On improving the performance of large parallel lattice Boltzmann flow simulations in heterogeneous porous media

Classical Cartesian domain decompositions for parallel lattice Boltzmann simulations of fluid flow through heterogeneous porous media are doomed to workload imbalance as the number of processors increases, thus leading to decreasing parallel performance. A one-lattice lattice Boltzmann method (LBM) implementation with vector data structure combined with even fluid node partitioning domain decomposition and fully-optimized data transfer layout is presented. It is found to provide nearly-optimal workload balance, lower memory usage and better computational performance than classical slice decomposition techniques using sparse matrix data structures. Predictive memory usage and parallel performance models are also established and observed to be in very good agreement with data corresponding to numerical fluid flow simulations performed through 3-dimensional packings of cylinders and polydisperse spheres.     

Modeling and control of the doubly fed induction generator wind turbine

The present paper deals with the modeling of wind turbine generation systems. The model of a doubly fed induction generator, along with the corresponding converter, crow bar protection and electrical grid is described. The different level control strategies both in normal operation and under voltage dig conditions are discussed, including speed control, torque and reactive power control for the rotor-side converter, reactive and DC voltage control for the grid-side converter and the corresponding current loops control. The results obtained with simulations are compared to experimental data obtained from voltage sags provoked to real wind turbines.     

Real-time massive convolution for audio applications on GPU

Massive convolution is the basic operation in multichannel acoustic signal processing. This field has experienced a major development in recent years. One reason for this has been the increase in the number of sound sources used in playback applications available to users. Another reason is the growing need to incorporate new effects and to improve the hearing experience. Massive convolution requires high computing capacity. GPUs offer the possibility of parallelizing these operations. This allows us to obtain the processing result in much shorter time and to free up CPU resources. One important aspect lies in the possibility of overlapping the transfer of data from CPU to GPU and vice versa with the computation, in order to carry out real-time applications. Thus, a synthesis of 3D sound scenes could be achieved with only a peer-to-peer music streaming environment using a simple GPU in your computer, while the CPU in the computer is being used for other tasks. Nowadays, these effects are obtained in theaters or funfairs at a very high cost, requiring a large quantity of resources. Thus, our work focuses on two mains points: to describe an efficient massive convolution implementation and to incorporate this task to real-time multichannel-sound applications.     

A geostatistical algorithm to reproduce lateral gradual facies transitions: Description and implementation

Valid representations of geological heterogeneity are fundamental inputs for quantitative models used in managing subsurface activities. Consequently, the simulation of realistic facies distributions is a significant aim. Realistic facies distributions are typically obtained by pixel-based, object-based or process-based methods. This work presents a pixel-based geostatistical algorithm suitable for reproducing lateral gradual facies transitions (LGFT) between two adjacent sedimentary bodies. Lateral contact (i.e. interfingering) between distinct depositional facies is a widespread geometric relationship that occurs at different scales in any depositional system. The algorithm is based on the truncation of the sum of a linear expectation trend and a random Gaussian field, and can be conditioned to well data. The implementation introduced herein also includes subroutines to clean and geometrically characterize the obtained LGFT. The cleaned sedimentary body transition provides a more appropriate and realistic facies distribution for some depositional settings. The geometric measures of the LGFT yield an intuitive measure of the morphology of the sedimentary body boundary, which can be compared to analogue data. An example of a LGFT obtained by the algorithm presented herein is also flow simulated, quantitatively demonstrating the importance of realistically reproducing them in subsurface models, if further flow-related accurate predictions are to be made.     

Robotics in Education

Collision avoidance is essential for safe robot manipulation. Especially with humans around, robots should work only when safety can be robustly guaranteed. In this paper, we propose using virtual impedance control for reactive, smooth, and consistent collision avoidance that interferes minimally with the original task. The virtual impedance control operates in the risk space, a vector space describing the possibilities of all forthcoming collisions, and is designed to elude all risks in a consistent response in order to create assuring human-robot interaction experiences. The proposed scheme intrinsically handles kinematic singularity and the activation of avoidance using a boundary layer defined on the spectrum of Jacobian. In cooperation with the original controller, the proposed avoidance scheme provides a proof of convergence if the original controller is stable with and without projection. In simulations and experiments, we verified the characteristics of the proposed control scheme and integrated the system with Microsoft Kinect to monitor the workspace for real-time collision detection and avoidance. The results show that the proposed approach is suitable for robot operation with humans nearby.     

Multi-scale stacked sequential learning

Sequential learning is the discipline of machine learning that deals with dependent data such that neighboring labels exhibit some kind of relationship. The paper main contribution is two-fold: first, we generalize the stacked sequential learning, highlighting the key role of neighboring interactions modeling. Second, we propose an effective and efficient way of capturing and exploiting sequential correlations that takes into account long-range interactions. We tested the method on two tasks: text lines classification and image pixel classification. Results on these tasks clearly show that our approach outperforms the standard stacked sequential learning as well as state-of-the-art conditional random fields.     

Homogenization-based multiscale crack modelling: From micro-diffusive damage to macro-cracks

The existence of a representative volume element (RVE) for a class of quasi-brittle materials having a random heterogeneous microstructure in tensile, shear and mixed mode loading is demonstrated by deriving traction–separation relations, which are objective with respect to RVE size. A computational homogenization based multiscale crack modelling framework, implemented in an FE² setting, for quasi-brittle solids with complex random microstructure is presented. The objectivity of the macroscopic response to the micro-sample size is shown by numerical simulations. Therefore, a homogenization scheme, which is objective with respect to macroscopic discretization and microscopic sample size, is devised. Numerical examples including a comparison with direct numerical simulation are given to demonstrate the performance of the proposed method.     

3D porosity prediction from seismic inversion and neural networks

In this work, we address the problem of transforming seismic reflection data into an intrinsic rock property model. Specifically, we present an application of a methodology that allows interpreters to obtain effective porosity 3D maps from post-stack 3D seismic amplitude data, using measured density and sonic well log data as constraints. In this methodology, a 3D acoustic impedance model is calculated from seismic reflection amplitudes by applying an L₁-norm sparse-spike inversion algorithm in the time domain, followed by a recursive inversion performed in the frequency domain. A 3D low-frequency impedance model is estimated by kriging interpolation of impedance values calculated from well log data. This low-frequency model is added to the inversion result which otherwise provides only a relative numerical scale. To convert acoustic impedance into a single reservoir property, a feed-forward Neural Network (NN) is trained, validated and tested using gamma-ray and acoustic impedance values observed at the well log positions as input and effective porosity values as target. The trained NN is then applied for the whole reservoir volume in order to obtain a 3D effective porosity model. While the particular conclusions drawn from the results obtained in this work cannot be generalized, such results suggest that this workflow can be applied successfully as an aid in reservoir characterization, especially when there is a strong non-linear relationship between effective porosity and acoustic impedance.     

Safety implications of cultural and cognitive issues in nuclear power plant operation

This research project was designed to investigate cultural and cognitive issues related to the work of nuclear power plant operators during their time on the job in the control room and during simulator training (emergency situations), in order to show how these issues impact on plant safety. The modeling of the operators work deals with the use of operational procedures, the constant changes in the focus of attention and the dynamics of the conflicting activities. The paper focuses on the relationships between the courses of action of the different operators and the constraints imposed by their working environment. It shows that the safety implications of the control room operators' cognitive and cultural issues go far beyond the formal organizational constructs usually implied. Our findings indicate that the competence required for the operators are concerned with developing the possibility of constructing situation awareness, managing conflicts, gaps and time problems created by ongoing task procedures, and dealing with distractions, developing skills for collaborative work.     

Petri net-based engine for adaptive learning

In this paper, an IMS LD engine based on a Petri net model that represents the operational semantics of units of learning based on this specification is presented. The Petri nets of this engine, which is called OPENET4LD, verify the structural properties that are desirable for a learning flow and also facilitate the adaptation of the engine if potential changes in the IMS LD specification were proposed. Furthermore, OPENET4LD has an open and flexible architecture based on a set of ontologies that describe both the semantics of the Petri nets execution and the semantics of each learning flow component of IMS LD. Furthermore, the implementation of this architecture has been exhaustively validated with a number of UoLs that are compliant with the levels A and B of IMS LD.