A straight-arc combined orbital welding robot system was designed for girth welding of box-type steel structures at construction sites. The kinematics analyses used the standard D-H model and the Craig modified D-H model for this welding robot with evluations based on the modeling accuracy and solution process for the welding torch trajectory planning for girth welding of box-type steel structures. The trajectory planning method could adjust for arbitrary spatial position and posture relative to the welding torch with the minimum degrees of freedom. The planning mehtod gave an efficient trajectory for right-angle welding of the box-type steel structure. Simulations show that the welding robot system can provide the girth welding of box-type steel structures and can ajust to various spatial positions and postures.

Aluminum alloys are important light metals for weight reduction in cars and aircrafts. However, aluminum alloy applications are limited when the aluminum alloy strength is not sufficient. The mechanical properties of Al5052 can be improved by using carbon fiber reinforced aluminum composites fabricated using multiple-pass friction stir processing (FSP). The mechanical properties and microstructures of the composites are investigated with the composite having 18.9% higher tensile strength and 19.7% higher elongation rate than the matrix. The microstructure observation indicates that the carbon fibers are uniformly dispersed in the matrix. Fractography results show pulled-out carbon fibers in the inner surfaces of dimples. The strengthening of the composites is due to load transfer to the homogeneous carbon fibers.

The stage is the key subsystem in a lithography machine. The cable-pipe assembly transmits the motor coil signal, power and coolant in stage systems. However, disturbances generated by the cable-pipe assembly reduce the ultra-precision of the stage. Therefore, the disturbances need to be reduced to improve the stage positioning accuracy. The cable-pipe assembly and its disturbance characteristics are analyzed to develop a simplified finite element model of the cable-pipe system. Then, the main structural parameters of the assembly are optimized with design guidelines for future cable-pipe assembly designs. Tests show that the finite element model is accurate with a first order frequency error of less than 10%. Simulations show that the disturbance amplitude of the L-direction is reduced by 80% relative to that without optimization.

Dynamic modeling was used to evaluate a 3-PRRU parallel manipulator in a high-speed hybrid machine tool. The virtual work principle was used to develop the dynamic model of the 3-PRRU parallel mechanism with two new indices defined to evaluate the isotropy of dynamics from a kinetic energy viewpoint. The indices were then used in an atlas method for a five dimensions image. The results show that these indices accurately describe the isotropic performance of the 3-PRRU parallel manipulator. The two indices have uniform dimensions, clear physical meaning, and can accurately describe the dynamics of parallel manipulators.

Directly preparation of a ⁶³Ni radioisotope source on the surface of a betavoltaic battery has problems such as low loadings, low activation, and functional deterioration of the PN junction. This research demonstrates the use of an indium tin oxide (ITO) membrane as a conductive layer to plate a ⁶³Ni source on the surface of a transparent glass package. Simulations of the radiation self-absorption effect give the optimal thickness of the ⁶³Ni-Si betavoltaic battery. Then, a 180 nm thick ITO conductive membrane is prepared on a 400 μm thick glass substrate via magnetron sputtering. ⁶³Ni is then electroplated on the ITO membrane surface. The micro-morphology and the Ni-electroplated membrane thickness are characterized. The electrical capabilities of the ⁶³Ni-Si betavoltaic battery fabricated based on this radiation source preparation method are then simulated. The results show that irradiation of a 2 μm thick, 7.25×10⁸ Bq activity, 36 mm² radiation source gives a radioisotope battery that can export 85.4 nW output power.

The cortical bone thickness in human ribs is location dependent and varies substantially among the population, which has significant effects on rib fracture characteristics in accidents. The objective of the current study is to quantify the cortical bone thickness distributions in Chinese subjects as functions of the age, sex, stature and body mass index (BMI). The cortical bone thicknesses are measured using clinical computed tomography (CT) data from 60 Chinese subjects for the 1^st to 12^th levels of the left ribs. Statistical analyses identify differences based on the locations and body characteristics. Sex, stature, and BMI significantly affect the cortical bone thickness distribution in the human ribs. This cortical bone thickness distribution model can serve as a geometric basis for developing a parametric human thorax finite element model for quantifying the effects of different human attributes on rib fractures.

The basic reproductive number is the most important parameter for measuring epidemic potential and assessing the disease spreading risk. To assess the 2013 Ebola virus spreading risk, a modified least squares method is used to fit the data to give a more self-consistent theoretical basis. Very early Ebola epidemic data sets from the three hardest hit countries in West Africa (Guinea, Sierra Leone and Liberia) are fit to calculate the basic reproductive number with the results in good agreement with the actual spread. A risk evaluation model which uses an extended susceptible, exposed, infectious, and removed (SEIR) model based on the multi-epidemic hypothesis gives better predictions than the uniformly mixed assumption based SEIR model used in previous studies. The model and fitting method can help predict virus propagation, assess the effects of control measures, and predict the future trends of the epidemic spread to enable better containment of future epidemics.

A principal component analysis (PCA) and Mahalanobis distance (MD) based anomaly gamma spectra detection method is developed to improve the reliability of radiation scanning of goods and human bodies. This method first extracts all the principal components (PCs) of large numbers of benign gamma spectra by PCA and selects several largest PCs to form a subspace. The algorithm then projects the benign, unknown gamma spectra to this subspace, calculates their MDs, and completes the anomaly detection by comparing these MDs. Monte Carlo simulations and actual tests show that the method is reliable and effective when the subspace has more than 99% of the original information.

The helium compressor is one of the key components in the high temperature gas-cooled reactor coupled with a helium turbine. The aerodynamic performance of the helium compressor significantly influences the whole system efficiency. Therefore, the effects of the tip and axial clearances on the helium compressor need to be further studied and optimized. A computational fluid dynamics model that has been verified against experimental data of the helium compressor is used to investigate the influence of the tip and axial clearances. The results show that a reduced tip clearance reduces the flow losses and increases the compression ratio and compressor efficiency, especially for clearances of less than 0.3 mm (2% of the relative blade height). A larger axial clearance lowers the compressor efficiency with an acceptable axial displacement of ±1.0 mm. The results show the influences of the tip and axial clearances on the model, and provide guidance for the design and optimization of the actual helium compressor.

The procurement of equipment and materials plays a critical role in international engineering, procurement, and construction (EPC) project delivery. However, there is a lack of empirical research to clarify the causal mechanism between supply chain integration and procurement. A conceptual model of integrated procurement is developed based on supply chain integration. Questionnaire data is used to explore the effect of supply chain integration on procurement using a partial least squares-structural equation model (PLS-SEM). The results show that relational integration directly affects procurement performance, relational integration affects the procurement processes which helps improve the procurement performance, and relational integration facilitates information integration which further improves the procurement processes and procurement performance. The relational integration indirect effect on the procurement processes via information integration is larger than its direct effect on procurement processes and the relational integration direct effect on procurement performance is larger than its indirect effect on the procurement performance via information integration and procurement processes. These results provide guidance for EPC contractors choosing procurement strategies.

Concrete pouring layer surfaces are seriously affected by the environment when hydraulic concrete pours in dry, windy environments which affects the interlayer strength. The effect of the concrete interlayer surface is analyzed for various environmental conditions by measuring the interlayer splitting tensile strength and the water content of the first concrete layer. The results show that the interlayer strength is reduced by the environmental conditions with the interlayer strength related to the water content in the first layer. Therefore, the water content can be used to predict interlayer strength. A layer remodeling method is proposed to improve the interlayer strength based on the extreme situation of a water shortage on concrete surface of the first layer surface.

This study distinguishes the direct and indirect effects of rainfall changes on the water balance in a watershed. The water balance is used to build a model based on Eagleson's ecohydrological optimality theory to describe the vegetation cover response to rainfall changes. An attribution analysis distinguishes the direct effect of rainfall on the water balance shifts and the indirect effect of vegetation cover which is influenced by the rainfall. This method is applied to the Xilamulun-Laoha River basin in the Horqin Sands area. The results show a positive correlation between the vegetation cover and the rainfall with the relationship affected by the rainfall intensity and frequency. Increased rainfall mainly increases the evapotranspiration and soil water storage during the non-growing season. Increased rainfall has a positive effect on every water balance term. However, the indirect effects of the vegetation cover result in negative changes to all the water balance terms, except for the evapotranspiration. The relationship between the evapotranspiration efficiency and the rainfall is affected by the response of the vegetation cover to rainfall changes.

The inter-frame independence assumption for speech recognition simplifies the computations. However, it also reduces the model accuracy and can easily give rise to recognition errors. Therefore, the objective of this paper is to search for a feature which can weaken the inter-frame dependence of the speech features and keep as much information of the original speech as possible. Two speech recognition feature extraction algorithms are given based on the k-means algorithm and the normalized intra-class variance. These algorithms provide adaptive clustering feature extraction. Speech recognition tests with these algorithms on a Gaussian mixture model-hidden Markov model (GMM-HMM), a duration distribution based HMM (DDBHMM), and a context dependent deep neural network HMM (CD-DNN-HMM) show that the adaptive feature based on the normalized intra-class variance reduces the relative recognition error rates by 10.53%, 5.17%, and 2.65% relative to the original features. Thus, this adaptive clustering feature extraction algorithm provides improved speech recognition.

When the propagation environment does not change, the only factor that can lead to uncertainties in the key generation rate is the changing position of the user. This study analyzes the relationship between the user position variation and the wireless key generation rate with the key generation rate limit analyzed for scenarios in two static environments. The results show that in an astatic channel environment where the user positions are changing, the channel distribution is not Gaussian and the secret key generation rate is only affected by changes in the user position. The upper bound on the secret key generation rate is estimated numerically based on the channel error for the simple reflection scene. For the scattering scene, the results show that the key generation rate is high enough when the user position changes about one wavelength.

The small-cell technology is one of the key technologies in cellular communications that should provide high system capacity and high energy efficiency. Although the energy consumption of one single base station diminishes in small cell networks (SCNs), the total energy consumption still increases along with the growth in traffic, as the number of base stations increases quickly. Queue state information (QSI) could be utilized for user scheduling, carrier allocation, and power allocation to enhance system performance. In this study, QSI is used to minimize the time average of the total power expenditure in multi-carrier SCNs, where no inter-cell interference exists, while satisfying the traffic demand. The problem is formulated according to Lyapunov optimization theory into a mixed integer programming problem. An optimal algorithm is given for user scheduling, carrier allocation, and power allocation in each slot. Simulations verify that the algorithm reduces the energy consumption with a similar sum rate as the algorithm aiming to maximize the system sum rate.

Fifth generation (5G) mobile communications using asynchronous multiple access machine-to-machine communications will use non-orthogonal multicarrier signals. A filter bank-based multicarrier signal with offset quadrature amplitude modulation (FBMC/OQAM) is one candidate waveform. This study analyzed the effect of narrow band interference (NBI) and residual intrinsic interference (RⅡ) on FBMC systems. Interference mitigation models were then designed for the NBI and RⅡ for a receiver based on a combination of FBMC and bit-interleaved coded modulation with iterative decoding (BICM-ID) to further improve the performance. Simulations show that the FBMC is more robust to NBI than cyclic prefix orthogonal frequency division multiplexing (CP-OFDM). After several iterations, the receiver approaches the Genie performance bound which assumes perfectly known interference. The iterative performance of different constellation types are also compared.

The neural machine translation based on deep learning significantly surpasses the traditional statistical machine translation in many languages, and becomes the current mainstream machine translation technology. This paper compares six influential neural machine translation methods from the level of word granularity in the task of Uyghur-Chinese machine translation. These methods are attention mechanism (GroundHog), vocabulary expansion (LV-groundhog), source language and target language with subword units (subword-nmt), characters and words mixed (nmt.hybrid), subword units and characters (dl4mt-cdec), and complete characters (dl4mt-c2c). The experimental results show that Uyghur-Chinese neural machine translation performs best when the source language is segmented into subword units and the target language is represented by characters (dl4mt-cdec). This paper is the first to use neural machine translation for Uyghur-Chinese machine translation and the first to compare different neural machine translation methods on the same corpus. This work is an important reference not only for Uyghur-Chinese machine translation, but also for general neural machine translation tasks.

Uncertainties make it impossible to cluster uncertain data streams using traditional clustering algorithms. This paper presents a density-based clustering algorithms for uncertain data stream environments. An uncertainty metric is used to measure the distribution information in the uncertain data. The uncertain data streams DENCLUE algorithm (USDENCLUE) is then modified to deal with uncertainty data to minimize the impact of the data uncertainty on the clustering results. A density-based clustering algorithm is then given for uncertain data streams with a sliding window to rapidly prune the clusters using an exponential histogram of the cluster features. This algorithm can efficiently handle noisy data in evolving data streams to generate arbitrary clusters to improve the clustering quality. Comparisons of this algorithm with the CluStream clustering algorithm on real and synthetic data sets show the efficiency and effectiveness of this algorithm.

Existing information system risk assessment methods based on fuzzy language variables generally requires that the language variable itemsets generated by the standard parameter trapezoidal fuzzy numbers which are generated by experts in the field. The effectiveness of this approach then often depends on its ability to deal with different types of fuzzy numbers and changing linguistic variable itemsets when the expert cannot participate. This study presents a method for an n-fold reduction of the linguistic variables based on the triangular fuzzy numbers that is validated using four types of fuzzy numbers for language variables. The results show that the method automates the equivalent transformation of n-fold language variable items without expert intervention to improve information security risk assessment systems.