There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

Three parallel algorithms for classical molecular dynamics are presented. The first assigns each processor a fixed subset of atoms; the second assigns each a fixed subset of inter-atomic forces to compute; the third assigns each a fixed spatial region. The algorithms are suitable for molecular dynamics models which can be difficult to parallelize efficiently—those with short-range forces where the neighbors of each atom change rapidly. They can be implemented on any distributed-memory parallel machine which allows for message-passing of data between independently executing processors. The algorithms are tested on a standard Lennard-Jones benchmark problem for system sizes ranging from 500 to 100,000,000 atoms on several parallel supercomputers--the nCUBE 2, Intel iPSC/860 and Paragon, and Cray T3D. Comparing the results to the fastest reported vectorized Cray Y-MP and C90 algorithm shows that the current generation of parallel machines is competitive with conventional vector supercomputers even for small problems. For large problems, the spatial algorithm achieves parallel efficiencies of 90% and a 1840-node Intel Paragon performs up to 165 faster than a single Cray C9O processor. Trade-offs between the three algorithms and guidelines for adapting them to more complex molecular dynamics simulations are also discussed.

Fast parallel algorithms for short-range molecular dynamics

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

Crop diseases are a major threat to food security, but their rapid identification remains difficult in many parts of the world due to the lack of the necessary infrastructure. The combination of increasing global smartphone penetration and recent advances in computer vision made possible by deep learning has paved the way for smartphone-assisted disease diagnosis. Using a public dataset of 54,306 images of diseased and healthy plant leaves collected under controlled conditions, we train a deep convolutional neural network to identify 14 crop species and 26 diseases (or absence thereof). The trained model achieves an accuracy of 99.35% on a held-out test set, demonstrating the feasibility of this approach. Overall, the approach of training deep learning models on increasingly large and publicly available image datasets presents a clear path toward smartphone-assisted crop disease diagnosis on a massive global scale.

/pdf/using-deep-learning-for-image-based-plant-disease-detection-3ep7b222h5.pdf

Using Deep Learning for Image-Based Plant Disease Detection

Deep learning in agriculture: A survey

Neural network-based approaches for solving partial differential equations (PDEs) have recently received special attention. However, the large majority of neural PDE solvers only apply to rectilinear domains, and do not systematically address the imposition of Dirichlet/Neumann boundary conditions over irregular domain boundaries. In this paper, we present a framework to neurally solve partial differential equations over domains with irregularly shaped (non-rectilinear) geometric boundaries. Our network takes in the shape of the domain as an input (represented using an unstructured point cloud, or any other parametric representation such as Non-Uniform Rational B-Splines) and is able to generalize to novel (unseen) irregular domains; the key technical ingredient to realizing this model is a novel approach for identifying the interior and exterior of the computational grid in a differentiable manner. We also perform a careful error analysis which reveals theoretical insights into several sources of error incurred in the model-building process. Finally, we showcase a wide variety of applications, along with favorable comparisons with ground truth solutions. shaped domains by building on well-established ﬁnite element and immersed boundary methods. Our neural PDE solver, coined IBN, demonstrates the ability to predict ﬁeld solutions for irregular boundaries immersed in the target domain. We highlight two speciﬁc PDE cases, Poisson and Navier-Stokes, which show promising results. Alongside the empirical results, we have included theoretical results for the error bounds of the optimization process of our ﬁnite element-based loss function. IBN opens up fast design exploration and topology optimization for various societally critical applications such as room ventilation for reduced disease risk, shape design for energy harvesters, and aerodynamic design of vehicles.

Neural PDE Solvers for Irregular Domains

Solidification of materials to near net shape is one of the most commonly used and economical methods of manufacturing. Different industries impose different restrictions on the solidification process. For example, single crystal growth requires a planar growth front, whereas casting requires homogenous material distribution. There are various techniques to control the flow in the melt to achieve the required objectives. Magnetic field based control techniques have become the most popular, commercially used methods. The application of a non-uniform external magnetic field produces an extra body force (Kelvin force) in the melt which along with the Lorenz force can be used to effectively control solidification and crystal growth processes. The physics behind the application of a time varying magnetic gradient on the flow is discussed and a coupled set of equations governing the fluid flow, thermal and concentration fields is determined. In this work, we only consider the solidification of non-conducting melts in the presence of magnetic fields and magnetic field gradients. A computational method for the design of solidification of non-conducting materials is developed such that a prescribed characteristic during solidification is achieved. An appropriate cost functional is defined. The cost function is here taken as the square of the L2 norm of the deviation of the velocity field in the melt region from conditions corresponding to convection-less growth. The adjoint method for the inverse design of continuum processes is adopted in this framework. A continuum adjoint system is derived to calculate the adjoint temperature, concentration and velocity such that the gradient of the cost functional can be expressed analytically. The cost functional is minimized using the conjugate gradient method with a finite element realization of the continuum direct, sensitivity and adjoint problems. An example of designing the time history of the magnetic field is presented. Similar developments to the ones discussed here have been achieved for the solidification of conducting melts. Thus the developed methodology has wide applications in crystal growth and the directional solidification of materials, organic compounds and biological macromolecules.

/pdf/melt-flow-control-using-magnetic-fields-and-magnetic-field-19l0i1owb2.pdf

Melt flow control using magnetic fields and magnetic field gradients

This article presents CyRSoXS – an open-source virtual instrument that uses GPUs to simulate polarized resonant soft X-ray scattering (P-RSoXS) patterns from real-space material representations. It is significantly faster than the current state-of-the-art software, and it enables quantitative extraction of orientation information from P-RSoXS data. This enables a wide range of applications, including pattern fitting, co-simulation, data exploration, machine learning workflows and multi-modal data assimilation approaches.

CyRSoXS: a GPU-accelerated virtual instrument for polarized resonant soft X-ray scattering

Phenotypic studies require large datasets for accurate inference and prediction. Collecting plant data in a farm can be very labor intensive and costly. This paper presents the design, architecture (hardware and software) and deployment of a distributed modular agricultural multi-robot system for row crop field data collection. The proposed system has been deployed in a soybean research farm at Iowa State University.

/pdf/a-novel-multirobot-system-for-distributed-phenotyping-3auzrtinhn.pdf

A Novel Multirobot System for Distributed Phenotyping

Accurate simulation of the printing process is essential for improving print quality, reducing waste, and optimizing the printing parameters of extrusion-based additive manufacturing. Traditional additive manufacturing simulations are very compute-intensive and are not scalable to simulate even moderately-sized geometries. In this paper, we propose a general framework for creating a digital twin of the dynamic printing process by performing physics simulations with the intermediate print geometries. Our framework takes a general extrusion-based additive manufacturing G-code, generates an analysis-suitable voxelized geometry representation from the print schedule, and performs physics-based (transient thermal and phase change) simulations of the printing process. Our approach leverages parallel adaptive octree meshes for both voxelated geometry representation as well as for fast simulations to address real-time predictions. We demonstrate the effectiveness of our method by simulating the printing of complex geometries at high voxel resolutions with both sparse and dense infills. Our results show that this approach scales to high voxel resolutions and can predict the transient heat distribution as the print progresses. This work lays the computational and algorithmic foundations for building real-time digital twins and performing rapid virtual print sequence exploration to improve print quality and further reduce material waste.

Baskar Ganapathysubramanian

Papers

Neural PDE Solvers for Irregular Domains

Melt flow control using magnetic fields and magnetic field gradients

CyRSoXS: a GPU-accelerated virtual instrument for polarized resonant soft X-ray scattering

A Novel Multirobot System for Distributed Phenotyping

Geometric Modeling and Physics Simulation Framework for Building a Digital Twin of Extrusion-based Additive Manufacturing