Search

A BCMO-ANN algorithm for vibration and buckling optimization of functionally graded porous (FGP) microplates is proposed in this paper. The theory is based on a unified framework of higher-order shear deformation theory and modified couple stress theory. A combination of artificial neural network (ANN) and balancing composite motion optimization (BCMO) is developed to solve the optimization problems and predict stochastic vibration and buckling behaviors of functionally graded porous microplates with uncertainties of material properties. The characteristic equations are derived from Hamilton's principle and approximation of field variables under Ritz-type exponential series. Numerical results are obtained to investigate the effects of the material distribution, material length scale, porosity density and boundary conditions on natural frequencies and critical buckling loads of functionally graded porous microplates. The novel results derived from this paper can be used as future references.

Structural damage identification has been researched for a long time and continues to be an active research topic. This paper proposes the use of the natural frequencies of a novel composite structures made of glass fibre reinforced polymer (GFRP). The proposed methodology consists of an improved Artificial Neural Network (ANN) using optimization algorithms to detect the exact crack length. In the first step, the characterization of fabricated material is provided to determine Young's modulus using an experimental static bending test, tensile test and modal analysis test. Next, numerical validation is performed using commercial software ABAQUS to extract more data for different crack locations in the structure. The comparison between experimental and numerical results shows a good agreement. ANN has been improved using recent optimization techniques such as Jaya, enhanced Jaya (E-Jaya), Whale Optimization Algorithm (WOA) and Arithmetic Optimization Algorithm (AOA) to calibrate the influential parameters during training. After considering several scenarios, the results show that the accuracy of E-Jaya is better than other optimization techniques. This study on crack identification using improved ANN can be used to investigate the safety and soundness of composite structures.

The main cable is the primary load-bearing component of a long-span multi-tower suspension bridge. The interaction between a dead load, vehicle load, wind load, and the corrosion environment leads the main cable wire to exhibit tribo-corrosion-fatigue behaviors. This behavior causes wire wear and deterioration, as well as a reduction in the effective cross-sectional area. This leads to the gradual deterioration of the wire’s load-bearing strength and seriously affects the load-bearing safety of the main cable. In order to ensure the safety of suspension bridges, it is critical to investigate the gradual deterioration behavior of the main cable wire’s load-bearing strength. A wire tribo-corrosion-fatigue test rig was established to test the wire under different friction pairs (saddle groove or parallel wires). The cross-sectional failure area of the wire with different pairs was obtained by super-depth electron microscopy and calculation. The damage degree evolution model and the deterioration model of the wire load-bearing strength were established by combining the theory of damage mechanics and the finite element method. The results show that, as contact and fatigue loads increase, so does the cross-sectional failure area of the fatigue steel wire. The fatigue wire’s damage degree has a good quadratic function relationship with fatigue cycles. The damage degree of the wire increases and the load-bearing strength decreases with increasing contact load and fatigue load. The load-bearing strength of the wire changes little at the beginning and decreases with increasing fatigue cycles. The results have fundamental significance for the life prediction of the main cable wires of suspension bridges.

There are a huge number of bolted joints used in engineering applications. However, unreasonable design and use may lead to degradation and failure of the connected structure. Herein, finite element analysis method is used to investigate the influence of various factors on the mechanical properties of bolted joints. To simulate the actual assembly and loading process, different analysis steps were considered, given as: tightening step, initial lose step and the loading step. In the loading step, transverse load is applied to obtain the force-displacement curve. Structural parameters, i.e. clamping force, effective clamp length and assembly clearance, and co-efficients of friction of all the contact surfaces are considered in the finite element analysis. The results showed that the performance of the bolted joint under transverse loading exhibits different stages, and this agrees well with the results of our previous experimental results. Through the analysis of the force-displacement curves, the characterization parameters for the mechanical properties of the bolted joint can be extracted, such as curve form, tangential stiffness, slip distance, transverse load threshold, etc. The characterization and evaluation of the mechanical properties integrates the process from design to service, which helps to provide reference for the reliability design and performance improvement of bolted joints.

An inverse input–output method is proposed for long-term condition monitoring of civil infrastructures through monitoring the prediction error of air temperature recorded at the site of a structure. It is known that structural natural frequencies are affected by temperature. Hence, the proposed method considers the structural natural frequencies as input and temperature as output to train a machine learning algorithm (MLA). To this end, after signal preprocessing using the variational mode decomposition (VMD), different MLAs are employed, and the error associated with this prediction is regarded as damage–sensitive feature. It is hypothesised and further confirmed through solving numerical and benchmark problems that the prediction error deviates significantly from the upper bond control limit of an R-chart (errors signal) constructed based on the prediction error of temperature as soon as the damage occurs. The frequency–temperature scatter plots indicate a linear dependency between the natural frequencies and temperature. Moreover, the similar slope obtained for the regression line fitted to different frequency–temperature scatter plots indicates high collinearity among pairs of natural frequencies. This observation implies that an interaction term must be considered for such pairs of natural frequencies in the linear regression model. The results of both numerical and experimental studies further confirm that the interaction linear regression model is the most accurate machine learning algorithm for solving the inverse problem of predicting temperature using natural frequencies for condition monitoring of structures. The results of the proposed method are also compared with the direct strategy, whereby its superiority is demonstrated.

The Gurson–Tvergaard–Needleman damage model (GTN) describes the three stages of ductile tearing of steel: nucleation, growth and coalescence of micro-voids. This work is divided into two main parts. In the first part, based on the inverse analysis and the comparison between the experimental and numerical data, the parameters of the GTN damage model in conjunction with the hardening law are determined. The identification is broadened to include a considerable number of experimental tests drawn from our previous works and other works done at ALFAPIPE Ghardaia laboratory. In the second part, an Artificial Neural Network model is developed to predict the parameters of the (GTN) model coupled with the hardening law that goes through the prediction of traction and impact properties of API X70 steel pipe depending on its chemical composition. The weight of the chemical elements in percentages is considered as the inputs and the GTN parameters are considered as the outputs. In order to validate the obtained ANNGTN parameters, traction and impact tests are simulated. The numerical results are compared with the experimental ones and revealed that the developed model is very precise and has the potential to capture the interaction of GTN parameters coupled with hardening law and chemical composition of steel pipelines.

The paper aims to determine the mechanical properties of the adhesive joints made with acrylic pressure-sensitive adhesives. Two types of double-sided acrylic pressure-sensitive adhesive tapes are used. Three construction materials are used to prepare the adhesive joints: structural steel sheet (C45), aluminium alloy sheet (EN-AW 5754), and titanium sheet (Grade 2). Strength tests of adhesive joints made with the pressure-sensitive adhesive tapes are carried out both after conditioning time at room temperature (23 °C) and subjected to thermal shocks (500 cycles: +60 °C / –40 °C). Strength tests are carried out based on the DIN EN 1465 standard on a Zwick/Roell Z150 testing machine. The main conclusion from the tests carried out was the positive effect of thermal shocks on the mechanical strength of joints bonded with pressure-sensitive adhesive tape.

Over recent decades, the artificial neural networks (ANNs) have been applied as an effective approach for detecting damage in construction materials. However, to achieve a superior result of defect identification, they have to overcome some shortcomings, for instance slow convergence or stagnancy in local minima. Therefore, optimization algorithms with a global search ability are used to enhance ANNs, i.e. to increase the rate of convergence and to reach a global minimum. This paper introduces a two-stage approach for failure identification in a steel beam. In the first step, the presence of defects and their positions are identified by modal indices. In the second step, a feedforward neural network, improved by a hybrid particle swarm optimization and gravitational search algorithm, namely FNN-PSOGSA, is used to quantify the severity of damage. Finite element (FE) models of the beam for two damage scenarios are used to certify the accuracy and reliability of the proposed method. For comparison, a traditional ANN is also used to estimate the severity of the damage. The obtained results prove that the proposed approach can be used effectively for damage detection and quantification.

7xxx aluminium series reach exceptional strength compared to other industrial aluminium alloys. However, 7xxx aluminium series usually exhibit Precipitate-Free Zones (PFZs) along grain boundaries, which favour intergranular fracture and low ductility. In this study, the competition between intergranular and transgranular fracture is experimentally investigated in the 7075 Al alloy. This is of critical importance since it directly affects the formability and crashworthiness of thin Al sheets. Using Friction Stir Processing (FSP), microstructures with similar hardening precipitates and PFZs, but with very different grain structures and intermetallic (IM) particle size distribution, were generated and studied. Experimental results showed that the effect of microstructure on the failure mode was significantly different for tensile ductility compared to bending formability. While the tensile ductility was significantly improved for the microstructure with equiaxed grains and smaller IM particles (compared to elongated grains and larger particles), the opposite trend was observed in terms of formability.