Energy-based Method Research Articles

Integrating Machine Learning with the Multilayer Energy-Based Fragment Method for Excited States of Large Systems.

In this work we have combined machine learning techniques with our recently developed multilayer energy-based fragment method for studying excited states of large systems. The photochemically active and inert regions are separately treated with the complete active space self-consistent field method and the trained models. This method is demonstrated to provide accurate energies and gradients leading to essentially the same excited-state potential energy surfaces and nonadiabatic dynamics compared with full ab initio results. Furthermore, in conjunction with the use of machine learning models, this method is highly parallel and exhibits low-scaling computational cost. Finally, the present work could encourage the marriage of machine learning with fragment-based electronic structure methods to explore photochemistry of large systems.

Application of a new energy-based collocation method for nonlinear progressive damage analysis of imperfect composite plates

Abstract In this paper, an efficient simple numerical method, known as energy-based collocation, is presented to analyze the nonlinear behavior of a geometrically imperfect composite plate subjected to compressive in-plane load while taking into account progressive damage phenomena. This method is based on the energy approach, in which the plate is discretized into the Legendre-Gauss-Lobatto points. The nonlinear governing equations are based on the First-order Shear Deformation Theory (FSDT) and the assumptions of large deflection. The system of nonlinear algebraic equations is solved using the quadratic extrapolation technique. Plates with various boundary conditions and different ranges of thickness are investigated. Hashin and Rotem failure criteria are established to predict failure. Several properties degradation models are proposed for the developed failure mechanism and description of ply degradation. The analysis is performed by a computer program developed based on FORTRAN language. The accuracy of the developed procedure is assessed by comparing numerical results taken from previous works.

Topology Optimization of Micro-Structured Materials Featured with the Specific Mechanical Properties

Micro-structured materials consisting of an array of microstructures are engineered to provide the specific material properties. This present work investigates the design of cellular materials under the framework of level set, so as to optimize the topologies and shapes of these porous material microstructures. Firstly, the energy-based homogenization method (EBHM) is applied to evaluate the material effective properties based on the topology of the material cell, where the effective elasticity property is evaluated by the average stress and strain theorems. Secondly, a parametric level set method (PLSM) is employed to optimize the microstructural topology until the specific mechanical properties can be achieved, including the maximum bulk modulus, the maximum shear modulus and their combinations, as well as the negative Poisson’s ratio (NPR). The complicated topological shape optimization of the material microstructure has been equivalent to evolve the sizes of the expansion coefficients in the interpolation of the level set function. Finally, several numerical examples are fully discussed to demonstrate the effectiveness of the developed method. A series of new and interesting material cells with the specific mechanical properties can be found.

Thermo-mechanical stability of single-layered graphene sheets embedded in an elastic medium under action of a moving nanoparticle

Using an energy-based method, this paper sought to analyze dynamic stability and parametric resonance of single-layered graphene sheets (SLGSs) embedded in thermal environment and elastic medium while carrying a nanoparticle moving along an elliptical path. In order to present a realistic model, all inertial effects of the moving nanoparticle are taken into account in the dynamic formulation of the system. Equations governing the transverse vibrations of the embedded SLGS are obtained using the Hamilton's principle. Small-scale effects based on the Eringen's nonlocal elasticity theory are considered in deriving the motion equations. The equations governing the reduced model are calculated based on the Galerkin method. To calculate the instability boundaries, the energy-rate method is applied on the ordinary differential equations (ODEs) governing the system oscillations. The effects of nonlocal parameter, the nanoparticle motion path radii, SLGS length-to-width ratio, temperature change of the thermal environment, stiffness of the elastic medium and boundary conditions of SLGS on the parametric instability regions are examined. The results show that these parameters influence the system stability, so that a decrease in the nonlocal parameter, the SLGS length-to-width ratio and the nanoparticle motion path radii and also an increase in the stiffness coefficients of the elastic medium improve the system stability. The model presented in this paper is validated by comparing the observations with those published in previous studies.

Prediction of Seismic Performance Levels of Corroded Reinforced Concrete Columns as a Function of Crack Width

Abstract An experimental study was conducted for the prediction of the seismic performance levels of corroded reinforced concrete (RC) columns as a function of the initial corrosion crack widths. A full-scale accelerated-corrosion pool was used to corrode 25 full-scale RC columns. The initial crack widths at different levels of corrosion were measured for three different concrete strength levels: 9, 27, and 37 MPa. The seismic performance levels of corroded RC columns under combined cyclic lateral-displacement excursions at two different axial-load ratios (0.20 and 0.40) were measured. The corrosion levels obtained for the initially measured corrosion crack widths were used for establishing a correlation with the lateral-displacement capacities of the RC columns according to an energy-based method. Three empirical models were developed. The first was for predicting the cross-sectional area reduction of reinforcement bars according to the initial corrosion crack widths. The second model was for predicting the percentage of energy capacity of RC columns as a function of the drift ratio and corrosion levels. The third model was for predicting the seismic performance levels of RC columns as a function of the initial corrosion crack widths and was used for an in situ structural assessment. The experimental test results showed that above a 2% drift ratio, all uncorroded RC columns (regardless of the concrete strength levels) exhibited a sudden reduction in energy capacity, and their total energy capacity was consumed. In the case of corrosion, the energy capacity of concrete with high strength levels was prematurely exhausted above a 1.4% drift ratio because of the effects of the corrosion products under the high strength levels of the concrete.

Full-scale experiments of gypsum-sheathed cavity-insulated cold-formed steel walls under different fire conditions

Most previous fire experiments of cold-formed steel (CFS) walls focused on ISO 834 fire. The corresponding results may not represent the actual fire performance of CFS walls in some special circumstances. This paper presents a full-scale experimental investigation of gypsum-sheathed cavity-insulated CFS walls under four different fire conditions. The results show that the axial load has an insignificant effect on the heat transfer of such walls in fire. This realization provides a basis for the indirect coupling method of thermal-mechanical analysis of CFS walls, in which the time-temperature profiles of non-load-bearing CFS walls are adopted in the thermal-mechanical analysis of load-bearing walls. In addition, for CFS walls with the same configuration and load level, different fire conditions do not change the failure mode of steel studs; in particular, the stud hot flanges exhibit almost the same temperature when the structural failure of CFS walls occurs. Hence, a new time equivalency method of CFS walls is proposed, which uses the same temperature of hot flange as the principle of equivalency. Meanwhile, both the equal area method and energy-based method are modified by using reference temperatures, and give a good prediction of the fire resistance time equivalency of CFS walls. Moreover, significant opening of gypsum board joints is observed during the fire exposure; however, this phenomenon is an inherent characteristic of heated gypsum boards and has insignificant correlation with the out-of-plane deflection of CFS walls. Furthermore, staggered board joints are recommended as an important and effective fire resistance configuration of CFS walls.

Micro indentation fracture of cement paste assessed by energy-based method: The method improvement and affecting factors

This paper reported an improved energy-based method to assess the fracture properties of a hardened cement paste (HCP) by using micro indentation technique. The model was initiated by the additionally generated fracture energy that is assumed to be averagely distributed during indentation, and only involves the load–displacement curves that follow power functions. An explicit equation for estimating the plastic energy considering the effect of holding procedure that has not been considered before was established, which, together with the principle of energy balance, allows us to assess the fracture energy, energy release rate and fracture toughness of the HCP. The accuracy, reliability and robustness of the model were validated by the experimental data with changing the maximum load, loading speed and holding time. Overall, the micro fracture testing method developed here provides an efficient way to assess the fracture properties of cement-based materials in micro scale.

Properties of Fiber-Reinforced Structural and Non-Structural Ultra Lightweight Aggregate Concrete

This study is focused on studying the effect of synthetic short fibers on the mechanical properties of ultra lightweight aggregate concrete (ULWAC). Both non-structural and structural fiber-reinforced ULWAC were considered. The data were collected from different studies, including 15 design papers submitted by universities in the USA and Canada to ASCE National Concrete Canoe Competition (NCCC) with 23 different ULWAC mixes. The data were analyzed, and new modified equations to determine the modulus of elasticity and modulus of rupture were proposed that could improve the accuracy of the current ACI equations. The statistical parameters of fiber reinforced structural ULWAC were determined, and high cost was associated with this type of concrete. Ductility indices for plates and beams, that have the capability to exhibit strain-hardening prior to failure, were calculated using energy-based method, and they were above the minimum required value of 3. Finally, a new model was proposed to predict the complete stress-strain behavior of fiber reinforced ULWAC under axial compression.

Load-slip behaviour of glue laminated timber connections with glued-in steel rod parallel to grain

This paper reports experiments performed on glulam connections with glued-in single steel rod to investigate the joint behaviour. A total of 68 glulam joints with glued-in steel rod were tested via pull-pull tests. Experimental variables include the bar diameter, bonded length, glue-line thickness, angle with respect to the grain, species, adhesive type, and rod type. The results indicated that the failure modes were classified as rod pull-out failure, rod yielding, timber splitting, and timber shear failure. The ultimate capacity and stiffness of the joints were mainly influenced by the rod diameter, bonded length, adhesive type, and rod type. Once the bonded length was ＞12.5d (rod diameter), the glued-in rebar joints usually failed by bar yielding and both the angle with respect to the grain and the bond-line thickness showed little influence on the joint ultimate capacity and stiffness. The joint ductility coefficient calculated based on the proposed equivalent energy-based method ranges from 2.9 to18.0 indicating that the glued-in steel rod timber joints can show considerable ductility with reasonable design. Finally, a proposed bond–slip model considering the rod-yielding effect was used to predict the bond–slip behaviour of the tested joints.

A new failure criterion for determining the cyclic resistance of low-plasticity fine-grained tailings

Reasonable determination of the cyclic resistance of low-plasticity fine-grained tailings is important for the engineering design of new tailings dams. In this paper, a series of monotonic and cyclic triaxial undrained tests were conducted to investigate the mechanical response of low-plasticity fine-grained tailings (Taiping tailings). A unified Viscous Energy Dissipation Ratio (VEDR) was then used to develop an energy-based method to study the failure mode under cyclic loading. The VEDR is a normalized energy ratio based on the relationship between cyclic stress and strain. Collapse failure was observed when the dynamic properties suddenly changed beyond the peak point of VEDR (VEDRpeak). These changes included a sudden decrease in VEDR, an abrupt increase in cyclic amplitude strain (εcyc), and the prominent failure pattern of hysteresis curve. In this regard, VEDR failure criterion was proposed by considering VEDRpeak as the failure point. Comparisons based on the effective stress path and the cyclic resistance ratio (CRR) indicate that the VEDR failure criterion is more suitable for determining the CRR than conventional strain-based failure criteria for Taiping tailings. Furthermore, a relationship between the VEDR failure criterion and strain-based failure criterion was introduced. In general, the VEDR failure criterion will provide an insight into the determination of CRR for low-plasticity fine-grained tailings.

Video anomaly detection and localization via multivariate gaussian fully convolution adversarial autoencoder

In this paper, we present a novel deep learning based method for video anomaly detection and localization. The key idea of our approach is that the latent space representations of normal samples are trained to accord with a specific prior distribution by the proposed deep neural network - Multivariate Gaussian Fully Convolution Adversarial Autoencoder (MGFC-AAE), while the latent representations of anomalies do not. In order to extract deep features from input samples as latent representations, a convolutional neural network (CNN) is employed for the encoder of the deep network. Based on the probability that the test sample is associated with the prior distribution, an energy-based method is applied to obtain its anomaly score. A two-stream framework is utilized to integrate the appearance and motion cues to achieve more comprehensive detection results, taking the gradient and optical flow patches as inputs for each stream. Besides, a multi-scale patch structure is put forward to handle the perspective of some video scenes. Experiments are conducted on three public datasets, results verify that our framework can accurately detect and locate abnormal objects in various video scenes, achieving competitive performance when compared with other state-of-the-art works.

Topology optimization of 2-D mechanical metamaterials using a parametric level set method combined with a meshfree algorithm

Two-dimensional (2-D) micro-architectured mechanical metamaterials are designed using a topology optimization approach that integrates a parametric level set method (PLSM) with a meshfree method based on compactly supported radial basis functions (CS-RBF). The PLSM is employed as the optimization algorithm to achieve desired microstructures with targeted material properties. The effective elastic properties, including the bulk modulus, shear modulus and Poisson’s ratio, are predicted using a strain energy-based homogenization method and the CS-RBF meshfree algorithm. Two sets of optimizations are implemented: one is for the case of a single solid material, and the other is for the case of two solid materials, each involving an additional void phase. Three numerical examples are provided for each case with the same optimization objectives: maximizing the effective bulk modulus, maximizing the effective shear modulus, and minimizing the effective Poisson’s ratio under given volume fraction constraints. The numerical results reveal that the newly proposed approach can generate smooth topological boundaries and optimal microstructures. In particular, the current method can topologically optimize auxetic metamaterials with a negative Poisson’s ratio. It can also be extended to design other periodic metamaterials, including those with a negative coefficient of thermal expansion or frequency bandgaps.

A Sustainability 3D Framework of the 20 Regions of Italy and Comparison With World Countries

An Input-State-Output framework has been recently introduced to investigate the multidimensional aspects of sustainability (namely environmental, social and economic ones) of economic systems through a thermodynamically and logically ordered scheme. This approach facilitates the convergence of information from sets of indicators without aggregating results into single numbers and, consequently, losing information. In this paper we experiment the application of the Input-State-Output framework for regional systems. In particular the 20 regions of Italy will be investigated, categorized and compared with each other. A triad of indicators is selected and calculated to represent the dimensions of sustainability. For the environmental dimension (input) we use Emergy, an energy-based accounting method that takes into account the flows of energy and matter feeding the system per unit time; for the social dimension (state) we use the Gini Index of income distribution as a proxy of the health status of a society as a whole and its organization; for the economic dimension (output) we use the Gross Regional Product or a similar measure, which represents the economic result of the system. As a general result, we identify and discuss the position of the systems under study, by means of the three coordinates within a 3D space, as well as the weaknesses and strengths of each indicator. We believe that this approach is able to give a more complete and rational view of sustainability. The application at the regional level can be of great help in identifying priorities and possible policies at both national and sub-national level.

Rock brittleness evaluation based on energy dissipation under triaxial compression

Evaluating the brittleness of rock is significant for strategy determination of hydraulic fracturing, such as candidate selection and parameter optimization. A series of definitions and indices of brittleness have been proposed to characterize the mechanical behavior rock failure. However, these existing indices failed to consider the residual state and the confinement effect. Based on the analysis of energy evolution during the whole process of rock failure, the brittleness was redefined in this study as the comprehensive capability of dissipating little energy during the pre-peak stage and self-sustaining complete failure during the post-peak stage. Accordingly, a new energy-based brittleness index B was proposed in terms of the complete stress-strain curve to quantify this capability from following three aspects: the ratio of accumulated elastic strain energy and total absorbed energy during the pre-peak stage (B1); the proportion of elastic strain energy in all energy source consumed for sustaining rock failure (B2); and the dissipation extent of accumulated elastic strain energy during the post-peak stage (B3). To verify the reliability of the new method, uniaxial and triaxial compression tests were performed on different types of rock samples. The application and comparison of various indices showed that the new brittleness index precisely characterized the stress-strain curves and failure behavior of rock samples under different confinement levels. The variation trends of brittleness with confining pressure were obviously distinct among different rock types. Three independent brittleness indices B1, B2, and B3 were helpful for analyzing sensitivity difference of brittleness to confining pressure among different rock types. Accordingly, this new energy-based method can provide reliable evaluation of rock brittleness.

Concurrent topology optimization of multiscale composite structures in Matlab

This paper presents the compact and efficient Matlab codes for the concurrent topology optimization of multiscale composite structures not only in 2D scenario but also considering 3D cases. A modified SIMP approach (Sigmund 2007) is employed to implement the concurrent topological design, with an energy-based homogenization method (EBHM) to evaluate the macroscopic effective properties of the microstructure. The 2D and 3D Matlab codes in the paper are developed, using the 88-line 2D SIMP code (Struct Multidisc Optim 43(1): 1–16, 2011) and the 169-line 3D topology optimization code (Struct Multidisc Optim 50(6): 1175–1196, 2014), respectively. This paper mainly contributes to the following four aspects: (1) the code architecture for the topology optimization of cellular composite structures (ConTop2D.m and ConTop3D.m), (2) the code to compute the 3D iso-parametric element stiffness matrix (elementMatVec3D.m), (3) the EBHM to predict the macroscopic effective properties of 2D and 3D material microstructures (EBHM2D.m and EBHM3D.m), and (4) the code to calculate the sensitivities of the objective function with respect to the design variables at two scales. Several numerical examples are tested to demonstrate the effectiveness of the Matlab codes, which are attached in the Appendix, also offering an entry point for new comers in designing cellular composites using topology optimization.

Wavelet energy-based stable and unstable power swing detection scheme for distance relays

Distance relays are susceptible to maloperation during postfault power oscillations in the system. This unintended operation of relays may lead to power system blackout. It is due to their inability to distinguish a stable from an unstable power swing and to take a tripping decision appropriately. This research paper proposes a fast and wavelet energy-based method to detect stable and unstable power swings for distance relays. The proposed method computes the angular velocity of an equivalent machine developed from the measurements available at a bus. The energy in the low frequency band of the equivalent machine angular velocity will assist the tripping decisions of transmission line relays during power swings. Compared to the conventional method using blinders, the advantage of this method is that it requires less hardware and less time to know whether the resulting power oscillations after fault clearing are going to be stable or not. The performance of the proposed method has been evaluated on 5-bus system, WSCC 9-bus system, and IEEE 14-bus and 30-bus systems considering different test scenarios. The results show that using this method improves the performance of distance relays in detecting stable and unstable power swings.

Specific cutting energy index (SCEI)-based process signature for high-performance milling of hardened steel

The specific cutting energy can characterize the machinability of the cutter for a material, and its change is a reflection of the size effect in the metal cutting process. It is critical to study the cutting process using an energy-based processing signature method with the purpose of improving the machining performance. In this paper, a specific cutting energy calculation method is presented based on a mechanistic model, and an exponential function is used to describe the trend in the specific cutting energy with the average undeformed chip thickness. A dimensionless index with value of greater than 1, referred to as the Specific Cutting Energy Index (SCEI), is proposed to address the energy efficiency of the milling process and reflect the machining performance for different machining parameters. Machining experiments with 300 M steel are conducted to validate the effectiveness of the proposed model. Analyses are performed to determine the relationships between SCEI and the material removal mode, chip morphology, tool wear rate, and surface roughness. Based on these results, a feed rate scheduling method is proposed to obtain optimal machining strategies using SCEI, which is an effective way to achieve high-performance machining.

Thermal performance of non-load-bearing cold-formed steel walls under different design fire conditions

Most previous fire experiments of cold-formed steel (CFS) walls focused on the ISO 834 or the E119 fire curve. Although the fire performance of CFS walls under ISO 834 or E119 curve can be theoretically converted to that under other fire curves, the corresponding time equivalency methods have not been experimentally examined. This paper presents an experimental investigation of non-load-bearing CFS walls under four different fire conditions. It is proved by the experiments that the maximum temperature of the design fire curve is one of the most important indexes for representing the severity of fire exposure. Especially, the maximum temperature of 680 °C for external fire was not severe to the present non-load-bearing CFS walls. Two specimens were subjected to 285 min of external fire but demonstrated an almost steady-state heat transfer process. Meanwhile, the hydrocarbon fire and realistic design fire demonstrated much more severe damage. It is also proved for the first time that neither the equal area method nor the energy-based method was suitable for the fire-resistance time equivalency of CFS walls. Based on post fire observation, significant local crushing of studs was identified, especially after the fall off of fire-side sheathing. Additionally, it was found that replacing the fire-side base-layer gypsum plasterboard with low-bulk-density calcium-silicate board extends the insulation and structure performance of CFS walls under fire conditions. Moreover, a simplified temperature distribution model and the critical temperature on the fall off of fire-side gypsum plasterboard were recommended for the numerical modeling of CFS walls.

Establishing a Framework of Using Residue-Residue Interactions in Protein Difference Network Analysis.

Detailed understanding of interactions between amino acid residues is critical in using promising difference network analysis approaches to map allosteric communication pathways. Using experimental data as benchmarks, we scan values of two essential residue-residue contact parameters: the distance cutoff (dc) and the cutoff of residue separation in sequence (nc). The optimal dc = 4.5 Å is revealed, which defines the upper bound of the first shell of residue-residue packing in proteins, whereas nc is found to have little effects on performance. We also develop a new energy-based contact method for network analyses and find an equivalency between the energy network using the optimal energy cutoff ec = 1.0 kBT and the structure network using dc = 4.5 Å. The simple 4.5-Å contact method is further shown to have comparable prediction accuracy to a contact method using amino acid type-specific distance cutoffs and chemical shift prediction-based methods. This study provides necessary tools in mapping dynamics to functions.

Finite element modeling of the puncture-cutting response of soft material by a pointed blade

Finite element analysis of combined puncture and cutting of a soft substrate by pointed blades allows further progress in investigating material failure properties, as well as in determining the stress state during the insertion process of the pointed blade. In the current work, a finite element model of the puncture-cutting test was developed using ABAQUS explicit scheme, assuming a non-linear hyperelastic neoprene rubber-like material. Based on the phenomenological model of Mooney-Rivlin, this study suggests an energy-based method, which was derived from both the curve of the puncture cutting force and the pointed blade displacement. The required energy was computed from the work variation as a function of the occurred fracture surface area. The obtained results revealed that finite element analysis was able to accurately predict the material resistance to a combined puncture and cutting motion. Moreover, the stress field analysis showed that the puncture cutting of soft substrates involved a combined stress state, generating then a mixed fracture mode.