Search NASA⌕ Search

SEARCH · Search NASA

Results for “Vectorized algorithm”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,765 records · Page 98

Microwave holography of large reflector antennas - Simulation algorithms

The performance of large reflector antennas can be improved by identifying the location and amount of their surface distortions and correcting them. To determine the accuracy of the constructed surface profiles, simulation studies are used to incorporate both the effects of systematic and random distortions, particularly the effects of the displaced surface panels. In this paper, different simulation models are investigated, emphasizing a model based on the vector diffraction analysis of a curved reflector with displaced panels. The simulated far-field patterns are then used to reconstruct the location and amount of displacement of the surface panels by employing a fast Fourier transform/iterative procedure. The sensitivity of the microwave holography technique based on the number of far-field sampled points, level of distortions, polarizations, illumination tapers, etc., is also examined.

Rahmat-Samii, Y.↗

Exploiting data representation for fault tolerance

Incorrect computer hardware behavior may corrupt intermediate computations in numerical algorithms, possibly resulting in incorrect answers. Prior work models misbehaving hardware by randomly flipping bits in memory. We start by accepting this premise, and present an analytic model for the error introduced by a bit flip in an IEEE 754 floating-point number. We then relate this finding to the linear algebra concepts of normalization and matrix equilibration. In particular, we present a case study illustrating that normalizing both vector inputs of a dot product minimizes the probability of a single bit flip causing a large error in the dot product's result. Moreover, the absolute error is either less than one or very large, which allows detection of large errors. Then, we apply this to the GMRES iterative solver. We count all possible errors that can be introduced through faults in arithmetic in the computationally intensive orthogonalization phase of GMRES, and show that when the matrix is equilibrated, the absolute error is bounded above by one.

97 MATHEMATICS AND COMPUTING↗

WE-DE-201-04: Cross Validation of Knowledge-Based Treatment Planning for Prostate LDR Brachytherapy Using Principle Component Analysis

Purpose: To validate a knowledge-based algorithm for prostate LDR brachytherapy treatment planning. Methods: A dataset of 100 cases was compiled from an active prostate seed implant service. Cases were randomized into 10 subsets. For each subset, the 90 remaining library cases were registered to a common reference frame and then characterized on a point by point basis using principle component analysis (PCA). Each test case was converted to PCA vectors using the same process and compared with each library case using a Mahalanobis distance to evaluate similarity. Rank order PCA scores were used to select the best-matched library case. The seed arrangement was extracted from the best-matched case and used as a starting point for planning the test case. Any subsequent modifications were recorded that required input from a treatment planner to achieve V100>95%, V150<60%, V200<20%. To simulate operating-room planning constraints, seed activity was held constant, and the seed count could not increase. Results: The computational time required to register test-case contours and evaluate PCA similarity across the library was 10s. Preliminary analysis of 2 subsets shows that 9 of 20 test cases did not require any seed modifications to obtain an acceptable plan. Five test cases required fewer than 10 seed modifications or a grid shift. Another 5 test cases required approximately 20 seed modifications. An acceptable plan was not achieved for 1 outlier, which was substantially larger than its best match. Modifications took between 5s and 6min. Conclusion: A knowledge-based treatment planning algorithm for prostate LDR brachytherapy is being cross validated using 100 prior cases. Preliminary results suggest that for this size library, acceptable plans can be achieved without planner input in about half of the cases while varying amounts of planner input are needed in remaining cases. Computational time and planning time are compatible with clinical practice.

60 APPLIED LIFE SCIENCES↗

Towards a high performance geometry library for particle-detector simulations

Thread-parallelization and single-instruction multiple data (SIMD) ”vectorisation” of software components in HEP computing has become a necessity to fully benefit from current and future computing hardware. In this context, the Geant-Vector/GPU simulation project aims to re-engineer current software for the simulation of the passage of particles through detectors in order to increase the overall event throughput. As one of the core modules in this area, the geometry library plays a central role and vectorising its algorithms will be one of the cornerstones towards achieving good CPU performance. Here, we report on the progress made in vectorising the shape primitives, as well as in applying new C++ template based optimizations of existing code available in the Geant4, ROOT or USolids geometry libraries. We will focus on a presentation of our software development approach that aims to provide optimized code for all use cases of the library (e.g., single particle and many-particle APIs) and to support different architectures (CPU and GPU) while keeping the code base small, manageable and maintainable. We report on a generic and templated C++ geometry library as a continuation of the AIDA USolids project. As a result, the experience gained with these developments will be beneficial to other parts of the simulation software, such as for the optimization of the physics library, and possibly to other parts of the experiment software stack, such as reconstruction and analysis.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

A High Performance Block Eigensolver for Nuclear Configuration Interaction Calculations

As on-node parallelism increases and the performance gap between the processor and the memory system widens, achieving high performance in large-scale scientific applications requires an architecture-aware design of algorithms and solvers. We focus on the eigenvalue problem arising in nuclear Configuration Interaction (CI) calculations, where a few extreme eigenpairs of a sparse symmetric matrix are needed. Here, we consider a block iterative eigensolver whose main computational kernels are the multiplication of a sparse matrix with multiple vectors (SpMM), and tall-skinny matrix operations. We then present techniques to significantly improve the SpMM and the transpose operation SpMM T by using the compressed sparse blocks (CSB) format. We achieve 3-4× speedup on the requisite operations over good implementations with the commonly used compressed sparse row (CSR) format. We develop a performance model that allows us to correctly estimate the performance of our SpMM kernel implementations, and we identify cache bandwidth as a potential performance bottleneck beyond DRAM. We also analyze and optimize the performance of LOBPCG kernels (inner product and linear combinations on multiple vectors) and show up to 15× speedup over using high performance BLAS libraries for these operations. The resulting high performance LOBPCG solver achieves 1.4× to 1.8× speedup over the existing Lanczos solver on a series of CI computations on high-end multicore architectures (Intel Xeons). We also analyze the performance of our techniques on an Intel Xeon Phi Knights Corner (KNC) processor.

97 MATHEMATICS AND COMPUTING↗

Few-view cone-beam CT reconstruction with deformed prior image

Purpose: Prior images can be incorporated into the image reconstruction process to improve the quality of subsequent cone-beam CT (CBCT) images from sparse-view or low-dose projections. The purpose of this work is to develop a deformed prior image-based reconstruction (DPIR) strategy to mitigate the deformation between the prior image and the target image. Methods: The deformed prior image is obtained by a projection-based registration approach. Specifically, the deformation vector fields used to deform the prior image are estimated through iteratively matching the forward projection of the deformed prior image and the measured on-treatment projections. The deformed prior image is then used as the prior image in the standard prior image constrained compressed sensing (PICCS) algorithm. A simulation study on an XCAT phantom and a clinical study on a head-and-neck cancer patient were conducted to evaluate the performance of the proposed DPIR strategy. Results: The deformed prior image matches the geometry of the on-treatment CBCT more closely as compared to the original prior image. Consequently, the performance of the DPIR strategy from few-view projections is improved in comparison to the standard PICCS algorithm, based on both visual inspection and quantitative measures. In the XCAT phantom study using 20 projections, the average root mean squared error is reduced from 14% in PICCS to 10% in DPIR, and the average universal quality index increases from 0.88 in PICCS to 0.92 in DPIR. Conclusions: The present DPIR approach provides a practical solution to the mismatch problem between the prior image and target image, which improves the performance of the original PICCS algorithm for CBCT reconstruction from few-view or low-dose projections.

60 APPLIED LIFE SCIENCES↗

A Novel Coarsening Method for Scalable and Efficient Mesh Generation

In this paper, we propose a novel mesh coarsening method called brick coarsening method. The proposed method can be used in conjunction with any graph partitioners and scales to very large meshes. This method reduces problem space by decomposing the original mesh into fixed-size blocks of nodes called bricks, layered in a similar way to conventional brick laying, and then assigning each node of the original mesh to appropriate brick. Our experiments indicate that the proposed method scales to very large meshes while allowing simple RCB partitioner to produce higher-quality partitions with significantly less edge cuts. Our results further indicate that the proposed brick-coarsening method allows more complicated partitioners like PT-Scotch to scale to very large problem size while still maintaining good partitioning performance with relatively good edge-cut metric. Graph partitioning is an important problem that has many scientific and engineering applications in such areas as VLSI design, scientific computing, and resource management. Given a graph G = (V,E), where V is the set of vertices and E is the set of edges, (k-way) graph partitioning problem is to partition the vertices of the graph (V) into k disjoint groups such that each group contains roughly equal number of vertices and the number of edges connecting vertices in different groups is minimized. Graph partitioning plays a key role in large scientific computing, especially in mesh-based computations, as it is used as a tool to minimize the volume of communication and to ensure well-balanced load across computing nodes. The impact of graph partitioning on the reduction of communication can be easily seen, for example, in different iterative methods to solve a sparse system of linear equation. Here, a graph partitioning technique is applied to the matrix, which is basically a graph in which each edge is a non-zero entry in the matrix, to allocate groups of vertices to processors in such a way that many of matrix-vector multiplication can be performed locally on each processor and hence to minimize communication. Furthermore, a good graph partitioning scheme ensures the equal amount of computation performed on each processor. Graph partitioning is a well known NP-complete problem, and thus the most commonly used graph partitioning algorithms employ some forms of heuristics. These algorithms vary in terms of their complexity, partition generation time, and the quality of partitions, and they tend to trade off these factors. A significant challenge we are currently facing at the Lawrence Livermore National Laboratory is how to partition very large meshes on massive-size distributed memory machines like IBM BlueGene/P, where scalability becomes a big issue. For example, we have found that the ParMetis, a very popular graph partitioning tool, can only scale to 16K processors. An ideal graph partitioning method on such an environment should be fast and scale to very large meshes, while producing high quality partitions. This is an extremely challenging task, as to scale to that level, the partitioning algorithm should be simple and be able to produce partitions that minimize inter-processor communications and balance the load imposed on the processors. Our goals in this work are two-fold: (1) To develop a new scalable graph partitioning method with good load balancing and communication reduction capability. (2) To study the performance of the proposed partitioning method on very large parallel machines using actual data sets and compare the performance to that of existing methods. The proposed method achieves the desired scalability by reducing the mesh size. For this, it coarsens an input mesh into a smaller size mesh by coalescing the vertices and edges of the original mesh into a set of mega-vertices and mega-edges. A new coarsening method called brick algorithm is developed in this research. In the brick algorithm, the zones in a given mesh are first grouped into fixed size blocks called bricks. These brick are then laid in a way similar to conventional brick laying technique, which reduces the number of neighboring blocks each block needs to communicate. Contributions of this research are as follows: (1) We have developed a novel method that scales to a really large problem size while producing high quality mesh partitions; (2) We measured the performance and scalability of the proposed method on a machine of massive size using a set of actual large complex data sets, where we have scaled to a mesh with 110 million zones using our method. To the best of our knowledge, this is the largest complex mesh that a partitioning method is successfully applied to; and (3) We have shown that proposed method can reduce the number of edge cuts by as much as 65%.

97 MATHEMATICS AND COMPUTING↗

Mean curl formulation on quadrilaterals with application to implicit magnetics diffusion equations in Alegra 2D

This report proposes a mean curl on quadrilateral elements for Alegra 2D magnetics calculations. The resulting quadrilateral element is an extension of the work [1, 4] to include the curl operator. The resulting quadrilateral element implementation is applicable to implicit calculations in Carte sian coordinates which use either the vector potential or B formulation. It is useful to note that in the vector potential formulation, the mean curl produces a mean flux density on the element, whereas in the B formulation the mean curl produces a mean current on the element. In both cases, the mean curl is, by construction, divergence free. The mean curl is further developed and applied to the nonlinear permeability capability that currently exists in Alegra 2D for the vector poten tial formulation. Using the proposed mean curl, a Jacobian operator is developed and presented which is applicable to nonlinear iterations associated with the implicit problem. Note that the exist ing nonlinear permeability does not include hysteresis; also, the current nonlinear implementation uses a Jacobian free solution algorithm because a Jacobian operator has neither been developed or implemented. A Jacobian for the full quadrature implementation that currently exists in Alegra 2D can be similarly derived based upon the results presented.

Mitchell, John Anthony↗

Optimization methods, flux conserving methods for steady state Navier-Stokes equation

Navier-Stokes equation as discretized by new flux conserving method proposed by Chang and Scott results in the system: vector F(vector x) = 0, where F is a vector valued function. The Optimization method we use is based on Quasi-Newton methods: given a nonlinear function vector F(vector x) = 0, we solve, Delta(vector x) = -BF(vector x), where Delta(vector x) is the correction term and B is the inverse Jacobian of F(x). Then, iteratively, vector(x(sub (i+1))) = vector(x (sub i)) + alpha.Delta(vector x(sub i)), where alpha is a line search correction term determined by a line search routine. We use the BFCG's update the Jacobian matrix B(sub k) at each iteration. It is well known that B(sub k) approaches B(*) at the solution X(*). This algorithm has several advantages over the Newton-Raphson method. For example, we do not need to calculate the Jacobian matrix at each iteration which is computationally very expensive.

Adeyeye, John↗

Acoustic Calibration of the Exterior Effects Room at the NASA Langley Research Center

The Exterior Effects Room (EER) at the NASA Langley Research Center is a 39-seat auditorium built for psychoacoustic studies of aircraft community noise. The original reproduction system employed monaural playback and hence lacked sound localization capability. In an effort to more closely recreate field test conditions, a significant upgrade was undertaken to allow simulation of a three-dimensional audio and visual environment. The 3D audio system consists of 27 mid and high frequency satellite speakers and 4 subwoofers, driven by a real-time audio server running an implementation of Vector Base Amplitude Panning. The audio server is part of a larger simulation system, which controls the audio and visual presentation of recorded and synthesized aircraft flyovers. The focus of this work is on the calibration of the 3D audio system, including gains used in the amplitude panning algorithm, speaker equalization, and absolute gain control. Because the speakers are installed in an irregularly shaped room, the speaker equalization includes time delay and gain compensation due to different mounting distances from the focal point, filtering for color compensation due to different installations (half space, corner, baffled/unbaffled), and cross-over filtering.

Faller, Kenneth J., II↗

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-of-the-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Deep-learning↗

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-ofthe-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Hesham ElAbd↗

Universal Decoder for PPM of any Order

A recently developed algorithm for demodulation and decoding of a pulse-position- modulation (PPM) signal is suitable as a basis for designing a single hardware decoding apparatus to be capable of handling any PPM order. Hence, this algorithm offers advantages of greater flexibility and lower cost, in comparison with prior such algorithms, which necessitate the use of a distinct hardware implementation for each PPM order. In addition, in comparison with the prior algorithms, the present algorithm entails less complexity in decoding at large orders. An unavoidably lengthy presentation of background information, including definitions of terms, is prerequisite to a meaningful summary of this development. As an aid to understanding, the figure illustrates the relevant processes of coding, modulation, propagation, demodulation, and decoding. An M-ary PPM signal has M time slots per symbol period. A pulse (signifying 1) is transmitted during one of the time slots; no pulse (signifying 0) is transmitted during the other time slots. The information intended to be conveyed from the transmitting end to the receiving end of a radio or optical communication channel is a K-bit vector u. This vector is encoded by an (N,K) binary error-correcting code, producing an N-bit vector a. In turn, the vector a is subdivided into blocks of m = log2(M) bits and each such block is mapped to an M-ary PPM symbol. The resultant coding/modulation scheme can be regarded as equivalent to a nonlinear binary code. The binary vector of PPM symbols, x is transmitted over a Poisson channel, such that there is obtained, at the receiver, a Poisson-distributed photon count characterized by a mean background count nb during no-pulse time slots and a mean signal-plus-background count of ns+nb during a pulse time slot. In the receiver, demodulation of the signal is effected in an iterative soft decoding process that involves consideration of relationships among photon counts and conditional likelihoods of m-bit vectors of coded bits. Inasmuch as the likelihoods of all the m-bit vectors of coded bits mapping to the same PPM symbol are correlated, the best performance is obtained when the joint mbit conditional likelihoods are utilized. Unfortunately, the complexity of decoding, measured in the number of operations per bit, grows exponentially with m, and can thus become prohibitively expensive for large PPM orders. For a system required to handle multiple PPM orders, the cost is even higher because it is necessary to have separate decoding hardware for each order. This concludes the prerequisite background information. In the present algorithm, the decoding process as described above is modified by, among other things, introduction of an lbit marginalizer sub-algorithm. The term "l-bit marginalizer" signifies that instead of m-bit conditional likelihoods, the decoder computes l-bit conditional likelihoods, where l is fixed. Fixing l, regardless of the value of m, makes it possible to use a single hardware implementation for any PPM order. One could minimize the decoding complexity and obtain an especially simple design by fixing l at 1, but this would entail some loss of performance. An intermediate solution is to fix l at some value, greater than 1, that may be less than or greater than m. This solution makes it possible to obtain the desired flexibility to handle any PPM order while compromising between complexity and loss of performance.

Moision, Bruce E.↗

Results of the Magnetometer Navigation (MAGNAV) Inflight Experiment

The Magnetometer Navigation (MAGNAV) algorithm is currently running as a flight experiment as part of the Wide Field Infrarad Explorer Post-Science Engineer- ing Testbed. Initialization of MAGNAV occured on September 4, 2004. MAGNAV is designed to autonomously estimate the spacecraft orbit, attitude, and rate using magnetometer and sun sensor data. Since the earth s magnetic field is a function of time and position, and since time is known quite precisely, the differences between the computed magnetic field and measured magnetic field components, as measured by the magnetometer throughout the entire spacecraft orbit, are a function of the spacecraft trajectory and attitude errors. Therefore, these errors are used to estimate both trajec- tory and attitude. In addition, the time rate of change of the magnetic field vector is used to estimate the spacecraft rotation rate. The estimation of the attitude and tra- jectory is augmented with the rate estimation into an Extended Kalman filter blended with a pseudc-linear Kalman filter. Sun sensor data is also used to improve the accu- racy and observability of the attitude and rate estimates. This test serves to validate MAGNAV as a single low cost navigation system which utilizes reliable, flight qualified sensors. MAGNAV is intended as a backup algorithm, an initialization algorithm, or possibly a prime navigation algorithm for a mission with coarse constraints.

Thienel, Julie↗

CUDA Computation of the Feynman Distribution

In 2006, NVIDIA Corporation introduced the Computer Unified Device Architecture (CUDA) for C, C++, and Fortran programming languages. The CUDA parallel computing platform uses the Graphics Processing Unit (GPU) of NVIDIA graphics card, and it is designed for parallel computations. In these computations, the same algorithms are executed on the data in parallel. The GPU contains many Streaming Multiprocessors (SM) that can independently execute blocks of threads. A thread is a sequence of serial programmed instructions. The SM executes threads in groups of 32 and a group of 32 threads is called warp. The threads within a block can be synchronized and have access to a shared memory that is visible only to them. Conversely, blocks of threads cannot be synchronized and have access only to the global memory of the graphics card. This memory is called device memory and is independent of the host memory controlled by the Central Processing Unit (CPU). A single thread has access to a local memory (register) that is not visible to any other thread. The part of the program that is executed on the graphics card is called kernel. Each thread of the kernel has a unique identification index that is accessible by the in-built variable threadIdx; the latter is a 3-component vector. A threads block can contain up to 1024 threads. Blocks of threads have a unique identification index that is accessible by the in-built variable blockIdx; the latter is a 3-component vector. The in-built variable blockDim contains the number of threads per block; this variable is a 3-component vector. Threads and blocks represent different levels of parallelism; a kernel is executed as a grid of blocks of threads. Figure 1 illustrates a grid of thread blocks for a GPU with 2 SM. A SM can execute multiple blocks of threads in parallel; a block of threads is executed on a single SM. A GPU can execute multiple kernels concurrently. This work has used the Quadro M5000 graphics card with the following technical specifications: - 2048 cores with 861 MHz frequency, - 16 streaming multiprocessors, - 5.2 trillion transistors, - 8.2 Gb device memory, - 64 kb register memory (per thread), - 48 kb shared memory (per block) - 211.6 GB/s memory bandwidth, and - 3.5 GFLOPS floating point performance.

97 MATHEMATICS AND COMPUTING↗

Improving qubit readout with hidden Markov models

We demonstrate the application of pattern recognition algorithms via hidden Markov models (HMM) for qubit readout. This scheme provides a state-path trajectory approach capable of detecting qubit-state transitions and makes for a robust classification scheme with higher starting-state assignment fidelity than when compared to a multivariate Gaussian or a support vector machine scheme. Therefore, the method also eliminates the qubit-dependent readout time optimization requirement in current schemes. Using a HMM state discriminator we estimate fidelities reaching the ideal limit. Unsupervised learning gives access to transition matrix, priors, and IQ distributions, providing a toolbox for studying qubit-state dynamics during strong projective readout.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Short-Term Distribution System State Forecast Based on Optimal Synchrophasor Sensor Placement and Extreme Learning Machine

This paper proposes an approach for distribution system state forecasting, which aims to provide an accurate and high speed state forecasting with an optimal synchrophasor sensor placement (OSSP) based state estimator and an extreme learning machine (ELM) based forecaster. Specifically, considering the sensor installation cost and measurement error, an OSSP algorithm is proposed to reduce the number of synchrophasor sensor and keep the whole distribution system numerically and topologically observable. Then, the weighted least square (WLS) based system state estimator is used to produce the training data for the proposed forecaster. Traditionally, the artificial neural network (ANN) and support vector regression (SVR) are widely used in forecasting due to their nonlinear modeling capabilities. However, the ANN contains heavy computation load and the best parameters for SVR are difficult to obtain. In this paper, the ELM, which overcomes these drawbacks, is used to forecast the future system states with the historical system states. The proposed approach is effective and accurate based on the testing results.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Efficient Probabilistic Visualization of Local Divergence of 2D Vector Fields with Independent Gaussian Uncertainty

This work focuses on visualizing uncertainty of local divergence of two-dimensional vector fields. Divergence is one of the fundamental attributes of fluid flows, as it can help domain scientists analyze potential positions of sources (positive divergence) and sinks (negative divergence) in the flow. However, uncertainty inherent in vector field data can lead to erroneous divergence computations, adversely impacting downstream analysis. While Monte Carlo (MC) sampling is a classical approach for estimating divergence uncertainty, it suffers from slow convergence and poor scalability with increasing data size and sample counts. Thus, we present a two-fold contribution that tackles the challenges of slow convergence and limited scalability of the MC approach. (1) We derive a closed-form approach for highly efficient and accurate uncertainty visualization of local divergence, assuming independently Gaussian-distributed vector uncertainties. (2) We further integrate our approach into Viskores, a platform-portable parallel library, to accelerate uncertainty visualization. In our results, we demonstrate significantly enhanced efficiency and accuracy of our serial analytical (speed-up up to 1946×) and parallel Viskores (speed-up up to 19698×) algorithms over the classical serial MC approach. We also demonstrate qualitative improvements of our probabilistic divergence visualizations over traditional mean-field visualization, which disregards uncertainty. We validate the accuracy and efficiency of our methods on wind forecast and ocean simulation datasets.

Ouermi, Timbwaoga [University of Utah]↗