800 publications from this institution
The transient stability assessment (TSA) problem can be mapped into a two-class classification problem in machine learning, which estimates the dynamic security boundary of the power system by learning from large amount samples. A representational learning approach is proposed to solve the problem based on big data collected from Phasor Measurement Units (PMUs), which includes four stages: (i) Construct original input features by using PMUs data to describe the dynamic characteristics of the power system. (ii) Unsupervised representational feature learning by using the original features. Stacked autoencoders (SAEs) perform representational learning for crucial features. (iii) Supervised classifier training. A powerful deep learning model, convolutional neural network, which is added to SAE, is trained and tested with the learned representation. (iv) Online application, the trained model is applied to the online evaluation for TSA. Simulation on the New England 39-bus test system shows that the proposed approach has high accuracy, rare misclassification of the unstable sample and excellent robustness with noise in PMUs for TSA.
This article proposes a deep sparse autoencoder framework for structural damage identification. This framework can be employed to obtain the optimal solutions for some pattern recognition problems with highly nonlinear nature, such as learning a mapping between the vibration characteristics and structural damage. Three main components are defined in the proposed framework, namely, the pre-processing component with a data whitening process, the sparse dimensionality reduction component where the dimensionality of the original input vector is reduced while preserving the required necessary information, and the relationship learning component where the mapping between the compressed dimensional feature and the stiffness reduction parameters of the structure is built. The proposed framework utilizes the sparse autoencoders based deep neural network structure to enhance the capability and performance of the dimensionality reduction and relationship learning components with a pre-training scheme. In the final stage of training, both components are jointly optimized to fine-tune the network towards achieving a better accuracy in structural damage identification. Since structural damages usually occur only at a small number of elements that exhibit stiffness reduction out of the large total number of elements in the entire structure, sparse regularization is adopted in this framework. Numerical studies on a steel frame structure are conducted to investigate the accuracy and robustness of the proposed framework in structural damage identification, taking into consideration the effects of noise in the measurement data and uncertainties in the finite element modelling. Experimental studies on a prestressed concrete bridge in the laboratory are conducted to further validate the performance of using the proposed framework for structural damage identification.
A pounding tuned mass damper (PTMD) can be considered as a passive device, which combines the merits of a traditional tuned mass damper (TMD) and a collision damper. A recent analytical study by the authors demonstrated that the PTMD base on the energy dissipation during impact is able to achieve better control effectiveness over the traditional TMD. In this paper, a PTMD prototype is manufactured and applied for seismic response reduction to examine its efficacy. A series of shaking table tests is conducted in a three-story building frame model under single-dimensional and two-dimensional broadband earthquake excitations with different excitation intensities. The ability of the PTMD to reduce the structural responses is experimentally investigated. The results show that the traditional TMD is sensitive to input excitations, while the PTMD mostly has improved control performance over the TMD to remarkably reduce both the peak and root-mean-square (RMS) structural responses under single-dimensional earthquake excitation. Unlike the TMD, the PTMD is found to have the merit of maintaining a stable performance when subjected to different earthquake loadings. In addition, it is also indicated that the performance of the PTMD can be enhanced by adjusting the initial gap value, and the control effectiveness improves with the increasing excitation intensity. Under two-dimensional earthquake inputs, the PTMD controls remain outperform the TMD controls; however, the oscillation of the added mass is observed during the test, which may induce torsional vibration modes of the structure, and hence, result in poor control performance especially after a strong earthquake period.
Structural Health Monitoring (SHM) has been attracting numerous research efforts around the world because it targets at monitoring structural conditions and performance to prevent catastrophic failure, and to provide quantitative data for engineers and infrastructure owners to design a reliable and economical asset management strategy. In the past decade, with supports from Australian Research Council (ARC), Cooperative Research Center for Infrastructure and Engineering Asset Management (CIEAM), CSIRO and industry partners, intensive research works have been conducted in the School of Civil, Environmental and Mining Engineering, University of Western Australia and Centre for Infrastructural Monitoring and Protection, Curtin University on various techniques of SHM. The researches include the development of hardware, software and various algorithms, such as various signal processing techniques for operational modal analysis, modal analysis toolbox, non-model based methods for assessing the shear connection in composite bridges and identifying the free spanning and supports conditions of pipelines, vibration based structural damage identification and model updating approaches considering uncertainty and noise effects, structural identification under moving loads, guided wave propagation technique for detecting debonding damage, and relative displacement sensors for SHM in composite and steel truss bridges. This paper aims at summarizing and reviewing the recent research advances on SHM of civil infrastructure in Western Australia.
Doppler beam sharpening (DBS) is an effective working mode for quickly scanning and imaging to a large-scale scene. In this mode, the antenna beam direction is changing so that images from different slant angles can be spliced into a wide scale of image. In practical applications, the aircraft attitude is always affected by the unstable atmosphere, so the beam pointing has deviation compared with the ideal value, which introduces errors to image stitching and geometric correction. For this problem, a robust image stitching and geometric correction method is proposed in this paper, it has better performance when the platform is unstable, which is uncomplicated and suitable for real-time processing. The effectiveness of the proposed method is verified by real measured data experiments.
Precast segmental columns, as one structures, have attracted increasing attention in bridge and building construction. To understand the performance and dynamic characteristics of precast segmental columns subjected to earthquake excitations, this paper develops a novel approach for time-varying system stiffness identification of precast segmental columns under seismic excitations, based on the adaptive unscented Kalman filtering (AUKF) method. In the proposed approach, a precast segmental column model is considered as an equivalent single-degree-of-freedom (SDOF) system with time-varying system parameters. The AUKF method is employed to identify the time-varying stiffness parameters by using measured acceleration responses and then calculate dynamic displacement responses of the SDOF system to verify the accuracy of the identified stiffness parameters. Since the system parameter identification process and measurement noise covariance matrices are updated at each time step, the estimated results can be reliably obtained to evaluate the conditions of segmental columns subjected to seismic excitations. To verify the feasibility and effectiveness of using the proposed approach for time-varying system stiffness identification of segmental columns, numerical studies on a planar precast segmental column subjected to different intensities of ground motion excitations are conducted. Shake table tests on a scaled precast segmental column and a scaled monolithic column subjected to earthquake excitations are conducted in the laboratory to further verify the performance of the proposed approach. Measured acceleration responses are used for time-varying system identification. Both numerical and experimental results demonstrate that the proposed approach can successfully identify the time-varying stiffness values of precast segmental columns subjected to earthquake excitations, and the displacement responses of these columns are predicted accurately as compared with the measured ones.
This paper presents a structural damage identification approach based on the time domain impulse response functions, which are extracted from the measured dynamic responses with the input available. The theoretical sensitivity of the impulse response function with respect to the system stiffness parameters considering the damping model is derived. The first-order sensitivity based model updating technique is performed for the iterative model updating. The initial structural finite element model and acceleration measurements from the damaged structure are required. Local damage is identified as a reduction in the elemental stiffness factors. The impulse response function sensitivity based optimal sensor placement strategy is employed to investigate the best sensor locations for identification. Numerical studies on a beam model are conducted to validate the proposed approach for the extraction of time domain impulse response functions and subsequent damage identification. The simulated damage can be identified effectively and accurately.
Perforation plays an important role in the fracture morphology near the wellbore and the propagation of hydraulic fracturing fractures. Therefore, it is of great significance to find out the fracture morphology and propagation law during perforation for optimizing perforation technology, enhancing fracture control, and realizing complementary advantages of different perforation schemes. Based on analyzing the characteristics of perforation fracturing at each stage and existing perforation technology, two types of deep-penetrating perforating bullets were used to carry out large-scale perforation shooting experiments. The real processes of spiral perforation, directional perforation, conventional fixed-plane perforation, and interlaced fixed-plane perforation were simulated, respectively. The near-wellbore fracture morphology, formation mechanism, and propagation rule during perforation with different perforation modes were analyzed. The results show that (1) perforation is accompanied by the formation of tunnels, and there are three kinds of source microfractures developed around the tunnels, namely Type I radial microfractures, Type II oblique microfractures, and Type III perforation tip divergent microfractures. The three microfractures are interconnected to form more complex near-wellbore fractures. (2) Under different perforation modes and parameters, the near-wellbore fracture morphology and propagation law formed by microfractures around tunnels are also different. (3) The existence and expansion of near-wellbore fractures validate Chen et al.’s (2005) conjecture that there are “pre-existing fractures” in perforation and negate the assumption that the perforation tunnels are complete. There are no near-wellbore fractures when the perforation method is optimized. The research results in this paper can provide guidance and reference for improving the perforation fracturing effect in oil and gas reservoirs.
In continuous annealing lines/hot galvanizing,strips are guided by upper and lower rolls in an alternate up and down vertical motion.This kind of process consisting of heating,soaking,and cooling furnaces(or zones) has several advantages in terms of productivity and product quality.Main problems encountered in continuous annealing lines are misguiding and formation of buckling.The strip buckling causes the strip to be rejected and,in the worst case,the strip breaks and the continuous annealing lines are compelled to stop.In this paper the elastic-plastic model of the strip traveling and the corresponding boundary condition were constructed dependant on the theory of mechanic.Simulations are performed in two steps.First,a homogeneous longitudinal tensile stress is applied at one end of the strip,the other end of the strip and the roll are restricted.In the second step,the applied tension remains constant.A uniform displacement is applied to the other end of the strip and a rotation to the roll. The transverse compressive stress,the buckling formation during strip traveling process,the location for the buckling taking place and the buckling criterion were investigated systematically.The result shows that the transverse compressive stress increases with the increase of the tensile stress and the crown of the tapered roll. Because the contact area between the tapered roll and the strip increase with the decrease of the crown,the tensile stress distribution is more homogeneous in the transverse direction of the strip.An additional transverse compressive stress is created and a buckling is also observed due to the friction during the strip motion in the second step.Once the buckling is appearance,it is very hard to be removed only through decreasing the applied longitudinal tension,because the strip wrinkle still keeps high level due to the effect of the buckled strip,the wrinkle would evolve into buckling eventually.The buckling formation process predicted by the finite element model is very similar to those observed in industrial processing lines.It takes place on the cylindrical part of the tapped roll and the buckling is moving to the middle part of the tapped roll with the increase of the crown,its width is about 2-4 mm.The criterion for the buckle occurrence is also discussed,in the prescribed situation,the transverse shell curvature increases with the strip displacement,if the transverse shell curvature reaches 0.1 mm~(-1), the buckling would take place.
Abstract Three dimensional (3D) FE models have been developed for simulation of the incremental forming processes employed in UOE linepipe manufacturing at Tata Steel Hartlepool 42” Pipe Mill, namely edge crimping (C-press) and pipe expansion (expander). Transitional zone behaviour resulted from both forming operations as previously identified in practice has been revealed, for the first time, by the 3D FE simulations and preliminarily analysed. It was demonstrated that the transitional zone features observed in both edge crimping and pipe expansion were initiated in the plate/pipe feed-in side/area of the dies, and then formed within the working length/face of the forming dies. Detailed examination and analysis of the FE results, in terms of plastic strain and residual stress patterns as well as contact pressure distributions, have shown that the initiation of the transitional zone features was due to the redundant plastic deformation caused by (cantilever) bending in the material feed-in area of the dies mainly along the longitudinal direction. It is therefore believed that minimisation/elimination of such redundant plastic bending effect in the longitudinal direction would lead to minimisation/elimination of the unfavourable deformation features in the so-called transitional zones, which could be achieved through improved die designs and possibly forming process parameter settings.
A goal programming model has been developed to analyze the system behavior for the water distribution networks under contingency situations due to failures of pipes and pumps, taking into account three aspects: (1) equity, or sharing inconvenience equally among consumers; (2) redistribution of the network flows to reduce the negative consequences of a failure of one portion on other portions of the network; and (3) consideration of pressure‐dependent demand delivery due to insufficient head, namely, if a nodal head falls below a desired level, the flow delivered to that node is reduced. The first priority of the goal program is to maximize the lowest nodal demand supply ratio (or the ratio of actually delivered demand to the required demand at a node). The second priority is to maximize the system demand supply ratio (or the ratio of actually delivered water to the required total system demand). Link flow directions in the model are not fixed but are determined by a set of criteria. The system behaviors with respect to the three aspects of reliability factors are examined through extensive numerical experiments. The impact of equity requirements on redistribution of network flows, link flow directions, nodal demand supply ratio, and system demand supply ratio when failure events become serious is examined in particular detail. It is found that equity requirements can satisfactorily bring about fair sharing of inconvenience among consumers. The model proposed also suggests that network operations should reverse some link flow directions in order to meet equity requirements under severe contingencies.
The original signal of rolling bearing fault contains a large number of phase coupling components and is easily submerged in the background noise, which make the fault information difficult to be extracted accurately. Aiming at the above problems, a method of fault feature extraction for rolling bearing is proposed, which combines empirical mode decomposition (EMD) with a 1.5-dimensional spectrum. The original signal is decomposed by EMD to obtain the intrinsic modal function (IMF) of different scales. The IMF is selected by the size of a correlation coefficient and a kurtosis value to eliminate the high-frequency components, which is reconstructed to achieve the purpose of noise reduction. The reconstructed Hilbert envelope signal is analysed by the 1.5-dimensional spectrum to extract the nonlinear characteristic of two phase couplings, so that the fault characteristic frequency of the bearing is obtained. By analysing the signal of actual rolling bearings, the fault characteristic frequency of bearing inner and outer rings can be effectively extracted, and the validity and feasibility of the method are proved.
With the increasing of spacecraft launch missions, the number of spacecraft module maintenance tasks also increases, so the path planning link plays a crucial role in the online virtual module maintenance task operation of spacecraft. Since A large amount of qualitative/quantitative method analysis is required in the process of writing the traditional maintenance report, the A* algorithm is adopted in this paper to search the path for the on-orbit module maintenance task of the spacecraft. Firstly, the abstract model of spacecraft in orbit cabin is built with unity 3D, and then the search point is controlled to move to the target point based on C#. Finally, in the process of gradually adjusting C# parameters, the optimal parameters are obtained and the path planned by A* algorithm is displayed. Simulation results show that this method can plan an efficient and safe maintenance path in virtual maintenance environment.
Combining actual face circumstance,elaborates key technique processing for high gas face with compound roof so as to guarantee safe production of coal mine,which solves the dificult preblem of gas govenence at blasting face with compound roof and high gas.
To investigate the dynamic performance of the multi-span girder bridge under moving vehicles, based on the synthesis modal method, the program for vehicle-bridge coupled vibration analysis is completed firstly. According to the numerical simulation, the estimation formulas in current codes (China and UK) and some other references (USA) have been proved problematic. Also, the roughness and the critical speed are fully discussed. In addition, the DLA in the side span is largely different from that in the other span, and the DLA in other spans are almost the same. The DLA of middle span is little influenced by the number of spans. However, the DLA of side span will be stable only when the number of spans is not less than five. And when the number of spans is not more than five, the DLA of the three-span bridge is biggest, that of the five-span bridge is second, and that of the four-span bridge is smallest. It has been proved that the DLA in current code is not rational enough, as it does not distinguish the positions. It may provide some basis for the design and evaluation of the dynamic performance of bridges to moving vehicles.
