We present new versions of sample CUDA programs for the GPU computing of the Swendsen-Wang multi-cluster spin flip algorithm. In this update, we add the method of GPU-based cluster-labeling algorithm without the use o...
详细信息
We present new versions of sample CUDA programs for the GPU computing of the Swendsen-Wang multi-cluster spin flip algorithm. In this update, we add the method of GPU-based cluster-labeling algorithm without the use of conventional iteration (Komura, 2015) to those programs. For high-precision calculations, we also add a random-number generator in the cuRAND library. Moreover, we fix several bugs and remove the extra usage of shared memory in the kernel functions.
The correlation dimensions in the financial market are calculated and used as a measure to study the cluster structure in the correlation coefficient matrix. First, based on the existing model, we present a toy model....
详细信息
The correlation dimensions in the financial market are calculated and used as a measure to study the cluster structure in the correlation coefficient matrix. First, based on the existing model, we present a toy model. Using the model-generated data, we find that the clearer cluster structure corresponds to a smaller dimension. It implies that the correlation dimension can be used as a measure of the cluster structure in the correlation coefficient matrix. Finally, we use the algorithm to compute the clusters in the real market and verify the previous empirical evidence. The results show that the cluster structure in the financial correlation coefficient matrix may change with time. The correlation dimension is smaller after the financial crisis, indicating that the cluster structure is clearer after the financial crisis. (C) 2017 Elsevier Ltd. All rights reserved.
Accurate forecasting of residential energy loads is highly influenced by the use of electrical appliances, which not only affect electrical energy use but also internal heat gains, which in turn affects thermal energy...
详细信息
Accurate forecasting of residential energy loads is highly influenced by the use of electrical appliances, which not only affect electrical energy use but also internal heat gains, which in turn affects thermal energy use. It is therefore important to accurately understand the characteristics of appliance use and to embed this understanding into predictive models to support load forecast and building design decisions. Bottom-up techniques that account for the variability in socio-demographic characteristics of the occupants and their behaviour patterns constitute a powerful tool to this end, and are potentially able to inform the design of Demand Side Management strategies in homes. To this end, this paper presents a comparison of alternative strategies to stochastically model the temporal energy use of low-load appliances (meaning those whose annual energy share is individually small but significant when considered as a group). In particular, discrete-time Markov processes and survival analysis have been explored. Rigorous mathematical procedures, including cluster analysis, have been employed to identify a parsimonious strategy for the modelling of variations in energy demand over time of the four principle categories of small appliances: audio-visual, computing, kitchen and other small appliances. From this it is concluded that a model of the duration for which appliances survive in discrete states expressed as bins in fraction of maximum power demand performs best. This general solution may be integrated with relative ease with dynamic simulation programs, to complement existing models of relatively large load appliances for the comprehensive simulation of household appliance use. Crown Copyright (C) 2016 Published by Elsevier B.V. All rights reserved.
We perform the high-performance computation of the ferromagnetic Ising model on the pyrochlore lattice. We determine the critical temperature accurately based on the finite-size staling of the Binder ratio. Comparing ...
详细信息
We perform the high-performance computation of the ferromagnetic Ising model on the pyrochlore lattice. We determine the critical temperature accurately based on the finite-size staling of the Binder ratio. Comparing with the data on the simple cubic lattice, we argue the universal finite-size scaling. We also calculate the classical XY model and the classical Heisenberg model on the pyrochlore lattice. (C) 2016 Elsevier B.V. All rights reserved.
Transmission System Operators must regularly perform different stability studies in order to detect hazard situations in the system and to define proper corrective measures. Due to the high computational burden requir...
详细信息
ISBN:
(纸本)9781538622124
Transmission System Operators must regularly perform different stability studies in order to detect hazard situations in the system and to define proper corrective measures. Due to the high computational burden required by dynamic simulations, the number of study cases must be strongly limited to only a few cases. Ideally, only worst cases from a system stability point of view should be identified and used. However, the increased complexity of power system dynamics under high levels of variable generation technologies makes it more difficult to define worst-case scenarios for stability assessments. This paper proposes a novel methodology for selecting critical operating conditions (including operating point and contingency) for frequency stability studies. The methodology considers key factors influencing the system frequency stability, such as the power imbalance, the frequency nadir, the system inertia, the ROCOF and the short circuit level at the point of the contingency. The proposed methodology is validated considering the northern power system of Chile. The dynamic simulations show that the proposed methodology is able to identify critical operating conditions for frequency stability assessments. Furthermore, obtained results show that for the system under study, some of the selected critical operating points have a worse dynamic performance than the one that would be normally chosen with the traditional worst-case approach.
We present sample OpenACC programs of the Swendsen-Wang multi-cluster spin flip algorithm. OpenACC is a directive-based programming model for accelerators without requiring modification to the underlying CPU code itse...
详细信息
We present sample OpenACC programs of the Swendsen-Wang multi-cluster spin flip algorithm. OpenACC is a directive-based programming model for accelerators without requiring modification to the underlying CPU code itself. In this paper, we deal with the classical spin models as with the sample CUDA programs (Komura and Okabe, 2014), that is, two-dimensional (2D) Ising model, three-dimensional (3D) Ising model, 2D Potts model, 3D Potts model, 2D XY model and 3D XY model. We explain the details of sample OpenACC programs and compare the performance of the present OpenACC implementations with that of the CUDA implementations for the 2D and 3D Ising models and the 2D and 3D XY models. Program summary Program title: SWspin_OpenACC Catalogue identifier: AEXU_v1_0 Program summary URL: http://***/summaries/AEXU_v1_*** Program obtainable from: CPC Program Library, Queen's University, Belfast, N. Ireland Licensing provisions: Standard CPC licence, http://***/licence/*** No. of lines in distributed program, including test data, etc.: 2898 No. of bytes in distributed program, including test data, etc.: 9729 Distribution format: *** Programming language: C, OpenACC. Computer: Any computer with an OpenACC-enabled accelerator (tested on NVIDIA GPU). Operating system: No limits (tested on Linux). RAM: About 1MiB for the parameters used in the sample programs. Classification: 23. Nature of problem: Monte Carlo simulation of classical spin systems. Ising model, q-state Potts model, and the classical XY model are treated for both two-dimensional and three-dimensional lattices. Solution method: Swendsen-Wang multi-cluster spin flip Monte Carlo method. The OpenACC implementation for the cluster-labeling is based on the work by Kalentev et al. [J. Parallel Distrib. Comput. 71 (2011) 615-620]. Restrictions: The system size is limited depending on the memory of an accelerator. Running time: A few minutes per each program for the parameters used in th
We propose a new cluster algorithm for the Baxter-Wu model that significantly reduces critical slowing down. We examine the behavior of the created clusters as we vary the temperature and then specify dynamic exponent...
详细信息
We propose a new cluster algorithm for the Baxter-Wu model that significantly reduces critical slowing down. We examine the behavior of the created clusters as we vary the temperature and then specify dynamic exponents. Comparison is made with the Metropolis algorithm and with the other existing cluster algorithm. (C) 2010 Elsevier B.V. All rights reserved.
In this article, we propose a new Bayesian variable selection (BVS) approach via the graphical model and the Ising model, which we refer to as the "Bayesian Ising graphical model" (BIGM). The BIGM is develop...
详细信息
In this article, we propose a new Bayesian variable selection (BVS) approach via the graphical model and the Ising model, which we refer to as the "Bayesian Ising graphical model" (BIGM). The BIGM is developed by showing that the BVS problem based on the linear regression model can be considered as a complete graph and described by an Ising model with random interactions. There are several advantages of our BIGM: it is easy to (i) employ the single-site updating and cluster updating algorithm, both of which are suitable for problems with small sample sizes and a larger number of variables, (ii) extend this approach to nonparametric regression models, and (iii) incorporate graphical prior information. In our BIGM, the interactions are determined by the linear model coefficients, so we systematically study the performance of different scale normal mixture priors for the model coefficients by adopting the global-local shrinkage strategy. Our results indicate that the best prior for the model coefficients in terms of variable selection should place substantial weight on small, nonzero shrinkage. The methods are illustrated with simulated and real data. Supplementary materials for this article are available online.
Node occlusion and edge congestion problems, which are caused by the increasing scale and complexity of network, had become a hot spot in network visualization research. To solve the visual clutter problem in network,...
详细信息
ISBN:
(纸本)9781509011926
Node occlusion and edge congestion problems, which are caused by the increasing scale and complexity of network, had become a hot spot in network visualization research. To solve the visual clutter problem in network, edges close to each other in network were bundled by curving them. A segmental forced-directed algorithm (FDA) simplification model and a community based compatible edge bundling network model were proposed and improved. To solve the problem of excessive bending of some edges in segmental FDA bundling model, network was divided into different communities by CNM cluster algorithm. Experimental result shows that the bundling simplification algorithm introduced above has a wide applicability, and network visualized by this algorithm has good visual effect and readability.
Network traffic classification algorithm based on the machine learning has attracted more and more attention. Because the traditional EM algorithm has the disadvantage that the algorithm has the sensitivity of initial...
详细信息
ISBN:
(纸本)9781467381390
Network traffic classification algorithm based on the machine learning has attracted more and more attention. Because the traditional EM algorithm has the disadvantage that the algorithm has the sensitivity of initial value and converge to local optimal point easily. This paper proposed a new improved EM algorithm based on the q-DAEM. The improved algorithm applies the EM algorithm to generate a constrained matrix, then combine the constrained matrix with the q-DAEM algorithm to reduce the search range, so that a better Gaussian mixture model can he derived from this algorithm. The algorithm was applied to the Moore datasets for evaluation, the experimental results show that this improved algorithm which applied to network traffic classification can lead to a higher precision and overall accuracy.
暂无评论