, is conducted to assess the sensors’ efficacy in detecting BVID. Moreover, a comparison between the two transverse loading types, static indentation and low-velocity impact, is presented. The results suggest that integrating sensors into composite structures has a minimal effect on mechanical properties such as structural stiffness and energy absorption, while substantially improving damage visibility. Additionally, the influence of fibre orientation of the sensing layer on sensor performance is evaluated, and correlations between internal and surface damage are demonstrated.

Gantry cranes play a pivotal role in various industrial applications, and their reliable operation is paramount. While routine inspections are standard practice, certain defects, particularly in less accessible components, remain challenging to detect early. In this study, first a finite element model is presented, and the damage is introduced using random changes in the stiffness of different parts of the structure. Contrary to the assumption of inherent reliability, undetected defects in crucial structural elements can lead to catastrophic failures. Then, the vibration equations of healthy and damaged models are analyzed to find the displacement, velocity, and acceleration of the different crane parts. The learning vector quantization neural network is used to train and detect the defects. The output is the location of the damage and the damage severity. Noisy data are then used to evaluate the network performance robustness. This research also addresses the limitations of traditional inspection methods, providing early detection and classification of defects in gantry cranes. The study’s relevance lies in the need for a comprehensive and efficient damage detection method, especially for components not easily accessible during routine inspections.

This paper investigates the structural performance of flat double-layer grids with various constitutive units, addressing a notable gap in the literature on diverse geometries. Six common types of flat double-layer grids are selected to provide a comprehensive comparison to understand their structural performance. Parametric models are built using Rhino and Grasshopper plugins. Single- and multi-objective optimization processes are conducted on the considered models to evaluate structural mass and maximum deflection. The number of constitutive units, the structural depth, and the cross-section diameter of the members are selected as design variables. The analysis reveals that the semi-octahedron upon square-grid configuration excels in minimizing structural mass and deflection. Furthermore, models lacking a full pyramid form exhibit higher deflections. Sensitivity analyses disclose the critical influence of the design variables, particularly highlighting the sensitivity of structural mass to the number of constitutive units and cross-section diameter. These findings offer valuable insights and practical design considerations for optimizing double-layer grid space structures.

This study proposes a machine learning (ML) model to predict the displacement response of high-rise structures under various vertical and lateral loading conditions. The study combined finite element analysis (FEA), parametric modeling, and a multi-objective genetic algorithm to create a robust and diverse dataset of loading scenarios for developing a predictive ML model. The ML model was trained using a recurrent neural network (RNN) with Long Short-Term Memory (LSTM) layers. The developed model demonstrated high accuracy in predicting time series of vertical, lateral (X), and lateral (Y) displacements. The training and testing results showed Mean Squared Errors (MSE) of 0.1796 and 0.0033, respectively, with R2 values of 0.8416 and 0.9939. The model’s predictions differed by only 0.93% from the actual vertical displacement values and by 4.55% and 7.35% for lateral displacements in the Y and X directions, respectively. The results demonstrate the model’s high accuracy and generalization ability, making it a valuable tool for structural health monitoring (SHM) in high-rise buildings. This research highlights the potential of ML to provide real-time displacement predictions under various load conditions, offering practical applications for ensuring the structural integrity and safety of high-rise buildings, particularly in high-risk seismic areas.

Fatigue Life Monitoring is crucial to ensuring the safety and durability of engineering structures. This paper presents an innovative approach for fatigue life monitoring using a PZT (Lead Zirconate Titanate) piezoelectric sensor operating in buckling mode to measure applied cyclic strains. A significant focus is placed on the utilization of a 3D-printed 'Extension' platform upon which the PZT sensor is installed, allowing it to operate in buckling mode while subjected to cyclic strains. An analytical framework is developed to establish a direct relationship between dynamic strain values and the sensor's output, considering critical strain attributes such as amplitude and strain rate. The analytical solutions are validated through a series of experimental tests conducted under various dynamic loading conditions. Integrating the piezoelectric sensor with the 3D-printed extension demonstrates high sensitivity and serves as a passive dynamic strain measurement sensor, with promising potential for application as sensor nodes in fatigue life monitoring of engineering structures.

The complexity of drilling carbon fiber reinforced polymers (CFRP) requires accurate predictive models. This study addresses the challenge using an ensemble machine learning (ML) approach with stacked generalization. The model captures the relationships between key input variables—such as graphene nanoplatelet (GNP) content, ultrasonic assistance, tool type, stacking sequence, and feed rate—and output parameters, specifically thrust force and delamination. A nested feature scoring (NFS) method was employed for importance analysis, revealing tooling type and feed rate as key features for minimizing delamination and reducing thrust force, respectively. The machinability results revealed that ultrasonic drilling lowered thrust force by improving chip evacuation and reducing fiber breakage. HSS tools with cobalt content, alongside symmetrical stacking sequence, helped to further minimize both thrust force and delamination. However, the inclusion of GNPs led to an increase in thrust force and delamination, attributed to the increased strength of the CFRP/GNP composite. The process involved meticulous training, resulting in four optimal-fit models serving as inputs for the stacked meta-model. Iterative enhancements fortified the ensemble robustness, with fine-tuning of hyperparameters through Bayesian optimization. The ensemble superiority over individual models manifested in a remarkable reduction of mean absolute error (MAE) and root mean squared error (RMSE) by up to 97% and 124% for delamination, and 205% and 154% for thrust force, compared to the best base learner. Visual and statistical assessments effectively illuminated the intricate interactions between variables in the drilling process. The methodology resulted in a highly adaptable predictive model with applications across diverse manufacturing contexts.