This thesis investigates failure in sandwich panels due to bending, specifically localised buckling or wrinkling, a predominant failure mechanism for thin gauge honeycomb sandwich panels loaded in bending or compression. Over the past 60 years, considerable work has been devoted to understanding wrinkling and trying to predict failure loads in damaged and undamaged panels accurately. Existing wrinkling expressions were shown to over-estimate failure loads by over 100%. Discrepancies between wrinkling expressions and experimental failure loads were previously attributed to imperfections and irregularities in the structure. The aim of this thesis is to investigate this problem and try to accurately predict failure loads and understand the underlying failure mechanisms in damaged and undamaged panels, using a combination of numerical and analytical techniques. Classical wrinkling models use a continuum core to model complex cellular honeycomb cores. This type of model reduces complex cellular geometry to a series of effective properties that provide constant support to the face sheet. In reality, honeycomb cores provide support around the periphery of the cell walls and not across the entire surface of the face sheet. Due to the nature of wrinkling and the size of the wavelength, incorrect representation of the core could affect the failure loads and model. This study made direct comparisons between linear buckling loads of a discrete-cored sandwich panel and a continuum-cored sandwich panel. Discrete properties were converted to continuum properties within a Finite Element package. The result conclusively showed that both models predict the same linear failure loads, disproving the theory that the core representations contribute to the difference between experimental and analytical models. It was also shown that existing wrinkling models can accurately predict linear wrinkling loads. These linear model loads do not necessarily match the collapse strength of the physical panel and in most cases predict a significantly higher value. The research then moves on to developing expressions to convert cellular geometry into continuum properties accurately. Expressions are developed for honeycomb structures with fillets in their junctions. Both out-of-plane and in-plane modulus properties are reviewed and the models are verified against Finite Elements models and experimental results. Studies showed that the restrained in-plane modulus can be up to ten times stiffer than the commonly used free modulus value. This has a significant effect on the wrinkling stress. By using the correct value, the discrete model and continuum models predict the same loads. The classical wrinkling expressions also predict the same wrinkling stress as the Finite Element models. After establishing that the core representation is not the cause of the prediction error, the thesis turns to non-linear Finite Element models to predict failure loads and failure mechanism of thin-gauge sandwich honeycomb structures loaded in bending. A continuum three-dimensional non-linear Finite Element model, with bilinear plasticity, is compared with a set of experiments that use different types of Nomex cores and face sheets. The models show that the panels fail prematurely due to core crushing because of wrinkles forming in the face sheets. Experimental results indicate similar trends. The final section examines the affect of impact damage in honeycomb sandwich structures. Due to the thin face sheets and thick cores used on many aircraft and marine components, sandwich panels offer little resistance to impact events. Resulting damage usually consists of a layer of crushed core and a shallow dent in the face sheet. This type of damage often leads to a significant reduction in the load-carrying capacity of the panel through a full range of damage sizes. Finite element and analytical models were developed to accurately predict and capture the localised wrinkling failure mechanism which occurs in the impacted area. Models were directly compared to experimental results, with a high degree of correlation. The numerical and analytical models showed that impact damaged panels were failing due to wrinkling instability and not due to premature core crushing, which is the case with undamaged panels. They showed that two factors influence the wrinkling failure load: damage depth and damage diameter.

This thesis investigates failure in sandwich panels due to bending, specifically localised buckling or wrinkling, a predominant failure mechanism for thin gauge honeycomb sandwich panels loaded in bending or compression. Over the past 60 years, considerable work has been devoted to understanding wrinkling and trying to predict failure loads in damaged and undamaged panels accurately. Existing wrinkling expressions were shown to over-estimate failure loads by over 100%. Discrepancies between wrinkling expressions and experimental failure loads were previously attributed to imperfections and irregularities in the structure. The aim of this thesis is to investigate this problem and try to accurately predict failure loads and understand the underlying failure mechanisms in damaged and undamaged panels, using a combination of numerical and analytical techniques. Classical wrinkling models use a continuum core to model complex cellular honeycomb cores. This type of model reduces complex cellular geometry to a series of effective properties that provide constant support to the face sheet. In reality, honeycomb cores provide support around the periphery of the cell walls and not across the entire surface of the face sheet. Due to the nature of wrinkling and the size of the wavelength, incorrect representation of the core could affect the failure loads and model. This study made direct comparisons between linear buckling loads of a discrete-cored sandwich panel and a continuum-cored sandwich panel. Discrete properties were converted to continuum properties within a Finite Element package. The result conclusively showed that both models predict the same linear failure loads, disproving the theory that the core representations contribute to the difference between experimental and analytical models. It was also shown that existing wrinkling models can accurately predict linear wrinkling loads. These linear model loads do not necessarily match the collapse strength of the physical panel and in most cases predict a significantly higher value. The research then moves on to developing expressions to convert cellular geometry into continuum properties accurately. Expressions are developed for honeycomb structures with fillets in their junctions. Both out-of-plane and in-plane modulus properties are reviewed and the models are verified against Finite Elements models and experimental results. Studies showed that the restrained in-plane modulus can be up to ten times stiffer than the commonly used free modulus value. This has a significant effect on the wrinkling stress. By using the correct value, the discrete model and continuum models predict the same loads. The classical wrinkling expressions also predict the same wrinkling stress as the Finite Element models. After establishing that the core representation is not the cause of the prediction error, the thesis turns to non-linear Finite Element models to predict failure loads and failure mechanism of thin-gauge sandwich honeycomb structures loaded in bending. A continuum three-dimensional non-linear Finite Element model, with bilinear plasticity, is compared with a set of experiments that use different types of Nomex cores and face sheets. The models show that the panels fail prematurely due to core crushing because of wrinkles forming in the face sheets. Experimental results indicate similar trends. The final section examines the affect of impact damage in honeycomb sandwich structures. Due to the thin face sheets and thick cores used on many aircraft and marine components, sandwich panels offer little resistance to impact events. Resulting damage usually consists of a layer of crushed core and a shallow dent in the face sheet. This type of damage often leads to a significant reduction in the load-carrying capacity of the panel through a full range of damage sizes. Finite element and analytical models were developed to accurately predict and capture the localised wrinkling failure mechanism which occurs in the impacted area. Models were directly compared to experimental results, with a high degree of correlation. The numerical and analytical models showed that impact damaged panels were failing due to wrinkling instability and not due to premature core crushing, which is the case with undamaged panels. They showed that two factors influence the wrinkling failure load: damage depth and damage diameter.

This thesis investigates failure in sandwich panels due to bending, specifically localised buckling or wrinkling, a predominant failure mechanism for thin gauge honeycomb sandwich panels loaded in bending or compression. Over the past 60 years, considerable work has been devoted to understanding wrinkling and trying to predict failure loads in damaged and undamaged panels accurately. Existing wrinkling expressions were shown to over-estimate failure loads by over 100%. Discrepancies between wrinkling expressions and experimental failure loads were previously attributed to imperfections and irregularities in the structure. The aim of this thesis is to investigate this problem and try to accurately predict failure loads and understand the underlying failure mechanisms in damaged and undamaged panels, using a combination of numerical and analytical techniques. Classical wrinkling models use a continuum core to model complex cellular honeycomb cores. This type of model reduces complex cellular geometry to a series of effective properties that provide constant support to the face sheet. In reality, honeycomb cores provide support around the periphery of the cell walls and not across the entire surface of the face sheet. Due to the nature of wrinkling and the size of the wavelength, incorrect representation of the core could affect the failure loads and model. This study made direct comparisons between linear buckling loads of a discrete-cored sandwich panel and a continuum-cored sandwich panel. Discrete properties were converted to continuum properties within a Finite Element package. The result conclusively showed that both models predict the same linear failure loads, disproving the theory that the core representations contribute to the difference between experimental and analytical models. It was also shown that existing wrinkling models can accurately predict linear wrinkling loads. These linear model loads do not necessarily match the collapse strength of the physical panel and in most cases predict a significantly higher value. The research then moves on to developing expressions to convert cellular geometry into continuum properties accurately. Expressions are developed for honeycomb structures with fillets in their junctions. Both out-of-plane and in-plane modulus properties are reviewed and the models are verified against Finite Elements models and experimental results. Studies showed that the restrained in-plane modulus can be up to ten times stiffer than the commonly used free modulus value. This has a significant effect on the wrinkling stress. By using the correct value, the discrete model and continuum models predict the same loads. The classical wrinkling expressions also predict the same wrinkling stress as the Finite Element models. After establishing that the core representation is not the cause of the prediction error, the thesis turns to non-linear Finite Element models to predict failure loads and failure mechanism of thin-gauge sandwich honeycomb structures loaded in bending. A continuum three-dimensional non-linear Finite Element model, with bilinear plasticity, is compared with a set of experiments that use different types of Nomex cores and face sheets. The models show that the panels fail prematurely due to core crushing because of wrinkles forming in the face sheets. Experimental results indicate similar trends. The final section examines the affect of impact damage in honeycomb sandwich structures. Due to the thin face sheets and thick cores used on many aircraft and marine components, sandwich panels offer little resistance to impact events. Resulting damage usually consists of a layer of crushed core and a shallow dent in the face sheet. This type of damage often leads to a significant reduction in the load-carrying capacity of the panel through a full range of damage sizes. Finite element and analytical models were developed to accurately predict and capture the localised wrinkling failure mechanism which occurs in the impacted area. Models were directly compared to experimental results, with a high degree of correlation. The numerical and analytical models showed that impact damaged panels were failing due to wrinkling instability and not due to premature core crushing, which is the case with undamaged panels. They showed that two factors influence the wrinkling failure load: damage depth and damage diameter.

This thesis investigates failure in sandwich panels due to bending, specifically localised buckling or wrinkling, a predominant failure mechanism for thin gauge honeycomb sandwich panels loaded in bending or compression. Over the past 60 years, considerable work has been devoted to understanding wrinkling and trying to predict failure loads in damaged and undamaged panels accurately. Existing wrinkling expressions were shown to over-estimate failure loads by over 100%. Discrepancies between wrinkling expressions and experimental failure loads were previously attributed to imperfections and irregularities in the structure. The aim of this thesis is to investigate this problem and try to accurately predict failure loads and understand the underlying failure mechanisms in damaged and undamaged panels, using a combination of numerical and analytical techniques. Classical wrinkling models use a continuum core to model complex cellular honeycomb cores. This type of model reduces complex cellular geometry to a series of effective properties that provide constant support to the face sheet. In reality, honeycomb cores provide support around the periphery of the cell walls and not across the entire surface of the face sheet. Due to the nature of wrinkling and the size of the wavelength, incorrect representation of the core could affect the failure loads and model. This study made direct comparisons between linear buckling loads of a discrete-cored sandwich panel and a continuum-cored sandwich panel. Discrete properties were converted to continuum properties within a Finite Element package. The result conclusively showed that both models predict the same linear failure loads, disproving the theory that the core representations contribute to the difference between experimental and analytical models. It was also shown that existing wrinkling models can accurately predict linear wrinkling loads. These linear model loads do not necessarily match the collapse strength of the physical panel and in most cases predict a significantly higher value. The research then moves on to developing expressions to convert cellular geometry into continuum properties accurately. Expressions are developed for honeycomb structures with fillets in their junctions. Both out-of-plane and in-plane modulus properties are reviewed and the models are verified against Finite Elements models and experimental results. Studies showed that the restrained in-plane modulus can be up to ten times stiffer than the commonly used free modulus value. This has a significant effect on the wrinkling stress. By using the correct value, the discrete model and continuum models predict the same loads. The classical wrinkling expressions also predict the same wrinkling stress as the Finite Element models. After establishing that the core representation is not the cause of the prediction error, the thesis turns to non-linear Finite Element models to predict failure loads and failure mechanism of thin-gauge sandwich honeycomb structures loaded in bending. A continuum three-dimensional non-linear Finite Element model, with bilinear plasticity, is compared with a set of experiments that use different types of Nomex cores and face sheets. The models show that the panels fail prematurely due to core crushing because of wrinkles forming in the face sheets. Experimental results indicate similar trends. The final section examines the affect of impact damage in honeycomb sandwich structures. Due to the thin face sheets and thick cores used on many aircraft and marine components, sandwich panels offer little resistance to impact events. Resulting damage usually consists of a layer of crushed core and a shallow dent in the face sheet. This type of damage often leads to a significant reduction in the load-carrying capacity of the panel through a full range of damage sizes. Finite element and analytical models were developed to accurately predict and capture the localised wrinkling failure mechanism which occurs in the impacted area. Models were directly compared to experimental results, with a high degree of correlation. The numerical and analytical models showed that impact damaged panels were failing due to wrinkling instability and not due to premature core crushing, which is the case with undamaged panels. They showed that two factors influence the wrinkling failure load: damage depth and damage diameter.

Being a widely used structure, composite shells have been studied for a long time. The features of small thickness, heterogeneity, and anisotropy of composite shells have created many challenges for analyzing them. A number of theories have been developed for modeling composite shells, while they are either not practical for engineering use, or rely on assumptions that do not always hold. Consequently, a better theory is needed, especially for the application on challenging problems such as shells involving thermoelasticity, viscoelasticity, or viscoplasticity.<br><br>In this dissertation, a shell theory based on mechanics of structure genome (MSG), a unified theory for multiscale constitutive modeling, is developed. This theory is capable of handling fully anisotropy and complex heterogeneity, and because the derivation follows principle of minimum information loss (PMIL) and using the variational asymptotic method (VAM), high accuracy can be achieved. Both a linear version and a nonlinear version using Euler method combined with Newton-Raphson method are presented. This MSG-based shell theory is used for analyzing the curing process of composites, deployable structures made with thin-ply high strain composite (TP-HSC), and material nonlinear shell behaviors.<br><br>When using the MSG-based shell theory to simulate the curing process of composites, the formulation is written in an analytical form, with the effect of temperature change and degree of cure (DOC) included. In addition to an equivalent classical shell theory, a higher order model with the correction from initial geometry and transverse shear deformation is presented in the form of the Reissner-Mindlin model. Examples show that MSG-based shell theory can accurately capture the deformation caused by temperature change and cure shrinkage, while errors exist when recovering three-dimensional (3D) strain field. Besides, the influence of varying transverse shear stiffness needs to be further studied.<br><br>In order to analyze TP-HSC deployable structures, linear viscoelasticity behavior of composite shells is modeled. Then, column bending test (CBT), an experiment for testing the bending stiffness of thin panels under large bending deformation, is simulated with both quasi-elastic (QE) and direct integration (DI) implementation of viscoelastic shell properties. Comparisons of the test and analysis results show that the model is capable of predicting most of the measured trends. Residual curvature measured in the tests, but not predicted by the present model, suggests that viscoplasticity should be considered. A demonstrative study also shows the potential of material model calibration using the virtual CBT developed in this work. A deployable boom structure is also analyzed. The complete process of flattening, coiling, stowage, deployment and recovery is simulated with the viscoelastic shell model. Results show that major residual deformation happens in the hoop direction.<br><br>A nonlinear version of the MSG-based general purpose constitutive modeling code SwiftComp is developed. The nonlinear solving algorithm based on the combined Euler-Newton method is implemented into SwiftComp. For the convenience of implementing a nonlinear material model, the capability of using user material is also added. A viscoelastic material model and a continuum damage model is tested and shows excellent match when compared with Abaqus results with solid elements and UMAT. Further validation of the nonlinear SwiftComp is done with a nonlinear viscoelastic-viscoplastic model. The high computational cost is emphasized with a preliminary study with surrogate model.

The initiation of high explosives (HEs) under shock loading lacks a comprehensive understanding: particularly at the particle scale. One common explanation is hot spot theory, which suggests that energy in the material resulting from the impact event is localized in a small area causing an increase in temperature that can lead to ignition. This study focuses on the response of HMX particles (a common HE) within a polymer matrix (Sylgard-184^®), a simplified example of a polymer bonded explosive (PBX). A light gas gun was used to load the samples at impact velocities ranging from 370 to 520 m/s. The impact events were visualized using X-ray phase contrast imaging (PCI) allowing real-time observation of the impact event. The experiments used three subsets of PBX samples: multiple particle (production grade and single crystal), drilled hole, and milled slot. Evidence of damage and deformation occurred in all of the sample types. While the necessary impact velocity for consistent hot spot formation leading to reactions was not reached, the damage (particularly cracking) that occurred provides a useful indication of where hot spots may occur when higher velocities are reached. With the multiple particle samples, evidence of cracking and debonding occurred throughout. One sample showed significant volume expansion due to possible reaction. The samples containing drilled holes demonstrated the expected pore collapse behavior at these velocities, as well as damage downstream from the holes under various two-hole arrangements. Milled slot samples were tested to simulate existing cracks in the HMX. These samples showed increased damage at the site of the milled slot, as well as unique cracking behavior in one of the samples.

<div>Sensing of mechanical state is critical in diverse fields including biomedical implants, intelligent robotics, consumer technology interfaces, and integrated structural health monitoring among many others. Recently, materials that are self-sensing via the piezoresistive effect (i.e. having deformation-dependent electrical conductivity) have received much attention due to their potential to enable intrinsic, material-level strain sensing with lesser dependence on external/ad hoc sensor arrays. In order to effectively use piezoresistive materials for strain-sensing, however, it is necessary to understand the deformation-resistivity change relationship. To that end, many studies have been conducted to model the piezoresistive effect, particularly in nanocomposites which have been modified with high aspect-ratio carbonaceous fillers such as carbon nanotubes or carbon nanofibers. However, prevailing piezoresistivity models have important limitations such as being limited to microscales and therefore being computationally prohibitive for macroscale analyses, considering only simple deformations, and having limited accuracy. These are important issues because small errors or delays due to these challenges can substantially mitigate the effectiveness of strain-sensing via piezoresistivity. Therefore, the first objective of this thesis is to develop a conceptual framework for a piezoresistive tensorial relation that is amenable to arbitrary deformation, macroscale analyses, and a wide range of piezoresistive material systems. This was achieved by postulating a general higher-order resistivity-strain relation and fitting the general model to experimental data for carbon nanofiber-modified epoxy (as a representative piezoresistive material with non-linear resistivity-strain relations) through the determination of piezoresistive constants. Lastly, the proposed relation was validated experimentally against discrete resistance changes collected over a complex shape and spatially distributed resistivity changes imaged via electrical impedance tomography (EIT) with very good correspondence. Because of the generality of the proposed higher-order tensorial relation, it can be applied to a wide variety of material systems (e.g. piezoresistive polymers, cementitious, and ceramic composites) thereby lending significant potential for broader impacts to this work. </div><div><br></div><div>Despite the expansive body of work on direct current (DC) transport, DC-based methods have important limitations which can be overcome via alternating current (AC)-based self-sensing. Unfortunately, comparatively little work has been done on AC transport-deformation modeling in self-sensing materials. Therefore, the second objective of this thesis is to establish a conceptual framework for the macroscale modeling of AC conductivity-strain coupling in piezoresistive materials. For this, the universal dielectric response (UDR) as described by Joncsher's power law for AC conductivity was fit to AC conductivity versus strain data for CNF/epoxy (again serving as a representative self-sensing material). It was found that this power law does indeed accurately describe deformation-dependent AC conductivity and power-law fitting constants are non-linear in both normal and shear strain. Curiously, a piezoresistive switching behavior was also observed during this testing. That is, positive piezoresistivity (i.e. decreasing AC conductivity with increasing tensile strain) was observed at low frequencies and negative piezoresistivity (i.e. increasing AC conductivity with increasing tensile strain) was observed at high frequencies. Consequently, there exists a point of zero piezoresistivity (i.e. frequency at which AC conductivity does not change with deformation) between these behaviors. Via microscale computational modeling, it was discovered that changing inter-filler tunneling resistance acting in parallel with inter-filler capacitance is the physical mechanism of this switching behavior.</div>

This work presents findings in the combined experimental and computational study of the effects of anisotropy and microstructure on the behavior of HMX-based energetic materials. Large single crystal samples of β-HMX were meticulously created by solvent evaporation for experimental purposes, and respective orientations were identified via x-ray diffraction. Indentation modulus and hardness values were obtained for different orientations of β-HMX via nanoindentation experiments. Small-scale dynamic impact experiments were performed, and a viscoplastic power law model fit, to describe the anisotropic viscoplastic properties of the crystal. The anisotropic fracture toughness and surface energy of β-HMX were calculated by studying indentation-nucleated crack system formations and fitting the corresponding data to two different models, developed by Lawn and Laugier. It was found that the {011} and {110} planes had the highest and lowest fracture toughnesses, respectively. Drop hammer impact tests were performed to investigate effects of morphology on the impact-induced thermal response of HMX. Finally, the anisotropic properties obtained in this work were applied in a cohesive finite element simulation involving the impact of a sample of PBX containing HMX crystals with varying orientations. Cohesive finite element models were generated of separate microstructure containing either anisotropic (locally isotropic) or global isotropic properties of HMX particle. In comparison, the isotropic model appeared to be more deformation resistant.

<div>Polymer matrices present in composite materials are prone to have time-dependent behavior very sensitive to changes in temperature. The modeling of thermoviscoelasticity is fundamental for capturing the performance of anisotropic viscoelastic materials subjected to both mechanical and thermal loads, or for manufacturing simulation of composites. In addition, improved plate/shell and beam models are required to efficiently design and simulate large anisotropic composite structures. Numerical models have been extensively used to capture the linear viscoelasticity in composites, which can be generalized in integral or differential forms. The hereditary integral constitutive form has been adopted by many researchers to be implemented into finite element codes, but its formulation is complex and time consuming as it is function of the time history. The differential formulation provides faster computation times, but its applicability has been limited to capture the behavior of three-dimensional thermoviscoelastic orthotropic materials.</div><div><br></div><div>This work extends mechanics of structure genome (MSG) to construct linear thermoviscoelastic solid, plate/shell and beam models for multiscale constitutive modeling of three-dimensional heterogeneous materials made of time and temperature dependent constituents. The formulation derives the transient strain energy based on integral formulation for thermorheologically simple materials subject to finite temperature changes. The reduced time parameter is introduced to relate the time-temperature dependency of the anisotropic material by means of master curves at reference conditions. The thermal expansion creep is treated as inherent material behavior. Exact three-dimensional thermoviscoelastic homogenization solutions are also formulated for laminates modeled as an equivalent, homogeneous, anisotropic solid. The new model is implemented in SwiftComp, a general-purpose multiscale constitutive modeling code based on MSG, to handle real heterogeneous materials with arbitrary microstructures, mesostructures or cross-sectional shapes.</div><div><br></div><div>Three-dimensional representative volume element (RVE) analyses and direct numerical simulations using a commercial finite element software are conducted to verify the accuracy of the MSG-based constitutive modeling. Additionally, MSG-based plate/shell results are validated against thin-ply high-strain composites experimental data showing good agreement. Numerical cases with uniform and nonuniform cross-sectional temperature distributions are studied. The results showed that unlike MSG, the RVE method exhibits limitations to properly capture the long-term behavior of effective coefficients of thermal expansion (CTEs) when time-dependent constituent CTEs are considered. The analyses of the homogenized properties also revealed that despite the heterogeneous nature of the composite material, from a multiscale analysis perspective, the temperature dependencies of the effective stiffness and thermal stress properties are governed by the same shift factor as the polymer matrix. This conclusion remains the same for MSG-based solid, plate/shell and beam models with uniform temperature distributions.</div>

<div>Structural health monitoring (SHM) has immense potential to improve the safety of aerospace, mechanical, and civil structures because it allows for continuous, real-time damage prognostication. However, conventional SHM methodologies are limited by factors such as the need for extensive external sensor arrays, inadequate sensitivity to small-sized damage, and poor spatial damage localization. As such, widespread implementation of SHM in engineering structures has been severely restricted. These limitations can be overcome through the use of multi-functional materials with intrinsic self-sensing capabilities. In this area, composite materials with nanofiller-modified polymer matrices have received considerable research interest. The electrical conductivity of these materials is affected by mechanical stimuli such as strain and damage. This is known as the piezoresistive effect and it has been leveraged extensively for SHM in self-sensing materials. However, prevailing conductivity-based SHM modalities suffer from two critical limitations. The first limitation is that the mechanical state of the structure must be indirectly inferred from conductivity changes. Since conductivity is not a structurally relevant property, it would be much more beneficial to know the displacements, strains, and stresses as these can be used to predict the onset of damage and failure. The second limitation is that the precise shape and size of damage cannot be accurately determined from conductivity changes. From a SHM point of view, knowing the precise shape and size of damage would greatly aid in-service inspection and nondestructive evaluation (NDE) of safety-critical structures. The underlying cause of these limitations is that recovering precise mechanics from conductivity presents an under determined and multi-modal inverse problem. Therefore, commonly used inversion schemes such as gradient-based optimization methods fail to produce physically meaningful solutions. Instead, metaheuristic search algorithms must be used in conjunction with physics-based damage models and realistic constraints on the solution search space. To that end, the overarching goal of this research is to address the limitations of conductivity-based SHM by developing metaheuristic algorithm-enabled methodologies for recovering precise mechanics from conductivity changes in self-sensing composites.</div><div><div><br></div><div>Three major scholarly contributions are made in this thesis. First, a piezoresistive inversion methodology is developed for recovering displacements, strains, and stresses in an elastically deformed self-sensing composite based on observed conductivity changes. For this, a genetic algorithm (GA) is integrated with an analytical piezoresistivity model and physics-based constraints on the search space. Using a simple stress based failure criterion, it is demonstrated that this approach can be used to accurately predict material failure. Second, the feasibility of using other widely used metaheuristic algorithms for piezoresistive inversion is explored. Specifically, simulated annealing (SA) and particle swarm optimization (PSO) are used and their performances are compared to the performance of the GA. It is concluded that while SA and PSO can certainly be used to solve the piezoresistive inversion problem, the GA is the best algorithm based on solution accuracy, consistency, and efficiency. Third, a novel methodology is developed for precisely determining damage shape and size from observed conductivity changes in self-sensing composites. For this, a GA is integrated with physics-based geometric models for damage and suitable constraints on the search space. By considering two specific damage modes —through-holes and delaminations —it is shown that this method can be used to precisely reconstruct the shape and size of damage. </div><div><br></div><div>In achieving these goals, this thesis advances the state of the art by addressing critical limitations of conductivity-based SHM. The methodologies developed herein can enable unprecedented NDE capabilities by providing real-time information about the precise mechanical state (displacements, strains, and stresses) and damage shape in self-sensing composites. This has incredible potential to improve the safety of structures in a myriad of engineering venues.</div></div>