Search

Vibration analysis of composite beams is carried out by using a finite element-based formal asymptotic expansion method. The formulation begins with three-dimensional (3D) equilibrium equations in which cross-sectional coordinates are scaled by the characteristic length of the beam. Microscopic two-dimensional and macroscopic one-dimensional (1D) equations obtained via the asymptotic expansion method are discretized by applying a conventional finite element method. Boundary conditions associated with macroscopic ID equations are considered to investigate the end effect. It is then described how one could form and solve the eigenvalue problems derived from the asymptotic method beyond the classical approximation. The results obtained are compared with those of 3D finite element method and those available in the literature for composite beams with solid cross section and thin-walled cross section. [DOI: 10.1115/1.4000972]

In this paper, a new damage detection technique able to identify the location of a breathing crack in an isotropic rod, relying only on real-time measurements, is proposed. The detection algorithm exploits the phase information associated with the superharmonic components produced, in the Fourier spectrum, by the nonlinear dynamic response of this kind of defect under the influence of an external dynamic excitation. The validity of the proposed algorithm for a weakly nonlinear system is supported by an analytical solution for a cracked beam obtained through the harmonic balance approach. A numerical investigation is conducted by means of a finite element model of an isotropic beam integrating nonlinear contact elements in the damaged area and solved for the steady-state response. Three different postprocessing approaches, incorporating the proposed damage detection algorithm, are formulated and compared to assess the capability of the current methodology. Results from the cracked beam model clearly show the generation of the superharmonics as a result of the nonlinear dynamic behavior of the breathing crack. The phase associated with the superharmonic components is then processed through the detection algorithm and the predicted location is compared with the actual position of the defect to assess the performances of the methodology. DOI: 10.2514/1.38947

For bladed-disk assemblies in turbomachinery, the elements are often exposed to aerodynamic loadings, the so-called engine order excitations. It has been reported that such excitations could cause significant structural vibration. The vibration level could become even more excessive when the bladed disk is mistuned, and may cause fatigue damage to the engine components. To effectively suppress vibration in bladed disks, a piezoelectric transducer networking concept has been explored previously by the authors. While promising, the idea was developed based on a simplified bladed-disk model without considering the disk dynamics. To advance the state of the art, this research further extends the investigation with focus on new circuitry designs for a more sophisticated and realistic system model with the consideration of coupled-blade-disk dynamics. A novel multicircuit piezoelectric transducer network is synthesized and analyzed for multiple-harmonic vibration suppression of bladed disks. An optimal network is derived analytically. The performance of the network for bladed disks with random mistuning is examined through Monte Carlo simulation. The effects of variations (mistuning and detuning) in circuit parameters are also studied. A method to improve the system performance and robustness utilizing negative capacitance is discussed. Finally, experiments are carried out to demonstrate the vibration suppression capability of the proposed piezoelectric circuitry network. [DOI: 10.1115/1.2948410] Keywords: piezoelectric network, bladed disk, vibration suppression

Byline: M. Cao (General Motors R&D, REB, 30500 Mound Road, Warren, MI 48071, USA, ming.cao@gm.com); K.W. Wang (Department of Mechanical and Nuclear Engineering, Pennsylvania State University, 157E Hammond Building, University [...]

The concept of using sensitivity-enhancing feedback control to improve the performance of frequency-shift-based structural damage identification has been recently explored. In previous studies, however, the feedback controller is designed to alter only the closed-loop eigenvalues, and the effect of closed-loop eigenvectors on the sensitivity enhancement performance has not been considered. In this research, it is shown that the sensitivity of the natural frequency shift to the damage in a multi-degree-of-freedom structure can be significantly influenced by the placement of both the eigenvalues and the eigenvectors. A constrained optimization problem is formulated to find the optimal assignment of both the closed-loop eigenvalues and eigenvectors, and then an optimal sensitivity-enhancing control is designed to achieve the desired closed-loop eigenstructure. Another advantage of this scheme is that the dataset of frequency measurement for damage identification can be enlarged by utilizing a series of closed-loop controls, which can be realized by activating different combinations of actuators in the system. Therefore, by using this proposed idea of multiple sensitivity-enhancing feedback controls, we can simultaneously address the two major limitations of frequency-shift-based damage identification: the low sensitivity of frequency shift to damage effects and the deficiency of frequency measurement data. A series of case studies are performed. It is demonstrated that the sensitivity of natural frequency shift to stiffness reduction can be significantly enhanced by using the designed sensitivity-enhancing feedback control where the optimal placement of closed-loop eigenvectors plays a very important role. It is further verified that such sensitivity enhancement can directly benefit the damage identification accuracy and robustness. [DOI: 10.1115/1.2748476] Keywords: structural damage identification, vibration-based damage identification, sensitivity-enhancing control eigenstructure assignment

Extensive investigations have been conducted to study the vibration localization phenomenon and the excessive forced response that can be caused by mistuning in bladed disks. Most previous researches have focused on analyzing/predicting localization or attacking the mistuning issue via mechanical tailoring. Few have focused on developing effective vibration control methods for such systems. This study extends the piezoelectric network concept, which has been utilized for mode delocalization in periodic structures, to the control of mistuned bladed disks under engine order excitation. A piezoelectric network is synthesized and optimized to effectively suppress vibration in bladed disks. One of the merits of such an approach is that the optimum design is independent of the number of spatial harmonics, or engine orders. Local circuits are first formulated by connecting inductors and resistors with piezoelectric patches on the individual blades. Although these local circuits can function as conventional damped absorber when properly tuned, they do not perform well for bladed disks under all engine order excitations. To address this issue, capacitors are introduced to couple the individual local circuitries. Through such networking, an absorber system that is independent of the engine order can be achieved. Monte Carlo simulation is performed to investigate the effectiveness of the network for a bladed disk with a range of mistuning level of its mechanical properties. The robustness issue of the network in terms of detuning of the electric circuit parameters is also studied. Finally, negative capacitance is introduced and its effect on the performance and robustness of the network is investigated. [DOI: 10.1115/1.2775511] Keywords: piezoelectric network, vibration suppression, bladed disk, negative capacitance

Abstract: The objective of this research is to investigate the feasibility of utilizing eigenvector assignment and piezoelectric circuitry for enhancing vibration isolation performance of periodic isolators. For a classical periodic [...]

In automatic transmission design, electronic control techniques have been adopted through proportional variable-force-solenoid valves, which typically consist of spool-type valves (Christenson, W. A., 2000, SAE Technical Paper Series, 2000-01-0116). This paper presents an experimental investigation and neural network modeling of the fluid force and flow rate for a spool-type hydraulic valve with symmetrically distributed circular ports. Through extensive data analysis, general trends of fluid force and flow rate are derived as functions of pressure drop and valve opening. To further reveal the insights of the spool valve fluid field, equivalent jet angle and discharge coefficient are calculated from the measurements, based on the lumped parameter models. By incorporating physical knowledge with nondimensional artificial neural networks (NDANN), gray-box NDANN-based hydraulic valve system models are also developed through the use of equivalent jet angle and discharge coefficient. The gray-box NDANN models calculate fluid force and flow rate as well as the intermediate variables with useful design implications. The network training and testing demonstrate that the gray-box NDANN fluid field estimators can accurately capture the relationship between the key geometry parameters and discharge coefficient/jet angle. The gray-box NDANN maintains the nondimensional network configuration, and thus possesses good scalability with respect to the geometry parameters and key operating conditions. All of these features make the gray-box NDANN fluid field estimator a valuable tool for hydraulic system design. [DOI: 10.1115/1.2232687] Keywords: spool valve, fluid control systems, discharge coefficient, jet angle, variable force solenoid (VFS), gray box, nondimensional artificial neural network (NDANN)

Space-based adaptive optic systems have gained considerable attention within the past couple of decades. Achieving the increasingly stringent performance requirements for these systems is greatly hindered by strict weight restrictions, size limitations, and subjected hostile environments. There has been considerable attention in developing lightweight adaptive optics where piezoelectric sheet actuators are attached' on the back of optical mirrors to achieve a high precision surface shape with minimum additional weight. Vibration control of such large flexible space structures is continually challenging to engineers due to the large number of actuators and sensors and the large number of vibration modes within the operational bandwidth. For these structures, any disturbed modes are likely to remain vibrating for an extended period of time due to the small amount of damping available. As a result, controller spillover should be minimized as much as possible to avoid exciting the residual modes. In recent investigations of circular plate shape control by [Philen and Wang, Int. Soc. Opt. Eng. 4327, pp. 709-719]. It was demonstrated that directional decoupling of the two-dimensional actuator (meaning that the actuation in one of the two directions is eliminated) improves the system performance when correcting for the lower order Zernike static deformations. This directional decoupling effect can be achieved through an active stiffener (AS) design. In this research, analytical and experimental efforts are carried out to examine the effect of the active stiffener actuators in reducing the controller spillover through the stiffeners' decoupling characteristics. It is shown that significant reductions in controller spillover can be achieved in systems using the active stiffener actuators when compared to systems having direct attached (DA) actuators, thus resulting in improved vibration control performance. The experimental results verify the analytical predictions and clearly demonstrate the merit of the active stiffener concept. [DOI: 10.1115/1.2013303]

In this study; a new hybrid-neural-network-based friction component model is developed for powertrain (PT) dynamic analysis and controller design. This new model, with significantly improved input-output scalability over conventional neural network configuration, has the capability to serve as a forward as well as an inverse system model. The structural information of the available physical and empirical correlations is utilized to construct a parallel-modulated neural network (PMNN) architecture consisting of small parallel sub-networks reflecting specific mechanisms of the friction component engagement process. The PMNN friction component model isolates the contribution of engagement pressure on engagement torque while identifying the nonlinear characteristics of the pressure-torque correlation. Furthermore, it provides a simple torque formula that is scalable with respect to engagement pressure. The network is successfully trained, tested and analyzed, first using analytical data at the component level and then using experimental data measured in a transmission system. The PMNN friction component model, together with a comprehensive powertrain model, is implemented to simulate the shifting process of an automatic transmission (AT) system under various operating conditions. Simulation results demonstrate that the PMNN model can be effectively applied as a part of powertrain system model to accurately predict transmission shift dynamics. A pressure-profiling scheme using a quadratic polynomial pressure-torque relationship of the PMNN model is developed for transmission shift controller design. The results illustrate that the proposed pressure profiling technique can be applied to a wide range of operating conditions. This study demonstrates the potential of the PMNN architecture as a new dynamic system-modeling concept: It not only outperforms the conventional network modeling techniques in accuracy and numerical efficiency, but also provides a new tool for transmission controller design to improve shift quality. [DOI: 10.1115/1.1978909] Keywords: Friction Component, Powertrain Simulation, Automatic Transmission, Dynamic Modeling, Hybrid Neural Network, Analytical Calibration

Abstract: Many researchers and engineers have employed active control techniques, such as active magnetic bearings (AMBs), to suppress imbalance vibration in various subcritical and supercritical speed rotor-dynamic applications. One issue [...]

Spaced-based adaptive optic systems have gained considerable attention within the past couple of decades. Achieving the increasingly stringent performance requirements for these systems is greatly hindered by strict weight restrictions, size limitations, and subjected hostile environments. There has been considerable attention in lightweight adaptive optics, where piezoelectric sheet actuators are directly attached on the back of optical mirrors to achieve a high-precision surface shape with minimum additional weight. In recent studies, it was discovered that the performance of the system could be further improved if the piezoelectric sheet actuator were decoupled in direction, meaning that the actuation in one of the two directions is eliminated. To realize the decoupling effect, the concept of utilizing the active stiffener for high-precision shape control is proposed and investigated in the study. The active stiffener is a stiffener-piezoceramic actuator pair that consists of an passive insert (stiffener) placed between the host structure and the active piezoelectric actuator. The basic premise is that the insert (stiffener) will reduce the action transmitted in one direction while allowing adequate action to transmit in the orthogonal direction. In this research, analytical and experimental efforts are carried out to examine the effect of the active stiffener actuators for shape control of circular plate structures. Analysis is performed on two large flexible circular plate structures, one having the direct attached actuators and the other utilizing the active stiffener actuators. It is shown that more reductions in the surface error can be achieved with the active stiffeners, as compared to systems with the direct attached actuators. It is illustrated that the direct attached actuators generate more localized deformation in the structure, whereas the decoupling capability of the active stiffener reduces the undesired localized deformation considerably. The experimental results verify the analytical predictions and clearly demonstrate the performance improvement of the active stiffener concept over the direct attached actuator.

In this research, a new hybrid neural network is developed to model engagement behaviors of automotive transmission wet friction component. Utilizing known first principles on the physics of engagement, special modules are created to estimate viscous torque and asperity contact torque as preprocessors to a two-layer neural network. Inside these modules, all the physical parameters are represented by neurons with various activation functions derived from first principles. These new features contribute to the improved performance and trainability over a conventional two-layer network model. Both the hybrid and conventional neural net models are trained and tested with experimental data collected from an SAE#2 test stand. The results show that the performance of the hybrid model is much superior to that of the conventional model. It successfully captures detailed characteristics of the friction component engagement torque as a function of time over a wide operating range. [DOI: 10.1115/1.1649980]

The underlying principle for vibration confinement is to alter the structural vibration modes so that the corresponding modal components have much smaller amplitude in concerned area than in the remaining part of the structure. In this research, the state-of-the-art in vibration confinement technique is advanced in two correlated ways. First, a new eigenstructure assignment algorithm is developed to more directly suppress vibration in regions of interest. This algorithm is featured by the optimal selection of achievable eigenvectors that minimizes the eigenvector components at concerned region by using the Rayleigh Principle. Second, the active control input is applied through an active-passive hybrid piezoelectric network. With the introduction of circuitry elements, which are much easier to implement than changing or adding mechanical components, the state matrices can be reformed and the design space for eigenstructure assignment can be greatly enlarged. To maximize the system performance, a simultaneous optimization/optimal eigenvector assignment approach to decide the passive and active parameters concurrently is outlined. The merits of the proposed system and scheme are demonstrated and analyzed using numerical examples.

This paper presents an application of probabilistic theory to the selection of robust PSS locations and parameters. The aim is to enhance the damping of multiple electromechanical modes in a multimachine system over a large and prespecified set of operating conditions. Conventional eigenvalue analysis is extended to the probabilistic environment in which the statistical nature of eigenvalues corresponding to different operating conditions is described by their expectations and variances. Probabilistic sensitivity indices to facilitate 'robust PSS' site selection and a probabilistic eigenvalue-based objective function for coordinated synthesis of PSS parameters are then proposed. The quasi-Newton technique of nonlinear programming is used to solve the objective function and its convergence properties are discussed and compared with the conventional steepest descent approach. The effectiveness of the proposed stabilizers, with a classical iead/lag structure, is demonstrated on an eight-machine system. Index Terms--Eigenvalue, optimization, power system stabilizer (PSS), probabilistic theory, sensitivity.

It is known that, when the mechanical coupling between the substructures is weak, small imperfections in a periodic structure can induce vibration localization. This phenomenon could lead to large vibration in certain regions of the structure and could be very harmful to the system. In this study, it is shown that the proposed coupled piezoelectric circuits can greatly relieve or even eliminate such localization problems. Part of the structural vibration energy will be transferred into electrical energy through the piezoelectric materials, and the newly created electro-mechanical wave/energy channel will sustain the energy propagation throughout the structure. The effectiveness and robustness of the coupled piezoelectric circuits on reducing vibration localization is demonstrated through analysis. Design guidelines are also established via approximation techniques and parametric studies. [DOI: 10.1115/1.1521951]