Search NASA⌕ Search

SEARCH · Search NASA

Results for “approximate computing”

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 1,423 records · Page 79

Evaluating Space Object Conjunction Probabilities Using Characteristic Function Inversion

This report discusses an approach to computing the probability of a conjunction between two space objects in the short-term encounter scenario. A conjunction is defined here as an event where the miss distance between the objects is less than some specified value. The scenario assumptions are that the motion of the objects is linear, their positions are Gaussian distributed, and their velocities are known and constant. Under these assumptions, the squared-miss distance is shown to have the generalized chi-square distribution. An established statistical technique called characteristic function inversion is employed to evaluate the distribution and obtain conjunction probabilities. The method is closely related to a recent approach based on moment generating function inversion, and a qualitative comparison of the approaches is provided. Last, the method is tested on several benchmark test cases where it agrees with numerical integration on the cases with conjunction probabilities above 10 –12 . However, the exact probability in these cases is usually not needed and this probability can be bounded above using an independent Gaussian approximation. Overall, the report shows how to compute conjunction probabilities using a standard statistical method, though numerical integration seems to perform equally well.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Gradient Coding With Iterative Block Leverage Score Sampling

Gradient coding is a method for mitigating straggling servers in a centralized computing network that uses erasure-coding techniques to distributively carry out first-order optimization methods. Randomized numerical linear algebra uses randomization to develop improved algorithms for large-scale linear algebra computations. In this study, we propose a method for distributed optimization that combines gradient coding and randomized numerical linear algebra. The proposed method uses a randomized ℓ 2 -subspace embedding and a gradient coding technique to distribute blocks of data to the computational nodes of a centralized network, and at each iteration the central server only requires a small number of computations to obtain the steepest descent update. The novelty of our approach is that the data is replicated according to importance scores, called block leverage scores, in contrast to most gradient coding approaches that uniformly replicate the data blocks. Furthermore, we do not require a decoding step at each iteration, avoiding a bottleneck in previous gradient coding schemes. We show that our approach results in a valid ℓ 2 -subspace embedding, and that our resulting approximation converges to the optimal solution.

97 MATHEMATICS AND COMPUTING↗

Large Scale Computing and Storage Requirements for Nuclear Physics Research

The National Energy Research Scientific Computing Center (NERSC) is the primary computing center for the DOE Office of Science, serving approximately 4,000 users and hosting some 550 projects that involve nearly 700 codes for a wide variety of scientific disciplines. In addition to large-scale computing resources NERSC provides critical staff support and expertise to help scientists make the most efficient use of these resources to advance the scientific mission of the Office of Science. In May 2011, NERSC, DOE’s Office of Advanced Scientific Computing Research (ASCR) and DOE’s Office of Nuclear Physics (NP) held a workshop to characterize HPC requirements for NP research over the next three to five years. The effort is part of NERSC’s continuing involvement in anticipating future user needs and deploying necessary resources to meet these demands. The workshop revealed several key requirements, in addition to achieving its goal of characterizing NP computing. The key requirements include: 1. Larger allocations of computational resources at NERSC; 2. Visualization and analytics support; and 3. Support at NERSC for the unique needs of experimental nuclear physicists. This report expands upon these key points and adds others. The results are based upon representative samples, called “case studies,” of the needs of science teams within NP. The case studies were prepared by NP workshop participants and contain a summary of science goals, methods of solution, current and future computing requirements, and special software and support needs. Participants were also asked to describe their strategy for computing in the highly parallel, “multi-core” environment that is expected to dominate HPC architectures over the next few years. The report also includes a section with NERSC responses to the workshop findings. NERSC has many initiatives already underway that address key workshop findings and all of the action items are aligned with NERSC strategic plans.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Large Scale Computing and Storage Requirements for Basic Energy Sciences: Target 2017. Report of the NERSC Requirements Review Conducted October 8–9, 2013, Gathersburg, MD

The National Energy Research Scientific Computing Center (NERSC) is the primary computing center for the DOE Office of Science, serving approximately 5,000 users working on some 700 projects that involve nearly 600 codes in a wide variety of scientific disciplines. In addition to large-scale computing and storage resources NERSC provides support and expertise that help scientists make efficient use of its systems. The latest review revealed several key requirements, in addition to achieving its goal of characterizing BES computing and storage needs. high-level findings are: 1) Scientists will need access to significantly more computational and storage resources to achieve their goalds and reach BES research objectives; 2) Users will need assistance from NERSC to prepare for Cori (NERSC-8) and follow-on manycore systems; 3) Research teams need to run complex jobs of many different types and scales; 4) BES is a leader in innovative use of HPC and requires a diverse set of resources and services from NERSC; and 5) BES facilities need computational analysis and data storage resources beyond what they can provide. This report expands upon these key points and add others. The results are based upon representative samples, called "case studies," of the needs of science teams within BES. The case study topics were selected by the NERSC meeting coordinators and BES program managers to represent the BES production computing workload. Prepared by BES workshop participants, the case studies contain a summary of science goals, methods of solution, current and future computing requirements, and special software and support needs. Also included are strategies for computing in the highly parallel "many-core" environment that is expected to dominate HPC architectures over the next few years.

97 MATHEMATICS AND COMPUTING↗

A guide to the design of the virtual element methods for second- and fourth-order partial differential equations

Here we discuss the design and implementation details of two conforming virtual element methods for the numerical approximation of two partial differential equations that emerge in phase-field modeling of fracture propagation in elastic material. The two partial differential equations are: (i) a linear hyperbolic equation describing the momentum balance and (ii) a fourth-order elliptic equation modeling the damage of the material. Inspired by, we develop a new conforming VEM for the discretization of the two equations, which is implementation-friendly, i.e., different terms can be implemented by exploiting a single projection operator. We use C 0 and C 1 virtual elements for the second-and fourth-order partial differential equation, respectively. For both equations, we review the formulation of the virtual element approximation and discuss the details pertaining the implementation.

42 ENGINEERING↗

Feynman formulae and phase space Feynman path integrals for tau-quantization of some Lévy-Khintchine type Hamilton functions

Evolution semigroups generated by pseudo-differential operators are considered. These operators are obtained by different (parameterized by a number τ) procedures of quantization from a certain class of functions (or symbols) defined on the phase space. This class contains Hamilton functions of particles with variable mass in magnetic and potential fields and more general symbols given by the Lévy-Khintchine formula. The considered semigroups are represented as limits of n-fold iterated integrals when n tends to infinity. Such representations are called Feynman formulae. Some of these representations are constructed with the help of another pseudo-differential operator, obtained by the same procedure of quantization; such representations are called Hamiltonian Feynman formulae. Some representations are based on integral operators with elementary kernels; these are called Lagrangian Feynman formulae. Langrangian Feynman formulae provide approximations of evolution semigroups, suitable for direct computations and numerical modeling of the corresponding dynamics. Hamiltonian Feynman formulae allow to represent the considered semigroups by means of Feynman path integrals. In the article, a family of phase space Feynman pseudomeasures corresponding to different procedures of quantization is introduced. The considered evolution semigroups are represented as phase space Feynman path integrals with respect to these Feynman pseudomeasures, i.e., different quantizations correspond to Feynman path integrals with the same integrand but with respect to different pseudomeasures. This answers Berezin’s problem of distinguishing a procedure of quantization on the language of Feynman path integrals. Moreover, the obtained Lagrangian Feynman formulae allow also to calculate these phase space Feynman path integrals and to connect them with some functional integrals with respect to probability measures.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Reduced-Cost Four-Component Relativistic Double Ionization Potential Equation-of-Motion Coupled-Cluster Approaches with 4-Hole–2-Particle Excitations and Three-Body Clusters

The double ionization potential (DIP) equation-ofmotion (EOM) coupled-cluster (CC) method with 4-hole−2- particle (4h-2p) excitations on top of the CC with singles, doubles, and triples calculation, abbreviated as DIP-EOMCCSDT(4h-2p), along with its perturbative DIP-EOMCCSD(T)(a)(4h-2p) approximation, are extended to a relativistic four-component (4c) framework. In addition, we introduce and test a new computationally practical DIP-EOMCC approach, which we call DIPEOMCCSD( T)(ã)(4h-2p), that approximates the treatment of 4h- 2p correlations within the DIP-EOMCCSD(T)(a)(4h-2p) method and reduces the $\mathcal{N}$ 8 scaling characterizing DIP-EOMCCSDT(4h- 2p) and DIP-EOMCCSD(T)(a)(4h-2p) to $\mathcal{N}$ 7 with the system size $\mathcal{N}$. Further improvements in computational efficiency are obtained using the frozen natural spinor (FNS) approximation to reduce the numbers of unoccupied spinors entering the correlated steps of the DIP-EOMCC calculations according to a well-defined occupation-number-based threshold. The resulting 4c-FNS-DIPEOMCC approaches are used to compute DIPs for the series of inert gas atoms from argon to radon as well as the vertical DIPs in Cl 2 , Br 2 , HBr, and HI, which have been experimentally examined in the past. We demonstrate that, when using complete basis set extrapolations and FNS truncation threshold of 10 −4.5 , the 4c-FNS-DIP-EOMCCSD(T)(ã)(4h-2p) calculations are capable of predicting DIPs in agreement with experimental data, improving upon their nonrelativistic and spin-free scalar-relativistic counterparts, particularly when examining DIPs characterized by stronger spin−orbit coupling effects.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Fast Reactor Methods Development Plan

Fast reactor design and safety analysis methods (computer codes) have been under development for approximately sixty years. During the first forty years, code development was driven by the immediate needs of operational facilities such as Experimental Breeder Reactors I and II and the Fast Flux Test Facility. Development was further driven by programmatic needs to support advanced reactor programs such as the Clinch River Breeder Reactor Project and the Advanced Liquid Metal Reactor Program. By the early 1990s, design and safety analysis codes had been well established to addressed the unique characteristics of fast reactors. During the subsequent twenty years, however, development was nearly stagnant as a result of U.S. Government policy changes that abruptly canceled nearly all of the fast reactor research in the United States. As a result, the state of fast reactor design and safety analysis methods has deteriorated.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Excitonic couplings between molecular crystal pairs by a multistate approximation

In this paper, we present a diabatization scheme to compute the excitonic couplings between an arbitrary number of states in molecular pairs. The method is based on an algebraic procedure to find the diabatic states with a desired property as close as possible to that of some reference states. In common with other diabatization schemes, this method captures the physics of the important short-range contributions (exchange, overlap, and charge-transfer mediated terms) but it becomes particularly suitable in presence of more than two states of interest. The method is formulated to be usable with any level of electronic structure calculations and to diabatize different types of states by selecting different molecular properties. These features make the diabatization scheme presented here especially appropriate in the context of organic crystals, where several excitons localized on the same molecular pair may be found close in energy. In this paper, the method is validated on the tetracene crystal dimer, a well characterized case where the charge transfer (CT) states are closer in energy to the Frenkel excitons (FE). The test system was studied as a function of an external electric field (to explore the effect of changing the relative energy of the CT excited state) and as a function of different intermolecular distances (to probe the strength of the coupling between FE and CT states). Additionally, we illustrate how the approximation can be used to include the environment polarization effect.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Mechanical properties of zeolite-templated carbons from approximate density functional theory calculations

Zeolite-templated carbon (ZTC) is a unique porous carbonaceous material whose structure is ordered at the nanometre scale, enabling a representative periodic description at the atomistic level. Utilizing an existing, well-defined reference model for ZTCs, a structural library of varying compositions was developed by refinement using density-functional tight-binding (DFTB) potentials parameterized for materials science applications. We first determined the quantum chemical-refined structures of models with CH, CHO, CHON, CHOB, and CHOBN compositions with various degrees of heteroatom substitution. These structural models comprise the characteristic morphological features of highly porous carbon materials, such as open-blade surfaces, edges, saddles, and closed-strut formations, spanning a range of curvatures and characteristic sizes. Second, we carried out alternating compression and expansion of the CHO model unit cell to determine the lowest energy structure as well as to obtain its bulk modulus in order to demonstrate a close connection between macroscopic observations and atomic-scale structures. Further, the agreement between experimental measurements and the computational model is remarkable and demonstrates the power of approximate density functional theory as a cost-effective computational tool with chemical accuracy for the investigation of structure/property relationships in real-world carbon-based solids.

03 NATURAL GAS↗

TU-AB-BRC-02: Accuracy Evaluation of GPU-Based OpenCL Carbon Monte Carlo Package (goCMC) in Biological Dose and Microdosimetry in Comparison to FLUKA Simulations

Purpose: One of the most accurate methods for radiation transport is Monte Carlo (MC) simulation. Long computation time prevents its wide applications in clinic. We have recently developed a fast MC code for carbon ion therapy called GPU-based OpenCL Carbon Monte Carlo (goCMC) and its accuracy in physical dose has been established. Since radiobiology is an indispensible aspect of carbon ion therapy, this study evaluates accuracy of goCMC in biological dose and microdosimetry by benchmarking it with FLUKA. Methods: We performed simulations of a carbon pencil beam with 150, 300 and 450 MeV/u in a homogeneous water phantom using goCMC and FLUKA. Dose and energy spectra for primary and secondary ions on the central beam axis were recorded. Repair-misrepair-fixation model was employed to calculate Relative Biological Effectiveness (RBE). Monte Carlo Damage Simulation (MCDS) tool was used to calculate microdosimetry parameters. Results: Physical dose differences on the central axis were <1.6% of the maximum value. Before the Bragg peak, differences in RBE and RBE-weighted dose were <2% and <1%. At the Bragg peak, the differences were 12.5% caused by small range discrepancy and sensitivity of RBE to beam spectra. Consequently, RBE-weighted dose difference was 11%. Beyond the peak, RBE differences were <20% and primarily caused by differences in the Helium-4 spectrum. However, the RBE-weighted dose agreed within 1% due to the low physical dose. Differences in microdosimetric quantities were small except at the Bragg peak. The simulation time per source particle with FLUKA was 0.08 sec, while goCMC was approximately 1000 times faster. Conclusion: Physical doses computed by FLUKA and goCMC were in good agreement. Although relatively large RBE differences were observed at and beyond the Bragg peak, the RBE-weighted dose differences were considered to be acceptable.

60 APPLIED LIFE SCIENCES↗

Reliable algorithms for power system analysis in the presence of data uncertainties

A robust and reliable power flow analysis represents an essential requirement for many power systems applications as far as network optimization, voltage control, state estimation, and service restoration are concerned. The most common power flow approach, referred to here as a deterministic power flow (PLF), requires precise or 'crisp' values chosen by the analyst for each input variable. The solution provides precise network voltages and flows through each line. The specified values rest upon assumptions about the operating condition derived from historical measurements or predictions about future conditions and thus, cannot be considered accurate. Even in the case where the inputs are based on measurements, inaccuracies arise from time-skew problems, three-phase unbalance, static modeling approximations of dynamic components (e.g., transformer tap changers), variations in line parameters, and so on. The advent of deregulation and competitive power markets will only exacerbate this problem as well-known generation patterns change, loading becomes less predictable and the transmission paths grow more diverse. Conventional methodologies proposed in literature address tolerance analysis of power flow solution by means of detailed probabilistic methods, accounting for the variability and stochastic nature of the input data, and sampling based approaches. In particular uncertainty propagation using sampling based methods, such as the Monte Carlo, requires several model runs that sample various combinations of input values. Since the number of model runs can sometimes be very large, the required computer resources can sometimes be prohibitively expensive resulting in substantial computational demands. As far as probabilistic methods are concerned, they represent a useful tool, especially for planning studies, but, as evidenced by the many discussions reported in literature, they could reveal some shortcomings principally arising from: (1) the non-normal distribution and the statistical dependence of the input data; and (2) the difficulty arising in accurately identifying probability distributions for some input data, such as the power generated by wind or photovoltaic generators. All these could result in time consuming computations with several limitations in practical applications especially in power flow analysis of complex power networks. In order to try and overcome some of these limitations, obtaining thereby comprehensive power flow solution tolerance analysis at adequate computational costs, self validated computation could play a crucial role. Armed with such a vision, this chapter will analyze two advanced techniques for power flow analysis in the presence of data uncertainty namely the boundary power flow and the affine arithmetic power flow.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Approximate quantum circuit synthesis using block encodings

One of the challenges in quantum computing is the synthesis of unitary operators into quantum circuits with polylogarithmic gate complexity. Exact synthesis of generic unitaries requires an exponential number of gates in general. We propose a novel approximate quantum circuit synthesis technique by relaxing the unitary constraints and interchanging them for ancilla qubits via block encodings. This approach combines smaller block encodings, which are easier to synthesize, into quantum circuits for larger operators. Due to the use of block encodings, our technique is not limited to unitary operators and can be applied for the synthesis of arbitrary operators. Overall, we show that operators which can be approximated by a canonical polyadic expression with a polylogarithmic number of terms can be synthesized with polylogarithmic gate complexity with respect to the matrix dimension.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Mixed Convection in the VHTR in the Event of a LOFA

The US Department of Energy, Office of Nuclear Energy (DOE-NE) is supporting the development of a very high temperature reactor (VHTR) concept as the primary focus of it next generation nuclear power plant (NGNP) program. The VHTR is cooled by forcing helium downwards through the core into the lower plenum and out the hot duct. In the event that the coolant circulators are lost, the driving pressure drop across the core will reduce to zero and there will be the opportunity for natural circulation to occur. During the time that the circulators are powering down, the heat transfer in the core from the graphite blocks to the helium coolant will transform from turbulent forced convection to mixed convection, where buoyancy effects become important, to free or natural convection, where buoyancy is dominant. Analysis of the nature of the forced, mixed and free convection is best done using computational fluid dynamic (CFD) software that can provide fine details of the flow and heat transfer. However, CFD analysis involves approximations in the results because of the finite nature of the spatial and temporal discretizations required, the inexact nature of the turbulence models that are used and the finite precision of the computers employed. Therefore, it is necessary to validate the CFD computations. Validation is accomplished by comparing results from specific CFD computations to experimental data that have been taken specifically for the purpose of validation and that are related to the physical phenomena in question. The present report examines the flow and heat transfer parameters (dimensionless numbers) that characterize the flow and reports ranges for their values based on specific CFD studies performed for the VHTR.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Fast increased fidelity samplers for approximate Bayesian Gaussian process regression

Gaussian processes (GPs) are common components in Bayesian non-parametric models having a rich methodological literature and strong theoretical grounding. The use of exact GPs in Bayesian models is limited to problems containing several thousand observations due to their prohibitive computational demands. We develop a posterior sampling algorithm using H-matrix approximations that scales at O(n log 2 n). We show that this approximation’s Kullback-Leibler divergence to the true posterior can be made arbitrarily small. Though multidimensional GPs could be used with our algorithm, d-dimensional surfaces are modeled as tensor products of univariate GPs to minimize the cost of matrix construction and maximize computational efficiency. We illustrate the performance of this fast increased fidelity approximate GP, FIFA-GP, using both simulated and non-synthetic data sets

97 MATHEMATICS AND COMPUTING↗

Optimal limits on primordial magnetic fields from CMB temperature bispectrum of passive modes

We investigate bounds on the strength of the primordial magnetic field (PMF) from the cosmic microwave background (CMB) bispectra of the intensity (temperature) modes induced from the auto- and cross-correlated bispectra of the scalar and tensor components of the PMF anisotropic stress. At first, we construct a general formula for the CMB intensity and polarization bispectra from PMFs composed of any type of perturbation. Then we derive an approximate expression which traces the exact shape of the CMB bispectrum in order to reduce the computation time with respect to a large number of the multipole configurations, and also show that the non-Gaussian structure coming from PMFs is classified as the local-type configuration. Computing the signal-to-noise ratio on the basis of the approximate formula with the information of the instrumental noises and resolutions, we find expected upper bounds on the magnetic field strength, when the magnetic spectrum is nearly scale invariant (n{sub B} = −2.9), smoothed on 1Mpc scale at 95% confidence level from the WMAP and PLANCK experiments as B{sub 1Mpc} < 4.0−6.7nG and 3.8−6.5nG, respectively, depending on the energy scale of the magnetic field production from 10{sup 14}GeV to 10{sup 3}GeV. Our new consequences imply slight overestimations by the previous rough discussions.

79 ASTRONOMY AND ASTROPHYSICS↗

Simple and efficient relaxation methods for interfaces separating compressible fluids, cavitating flows and shocks in multiphase mixtures

Numerical approximation of the five-equation two-phase flow of Kapila et al. [A.K. Kapila, R. Menikoff, J.B. Bdzil, S.F. Son, D.S. Stewart, Two-phase modeling of deflagration-to-detonation transition in granular materials: reduced equations, Physics of Fluids 13(10) (2001) 3002-3024] is examined. This model has shown excellent capabilities for the numerical resolution of interfaces separating compressible fluids as well as wave propagation in compressible mixtures [A. Murrone, H. Guillard, A five equation reduced model for compressible two phase flow problems, Journal of Computational Physics 202(2) (2005) 664-698; R. Abgrall, V. Perrier, Asymptotic expansion of a multiscale numerical scheme for compressible multiphase flows, SIAM Journal of Multiscale and Modeling and Simulation (5) (2006) 84-115; F. Petitpas, E. Franquet, R. Saurel, O. Le Metayer, A relaxation-projection method for compressible flows. Part II. The artificial heat exchange for multiphase shocks, Journal of Computational Physics 225(2) (2007) 2214-2248]. However, its numerical approximation poses some serious difficulties. Among them, the non-monotonic behavior of the sound speed causes inaccuracies in wave's transmission across interfaces. Moreover, volume fraction variation across acoustic waves results in difficulties for the Riemann problem resolution, and in particular for the derivation of approximate solvers. Volume fraction positivity in the presence of shocks or strong expansion waves is another issue resulting in lack of robustness. To circumvent these difficulties, the pressure equilibrium assumption is relaxed and a pressure non-equilibrium model is developed. It results in a single velocity, non-conservative hyperbolic model with two energy equations involving relaxation terms. It fulfills the equation of state and energy conservation on both sides of interfaces and guarantees correct transmission of shocks across them. This formulation considerably simplifies numerical resolution. Following a strategy developed previously for another flow model [R. Saurel, R. Abgrall, A multiphase Godunov method for multifluid and multiphase flows, Journal of Computational Physics 150 (1999) 425-467], the hyperbolic part is first solved without relaxation terms with a simple, fast and robust algorithm, valid for unstructured meshes. Second, stiff relaxation terms are solved with a Newton method that also guarantees positivity and robustness. The algorithm and model are compared to exact solutions of the Euler equations as well as solutions of the five-equation model under extreme flow conditions, for interface computation and cavitating flows involving dynamics appearance of interfaces. In order to deal with correct dynamic of shock waves propagating through multiphase mixtures, the artificial heat exchange method of Petitpas et al. [F. Petitpas, E. Franquet, R. Saurel, O. Le Metayer, A relaxation-projection method for compressible flows. Part II. The artificial heat exchange for multiphase shocks, Journal of Computational Physics 225(2) (2007) 2214-2248] is adapted to the present formulation.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NEW DEVELOPMENTS ON INVERSE POLYGON MAPPING TO CALCULATE GRAVITATIONAL LENSING MAGNIFICATION MAPS: OPTIMIZED COMPUTATIONS

We derive an exact solution (in the form of a series expansion) to compute gravitational lensing magnification maps. It is based on the backward gravitational lens mapping of a partition of the image plane in polygonal cells (inverse polygon mapping, IPM), not including critical points (except perhaps at the cell boundaries). The zeroth-order term of the series expansion leads to the method described by Mediavilla et al. The first-order term is used to study the error induced by the truncation of the series at zeroth order, explaining the high accuracy of the IPM even at this low order of approximation. Interpreting the Inverse Ray Shooting (IRS) method in terms of IPM, we explain the previously reported N {sup -3/4} dependence of the IRS error with the number of collected rays per pixel. Cells intersected by critical curves (critical cells) transform to non-simply connected regions with topological pathologies like auto-overlapping or non-preservation of the boundary under the transformation. To define a non-critical partition, we use a linear approximation of the critical curve to divide each critical cell into two non-critical subcells. The optimal choice of the cell size depends basically on the curvature of the critical curves. For typical applications in which the pixel of the magnification map is a small fraction of the Einstein radius, a one-to-one relationship between the cell and pixel sizes in the absence of lensing guarantees both the consistence of the method and a very high accuracy. This prescription is simple but very conservative. We show that substantially larger cells can be used to obtain magnification maps with huge savings in computation time.

97 MATHEMATICS AND COMPUTING↗