Search

In this research, an attention-based feature fusion network (AFFNet), with a backbone residual network (ResNet101) enhanced with two attention mechanism modules, is proposed for automatic pixel-level detection of concrete crack. In particular, the inclusion of attention mechanism modules, for example, the vertical and horizontal compression attention module (VH-CAM) and the efficient channel attention upsample module (ECAUM), is to enable selective concentration on the crack feature. The VH-CAM generates a feature map integrating pixel-level information in vertical and horizontal directions. The ECAUM applied on each decoder layer combines efficient channel attention (ECA) and feature fusion, which can provide rich contextual information as guidance to help low-level features recover crack localization. The proposed model is evaluated on the test dataset and the results reach 84.49% for mean intersection over union (MIoU). Comparison with other state-of-the-art models proves high efficiency and accuracy of the proposed method.

Magnetorheological (MR) grease, which is a type of magneto-sensitive smart material, exhibits complex viscoelastic behavior under different magnetic fields. Storage and loss moduli from oscillatory test are commonly used to characterize such field-dependent viscoelastic behavior. However, they are not able to represent the higher harmonic nonlinearity appeared in oscillatory responses under large amplitude oscillatory (LAOS) shear. In this paper, an investigation is conducted on the characterization of the filed-dependent nonlinearity of MR grease under LAOS by utilizing of Fourier transform (FT)-Chebyshev analysis. To capture the higher harmonic nonlinearity for the modeling and analysis, a comprehensive test program has been conducted. Firstly, MR grease with 70% weight percentage of carbonyl iron particles (MRG-70) is prepared and the stress response of MRG-70 under the oscillatory shear from small to large strain amplitudes with different magnetic field are measured. Then the higher harmonics in stress response signal are used to detect the intercycle nonlinearity of MRG-70 and compare with the traditional analysis method of dynamic modulus. Further, the elastic and viscous measures from FT-Chebyshev analysis are obtained to capture the inter-/intracycle nonlinear behaviors, that is, strain stiffening/softening, shear thickening/thinning, of MRG-70 under LAOS. It is found that, in the presence of magnetic fields, the onset of the intercycle nonlinearities of MRG-70 is originated from shear thickening of MR grease. With the further increase of the shear strain, MRG-70 exhibits the nonlinearities with different combinations of strain stiffening/softening and shear thickening/thinning.

In recent years, negative stiffness vibration isolation device with nonlinear characteristic has become an emerging research area and attracted a significant amount of attentions in the community due to the promising potentials it brought into the field. Its high-static-low-dynamic stiffness property endows the capacity to realize effective vibration isolation and in the meantime to maintain the system stability. This article presents a comprehensive review of the recent research and developments on negative stiffness vibration isolation device. It begins with an introduction on the concept of negative stiffness and then provides a summary and discussion regarding the realization and characteristics of negative stiffness vibration isolation device. The article places its special interest on the principles, structure design, and device characterisation of different types of negative stiffness vibration isolation devices, including spring type, pre-bucked beam type, magnetism type, geometrically nonlinear structural type, and composite structural type. Besides, the applications of negative stiffness vibration isolation device, as well as negative stiffness damper, are summarized and discussed based on the current state-of-the-art. Finally, the conclusions and further discussion provide highlights of the investigation.

Tuned inerter damper (TID) has recently gained increasing attention as a new structural control mechanism for seismic protection of structures. Currently, theoretical investigations are undertaken by researchers to reveal its fundamentals and to understand its underline principles in altering the structural performances of structures against dynamic loadings. However, the comprehensive study of optimization design of TID for undamped structures is lacking and the majority of the research focuses on the optimization of TID for structures without any damping. This research evaluates the [Formula: see text] and [Formula: see text] optimal designs of TIDs on the structures with damping. Using SDOF structure as an example, the frequency response function of the system equipped with TID underground motion excitation is obtained. The [Formula: see text] and [Formula: see text] designs of TID for structures without damping are derived considering various response parameters using analytical method. A numerical search method is utilized for the [Formula: see text] and [Formula: see text] designs of TID for structures with damping; meanwhile, a set of explicit formulae are obtained by curve-fitting for convenience in the application. Finally, the relative motion response of the inerter is explored, and an optimal design formula of TID which can reduce the displacements of primary mass and inerter simultaneously, is proposed.

© Institution of Mechanical Engineers 2016. This paper presents the design and multi-physics optimization of a novel multi-coil magnetorheological (MR) damper with a variable resistance gap (VRG-MMD). Enabling four electromagnetic coils (EMs) with individual exciting currents, a simplified magnetic equivalent circuit was presented and the magnetic flux generated by each voltage source passing through each active gap was calculated as vector operations. To design the optimal geometry of the VRG-MMD, the multi-physics optimization problem including electromagnetics and fluid dynamics has been formulated as a multi-objective function with weighting ratios among total damping force, dynamic range, and inductive time constant. Based on the selected design variables (DVs), six cases with different weighting ratios were optimized using Bound Optimization BY Quadratic Approximation (BOBYQA) technique. Finally, the vibration performance of the optimal VRG-MMD subjected to sinusoidal and triangle displacement excitations was compared to that of the typical multi-coil MR damper.

© The Author(s) 2015. In the past decades, base isolation techniques have become increasingly popular for seismic protection of civil structures owing to its capability of decoupling buildings from harmful ground motion. However, it has been recognised recently that the traditional passive base isolation technique could encounter a serious problem during earthquakes due its incapability in adjusting the isolation frequency to cope with the unpredictability and diversity of earthquakes. To address this challenge, a great deal of research efforts have been conducted to improve traditional base isolation systems, most of which focused on hybrid supplementary devices (passive, active and semi-active types) for the isolators to control displacement or to dissipate seismic energy. On the other hand, the most effective approach to address the aforementioned challenge should lay on varying isolator stiffness in real-time to achieve real-time spontaneous decoupling. A recent advance of the development of an adaptive magneto-rheological elastomer base isolator has brought such idea to reality as the new magneto-rheological elastomer base isolator is capable to alter its stiffness significantly in real-time. In this article, an innovative smart base isolation system employing such magneto-rheological elastomer isolator is proposed and a novel frequency control algorithm is developed to shift the fundamental frequency of the structure away from the dominant frequency range of earthquakes. Such design enables the building to avoid resonant state in real-time according to the on-coming spectrum of the earthquakes. Extensive simulation has been conducted using a five-storey benchmark model with the isolation system, and testing results indicate that the proposed control system is able to significantly suppress both the floor accelerations and inter-storey drifts simultaneously under different earthquakes.

© SAGE Publications. Available magnetorheological elastomer devices normally consist one to two layers of small-size magnetorheological elastomer materials. To be used in large-scale structures, magnetorheological elastomer devices with multiple larger magnetorheological elastomer materials are expected. This article addresses the critical issue in designing a large-scale device with multiple layers of low magnetic conductive magnetorheological elastomer materials, that is, magnetic circuit design. The primary target in magnetic circuit design for magnetorheological elastomer devices is to provide sufficient and uniform magnetic field to all magnetorheological elastomer layers in the device. In this article, finite element investigations are conducted. An innovative magnetic circuit design is proposed for magnetorheological elastomer base isolator with multi-layer of magnetorheological elastomer materials. In the design, laminated magnetorheological elastomer and steel structure is adopted as part of the magnetic core together with two steel blocks. Cylindrical steel tube is used as the yoke of the magnetic circuit. Two plates are placed on the top and bottom of the device to form enclosed magnetic path in the device. Finite element results showed that such innovative magnetic design is able to provide sufficient and uniform magnetic field to all magnetorheological elastomer layers, that is, 25 magnetorheological elastomer layers with thickness of 1 mm and diameter of 120 mm in this case. Finally, the influence of lateral deformation of the magnetorheological elastomer base isolator on the magnetic field is investigated. It is found that the magnetic field in magnetorheological elastomer materials deteriorates when the deformation of the device increases.

© SAGE Publications. Laminated magnetorheological elastomer base isolator is regarded as one of the most promising candidates for realizing adaptive base isolation for civil structures. However, the intrinsic hysteretic and nonlinear behavior of magnetorheological elastomer base isolators imposes challenge for adopting the device to accomplish high-accuracy performance in structural control. Therefore, it is essential to develop an accurate model for symbolizing this unique characteristic before designing a feedback controller. So far, some classical parametric models, such as Bouc-Wen, Dahl, and LuGre, have been proposed to depict the hysteretic response of magnetorheological devices, that is, magnetorheological damper, which may also be used for describing the nonlinear behavior of magnetorheological elastomer base isolator. However, the parameter identification is difficult to implement due to the nonlinear differential equations existing in these models. Considering this problem, this article proposes a nonparametric model, that is, an artificial neural network-based model with 3 input neurons, 18 hidden neurons, and 1 output neuron, to predict the magnetorheological elastomer isolator behavior. In this model, the ant colony algorithm is employed for model training to obtain the optimal weights based on the force-displacement/velocity data sampled from the magnetorheological elastomer isolator. Finally, experimental data are used to validate the effectiveness of the proposed artificial neural network-based model with the good forecasting results.

In this article, an MRD50 type of large-scale magnetorheological shock absorber was designed and manufactured in Smart Materials and Structures Laboratory of Nanjing University of Science and Technology. Upon providing a brief background on magnetorheological dampers, the detailed structure of this developed large-scale magnetorheological shock absorber was depicted. A suit of hardware-in-the-loop simulation platform under high impact load excitation was introduced for a weapon system. A series of tests were conducted to establish the dynamic behaviors of magnetorheological shock absorber under impact loads. The test results show that the inertia damping force should not be ignored like a common magnetorheological damper because of the large acceleration from the impact load. Based on the theory model and the experimental data, index parameters of magnetorheological fluid and other structural parameters in Herschel-Bulkley-Inertia model were identified by using the least square algorithm. In order to evaluate the controllability of large-scale magnetorheological shock absorber applied into high impact loads, three control algorithms, including on-off control, proportional-integral-derivative control, and fuzzy control algorithm, were used in tests to control the dynamic behavior of magnetorheological shock absorber, and some results of the controllability tests were exhibited in this article. In conclusion, the results indicated that the developed large-scale magnetorheological shock absorber was indeed able to effectively control the recoil dynamics. © The Author(s) 2012.

The magnetorheological (MR) shock absorber is one of the most promising new devices for vibration reduction. Many investigations have been carried out on low velocity and frequency applications of MR devices. The use of the MR shock absorber under impact load is of great interest. The now widely used MR damper models, such as the Bingham model, cannot explain sufficiently the shear thinning behavior under impact loads. However, the Herschel–Bulkley model can be used to explain the same. The main purpose of this study is to analyze the behavior of the MR shock absorber under impact load and to verify the analytical conclusions and experiments. First, some dynamic simulations on the MR shock absorber under impact loads in Matlab and Simulink are carried out. The model describes the dynamic characteristics especially the shear thinning behavior of the shock absorber, based on this model, then analyzes the variations of piston acceleration and back cavity pressure of MR shock absorber at both same and different flow indices. A test rig is developed to test the characteristics of the long-stroke MR shock absorber under impact loads. Comparisons between the simulation and the test results are made to validate our conclusions. The results indicate that the peak acceleration value of the piston rod and the pressure of the back cavity are decided by the peak value and the duration of the impact force. The peak value of the acceleration and the cavity pressure cannot be changed, but the transitional time of the acceleration of the piston and the pressure of the back cavity of MR shock absorber can be controlled by changing the applied current in the electromagnetic coil. MR shock absorbers will be the most promising shock reduction device under an impact environment.