Experimental Techniques最新文献

Due to the complex structure of most frame structure, a large amount of sensor data needs to be processed for damage diagnosis, which increases the computational cost of diagnosis models and poses a serious challenge to their fast, accurate, and efficient damage diagnosis. In order to address this issue, this paper proposes a novel lightweight damage diagnosis method of frame structure for mobile devices based on convolutional neural networks. This method first uses mean filtering to process the vibration data collected by sensors, and then innovatively combines two convolutional neural network models, ShuffleNet and GhostNet, to form a new lightweight convolutional neural network model called SGNet, thereby reducing the computational cost of the model while ensuring diagnosis accuracy. In order to test the performance of the method proposed in this article, experimental research on damage degree diagnosis and damage type diagnosis is conducted by taking the frame structure provided by Columbia University as the research object, and comparative experiments of performance are conducted with MobileNet, GhostNet, and ShuffleNet. The experimental results show that the lightweight damage diagnosis method for frame structure proposed in this article not only has high damage diagnosis accuracy, but also has fewer computational parameters, when the highest accuracy is 99.8%, the computational parameters are only 1 million. At the same time, it is superior to MobileNet, GhostNet, ShuffleNet in terms of diagnosis accuracy and computational cost, so it is an effective high-precision lightweight damage diagnosis method for frame structure.

Effective mass models are a powerful framework in mechanical design and qualification. They are a specialized modal model that use a base-acceleration input to yield the response of a base-mounted structure in single-axis dynamic environments. Applications include the computation of reaction forces of each mode of the base-mounted payload, the identification of the most “important” modes for loads analysis, and many others. Originally derived from a finite element model of the structure, many experimental techniques for computing effective mass models have also been developed. This work describes the derivation of a novel technique which utilizes the Craig-Bampton Modal framework to extract an effective mass model from a modal test and provides a much simpler formulation than previous methods while maintaining the accuracy of the computed effective mass model. Analytical and experimental demonstrations are provided along with practical recommendations for successful implementation.

Substructuring is a term used to describe the estimation of the dynamics of a coupled system assembly when only the dynamics of each uncoupled component is available. Existing approaches allow for the coupling of physical-to-physical models, physical-to-modal models, modal-to-modal models referred to as Component Mode Synthesis (CMS), and impedance-to-impedance models referred to as Frequency Based Substructuring (FBS). Often times, the component information may not be just modal data for both components or just FRF data for both components so that modal substructuring or FRF substructuring can be performed. In these cases, the component data needs to be converted from either modal data or FRF data to match the data of the other component. A method for directly coupling impedance- and modal-based components has not yet been addressed. A proposed Impedance to Modal Substructuring (IMS) approach addresses this situation by writing the equations in a form that allows the user to directly utilize modal data for one component and FRF data for the other component, offering more flexibility in coupling different component data sets. While intended to be used with experimental data, this approach may also implement analytical components. In this work, an approach was developed to allow for the direct coupling of impedance and modal models without the need for the user to convert component data type. The IMS approach derived in this work was validated using analytical and experimental data with various models.

The design and development of Indium Tin Oxide (ITO) thin film based piezoresistive pressure sensor is presented in this paper. ITO (90:10) nanoparticles were synthesized by green combustion method using indium and tin as precursors and, carica papaya seed extract as fuel. ITO (90:10) thin film piezoresistors were deposited using synthesized nanoparticles on AlN coated circular steel (SS 304) diaphragm using E-beam evaporation technique. Diaphragm models of different thickness (0.75, 1 and 1.25 mm) were created using ANSYS finite element analysis in order to determine the maximum stress and deflection region for applied pressure of 1 to 10 bar. ANSYS results exhibited that maximum stress and deflection occurred at the center and circumference of diaphragm. ITO thin film piezoresistors were deposited at these regions using mechanical mask. TiW metal contact was established to these ITO thin film piezoresistors using DC sputtering method. ITO thin film piezoresistive pressure sensor with TiW contact connected in Wheatstone full bridge configuration was calibrated and tested for 50 pressure cycles by applying 2 V DC supply. Sensitivity (S) of the developed ITO thin film pressure sensor was obtained as 0.686, 0.566 and 0.495 mV/bar for diaphragm thickness of 0.75, 1, and 1.25 mm pressure sensors respectively. The non-linearity (NLi) in the output response of the pressure sensors was found to be 9.14, 9.82 and 11.27% for diaphragm thickness of 0.75, 1, and 1.25 mm respectively. Hysteresis errors were found to be 0.0344, 0.0525 and 0.054 for diaphragm thickness of 0.75, 1, and 1.25 mm respectively.

The sandwich panels are widely used in many industrial applications due to their high mechanical properties. Their core design is most important parameter in enhancing their mechanical strength. Flexibility in the design of the core structure leads to the achievement of high strength and light structures. In this paper, the results of the optimized geometry in the previous work are used to investigate the capability of the core geometry design with different materials. Therefore, using the different materials, the peak enhancement of strength-to-weight ratio in sandwich panels besides core behavior during pressure testing are investigated. To this end, a new lattice core is brought forth as the first level; then, three types of materials including AL3105, glass, and innegra fiber/epoxy composites are used to fabricate the cores, in order to compare the compressive strength and the peak. The Nano-clay cloisite 20A is also utilized in construction of sandwich panels. The result indicates that the AL3105 lattice core has the highest strength-to-weight ratio, while the innegra fiber composite core has the highest toughness. Applying curve studies and the SEM Fig. 13, it is concluded that the addition of Nano-clay to composites leads to an increase in both of the strain and the core strength. Comparing the results of experimental and finite element modeling (FEM) data (in ABAQUS software) represented that there is a suitable compliance between them. Our results with the positional variation in core design can pave way in designing advanced engineered sandwich structures in aerospace, shipping, automotive industries. Therefore, these structures will have wide applications in the field of light structure, heat and fluid transfer, sound and vibration control.

Hybrid simulation is an innovative method that combines an analysis model of a structural system with physical tests of one or more substructures. The analysis model is typically a finite element analysis (FEA) model that outputs displacements applied to the physical substructure using a control system operated in displacement control. For stiff specimens, the displacement commands can be so small that the control system has difficulty imposing the command displacements accurately. To do hybrid simulation with a stiff specimen, force control is desirable. Cascade control, which features two layers of closed loop control, is proposed to address this issue. The inner control loop has force control mode that provides accurate control for hybrid tests with stiff specimens. The outer control loop is in displacement control mode for accepting displacement commands from an FEA model. The effectiveness of the cascade control method in conducting hybrid simulation of stiff test specimens was evaluated with three sets of tests. For each set of tests, the results of both cascade control and displacement control methods were compared. The three test cases covered a wide range of variation from specimen size, test equipment, model type (2-D vs. 3-D), experimental element type (beam-column vs. truss), and test speed (slowdown 10 times in Test Case 1 and 2 versus 100 times in Test Case 3). In all cases, cascade control proved to be an effective method for conducting hybrid simulation with a stiff specimen.

Artificial muscles actuated by negative pressure offer significant benefits over those driven by positive pressure, such as high contraction ratios and improved safety, making them a promising option for various applications. This paper studies the contraction force characteristic of a bellows-like artificial muscle actuated by negative pressure. Initially, the structure, fabrication, and working principle of the artificial muscle were introduced. Subsequently, based on the force balance method, the contraction force was decomposed as the forces acted by the difference value of the inner and the outer pressures on the end plate, and the tension force derived from the adjacent contraction unit. To reduce complexity, the contraction process was divided into three phases according to the distinct contact conditions of the contraction units: uncontacted, locally contacted, and fully contacted with crests. The deformations of the contraction units in each phase were analyzed, and the corresponding contraction forces were derived. An experiment platform was constructed to test the force by changing the dimension parameters and pressure, obtaining the output force data during isobaric contraction. Finally, a comparison of the experimental and calculated results substantiated the aptness of the theorem model.

The efficient characterization of material properties of porous multi-phase sintered steels by instrumental indentation is still an open question. To the authors’ knowledge, so far only a characterization of single-phase porous sintered steel by nanoindenation has been reported in literature. This paper for the first time offers a study about the applicability of microindentation techniques for characterizing the matrix material in a multi-phase sintered steel. This preliminary study is motivated by the relatively wide availability of necessary equipment, and simplicity of material identification procedures.

Herein, a dual-phase ferrite/bainite Astaloy steel with 9% porosity is studied. Various commonly used methods for the reconstruction of stress–strain curves from microindentation data are considered, whereby both Vickers and spherical tips are used. In addition, some homogeneous solid materials are investigated to better asses the performance of applied identification procedures. Two approaches for the mesoscale identification of the considered sintered steel are attempted. The first one is based on the identification of individual material phases, while in the other one the homogenization of the metallic matrix is adopted. To assess the reliability of obtained parameters, the direct numerical simulation of representative volume elements of realistic steel microstructure subjected to uniaxial tension is conducted. Numerical results are compared with the data from the macroscopic uniaxial tensile test.

The obtained results indicate that microindentation is adequate for the identification of elastic properties of individual material phases, but results for local plastic parameters are largely inconclusive and a further analysis is needed, focusing on applying smaller forces and investigating the influence of pores on identification results. Nevertheless, it seems that macroscopic stress–strain curves could be captured more accurately by the methodology based on the matrix homogenization if relatively large indentation forces are applied.

An essential consideration for metal-polymer applications is that the sound joining of these materials is challenging due to a significant surface energy differential in different structural characteristics between polymer and metal. However, the joining methods have some drawbacks, such as low-reliability joints, long curing time, stress concentration, and polymer degradation. A new novel metal-polymer hybrid joining technique is proposed in this work to overcome these issues and cost perspectives, manufacturing, and overcoming the problem of PVC degradation due to heat generation of other joining methods. In this study, we managed to join PVC to AA5053 sheets using a cold joining technique based on extruding PVC through a conical hole of an aluminum specimen using a punching tool. Experiments consisted of three parameters (the hole diameter, plunging depth, and radius of the punch), with four levels for each parameter. The experiments were designed, and mechanical characterizations of the joints were optimized using the design of the experiment's method. The hole diameter was the effective parameter on the mechanical characterizations and dimensions of the extruded PVC. Increasing the diameter of the AA5053 sheet increased the maximum diameter of the extruded PVC, shear force, and pull-out force of the joints and decreased the shear stress of the joints. We obtained a maximum shear strength of 106.15 MPa, which is ~3 times higher than the tensile strength of PVC (37 MPa).