Perfect Loop Research Articles

Research on uncovering parallelism in perfect loop nests for GCC compiler

Abstract To improve the ability to uncover parallelism in perfect loop nests for the GCC compiler, a method that utilizes the lambda transformation technique to uncover potential parallelism is proposed. Based on data references, dependence distance vectors and dependence direction vectors generated during the GIMPLE phase of the GCC compilation process, the GIMPLE form of perfect loop nests is converted into lambda representation in this paper, and the proposed method implements legal transformations on lambda representation for uncovering parallel loops in arbitrary loop nests. This method is integrated into the dependency analysis pass of the GCC compiler and has been experimentally validated on integer and floating-point benchmarks from the SPEC CPU2006 benchmarks. The results demonstrate that it can effectively enhance the existing loop analysis capabilities, loop transformation abilities, and loop parallelism uncovering capabilities of GCC. This method provides a reference for research on automatic parallelization techniques and holds practical application value within the GCC compiler.

Quadrotor Trajectory Tracking Using Model Reference Adaptive Control, Neural Network-Based Parameter Uncertainty Compensator, and Different Plant Parameterizations

A quadrotor trajectory tracking problem is addressed via the design of a model reference adaptive control (MRAC) system. As for real-world applications, the entire quadrotor dynamics is typically unknown. To take that into account, we consider a plant model, which contains uncertain nonlinear terms resulting from aerodynamic friction, blade flapping, and the fact that the mass and inertia moments of the quadrotor may change from their nominal values. Unlike many known studies, the explicit equations of the parameter uncertainty for the position control loop are derived in two different ways using the differential flatness approach: the control signals are (i) used and (ii) not used in the parametric uncertainty parameterization. After analysis, the neural network (NN) is chosen for both cases as a compensator of such uncertainty, and the set of NN input signals is justified for each of them. Unlike many known MRAC systems with NN for quadrotors, in this study, we use the kxx+krr baseline controller, which follows from the control system derivation, with both time-invariant (parameterization (i)) and adjustable (parameterization (ii)) parameters instead of an arbitrarily chosen non-tunable PI/PD/PID-like one. Adaptive laws are derived to adjust the parameters of NN uncertainty compensator for both parameterizations. As a result, the position controller ensures the asymptotic stability of the tracking error for both cases under the assumption of perfect attitude loop tracking, which is ensured in the system previously developed by the authors. The results of the numerical experiments support the theoretical conclusions and provide a comparison of the effectiveness of the derived parameterizations. They also allow us to make conclusions on the necessity of the baseline controller adjustment.

In-situ TEM investigation on the evolution of He bubbles and dislocation loops in Ni-based alloy irradiated by H2+ & He+ dual-beam ions

• Synergistic evolution mechanism of He bubbles and loops in Ni-based alloy irradiated by dual-beam ions were investigated via in-situ TEM. • Several evolution stages of the loop-punching mechanism and its effect on evolution of neighboring bubbles and loops were revealed. • The details of the unfaulting processes of the Frank loops and the formation mechanism of rhombic loops were discussed. • Four Frank loop variations were identified, and their different evolution behaviors were studied. The synergistic evolution mechanisms of He bubbles and dislocation loops under 30 keV H 2 + & H e + dual-beam ions irradiation at 650°C in the Ni-based alloy GH3535, which is the most promising candidate structure material for molten salt reactors (MSRs), were revealed via in-situ TEM. The nucleation, merging, and change in the size of the dislocation loops and He bubbles were characterized in detail to study the influences of irradiation fluence and pre-existing dislocation loops on their evolutions. The number density of both the He bubbles and dislocation loops increases rapidly and subsequently saturates, whereas their size continuously increases with the increasing ion fluence. Pre-existing dislocation loops with strong absorption characteristics grow preferentially and suppress the nucleation of dislocation loops during the dual-beam ions irradiation. Moreover, the bubbles tend to nucleate within the dislocation loops to form bubble-loop complexes, and then decrease in their number density. The details of the unfaulting processes of the Frank loops were discussed, where the energy difference between the two types of loops as well as the evolution of the inside Shockley dislocation loops dominates the unfaulting behavior. The several evolution stages of the loop-punching mechanism are revealed, and the emitted loops can directly form perfect loops as well as unfault the neighboring Frank loops. The He bubbles inside the loops provide corresponding stress for the formation of rhombic loops, which can achieve rapid growth and sweep ability by merging with the neighboring loops. Additionally, its dissociation to Shockley dislocation can unfault the Frank loops along their slip direction. Among the four Frank loop variations, the edge-on Frank loop variations have the highest growth rate, followed by the perfect loop. The related mechanisms based on in-situ experimental observation are discussed in depth. Synergistic evolution mechanism of He bubbles and dislocation loops, including loop-punching mechanism and bubble-loop complexes, in Ni-based alloy irradiated by dual-beam ions were investigated via in-situ TEM.

Impact of micro-alloying in ion-irradiated nickel: From the inhibition of point-defect cluster diffusion by thermal segregation to the change of dislocation loop nature

Micro-alloying strongly affects the incubation period of void swelling in irradiated face-centered cubic materials. However, the underlying mechanism, which relates to the formation of dislocation loops, is still unclear. Here, we investigate pure Ni, Ni-0.4wt.%Cr and Ni-0.4/0.8/1.2wt.%Ti as model materials, to gain insight into the solute effects on the loops evolution in the early stage of irradiation. The dislocation loop characteristics (mobility, Burgers vector, nature) are studied using in-situ transmission electron microscopy and ex-situ irradiation with Ni+ ions at 450°C and 510°C for doses from 0.06 to 0.7 dpa. It appears that a tiny amount of Ti effectively increases the loop density, reduces the loop mobility and the stacking fault energy. It leads to an equal distribution among a/2<110> perfect loop families. It also stabilizes self-interstitial loops against vacancy loops depending on Ti content and temperature. Our modeling of radiation-induced segregation, based on experiments and recent ab initio calculations of flux couplings, predicts a Cr enrichment and a Ti depletion nearby dislocation loops. It is in good agreement with our observations by X-ray spectroscopy in TEM and by atom probe tomography. However, the lowered loop mobility must be the signature of a thermal segregation rather than the impact of radiation-induced depletion. Indeed, oversized Ti atoms subsequently trapped at strained lattice sites around the dislocation line of the loop due to thermal segregation would inhibit its diffusion. This opens new perspectives for future experimental investigations and radiation-effect modeling.

Arc-descent for the perfect loop functor andp-adic Deligne–Lusztig spaces

AbstractWe prove that the perfect loop functorLXof a quasi-projective schemeXover a local non-archimedean fieldksatisfies arc-descent, strengthening a result of Drinfeld. Then we prove that for an unramified reductive groupG, the mapL⁢G→L⁢(G/B){LG\rightarrow L(G/B)}is av-surjection. This gives a mixed characteristic version (forv-topology) of an equal characteristic result (in étale topology) of Bouthier–Česnavičius. In the second part of the article, we use the above results to introduce a well-behaved notion of Deligne–Lusztig spacesXw⁢(b){X_{w}(b)}attached to unramifiedp-adic reductive groups. We show that in various cases these sheaves are ind-representable, thus partially solving a question of Boyarchenko. Finally, we show that the natural covering spacesX˙w˙⁢(b){\dot{X}_{\dot{w}}(b)}are pro-étale torsors over clopen subsets ofXw⁢(b){X_{w}(b)}, and analyze some examples.

The evolution and interaction of irradiation defects in CeO2 during H2+ & He+ dual-beam synergistic irradiation investigated by in-situ TEM

As a surrogate material of UO2, non-radioactive CeO2 was investigated by in-situ transmission electron microscopy during 30 keV H2+ & He+ dual-beam synergistic irradiation at 1073 K. The relationships between the size/density of dislocation loops and irradiation fluence were established, as well as for gas bubbles. Based on the analysis of the Burgers vectors, the formation of perfect dislocation loop (PDL) through the reaction between PDLs and the direct transformation from Frank dislocation loop (FDL) to PDL were obtained. The shrinkage of FDLs at high dose was observed. It was not only found that the number density and size of bubbles near the grain boundary were lower than those in the crystal, but also found that no loop-bubbles complex was formed. The mechanism of irradiation-induced excess oxygen vacancies should be the reason for the shrinkage of dislocation loops at high dose and the depletion of bubbles in the dislocation loops.

Analysis, Modeling, and Simulation of Adaptive Control Based on Dynamic Conductance Estimation of Photovoltaic Generator Interfaced Current-Mode Buck Converter

In this paper, a novel adaptive control method is presented, aimed at robustifying the terminal voltage of the photovoltaic generator, interfaced by the current-mode-controlled buck DC-DC converter load, and based on conductance estimation. The photovoltaic generator, which is integrated into the buck converter and a battery storage unit, is continuously affected by the operating point of the system and environmental variables, thus presenting a nonlinear behavior. Furthermore, the development of small-signal equations reveals a potentially unstable condition when the system is used as a micro-grid with a battery storage unit. This study shows that when the nonlinear behavior of the photovoltaic generator is combined with a typical nominal controller, designed for a single nominal operating point and due to the possibility of an unstable condition, it forces the controller to operate mostly outside the nominal operating point. These conditions result in significantly varied closed-loop performance. In contrast, an almost perfect loop gain performance can be achieved when implementing an adaptive controller based on an online conductance estimation method. Applying estimator results and injecting its value in real-time into the inverse-based plant controller results in an adaptive controller. Therefore, the closed-loop performance of the system integrated with an adaptive controller achieves an almost nominal response throughout the operating range.

Transfer of dislocation slip through grain boundaries in metal-graphene nanocomposites

A model is suggested that describes the onset of plastic deformation in metal-graphene nanocomposites. Within the model, the plastic deformation of a deformed metal-graphene nanocomposite is realized through the generation of perfect or partial dislocation loops. These loops nucleate at grain boundaries (GBs) of the metallic matrix and expand across grains. The expanded dislocation loops promote the nucleation and expansion of secondary dislocation loops in adjacent grains, thereby realizing dislocation transmission across GBs. Within the model, we have calculated the critical stresses, τc1 and τc2, for the generation of a secondary perfect or partial dislocation loop and for the bypass of the graphene inclusion by the secondary loop. The critical stress τc2 increases with an increase in the graphene concentration and with a decrease in the thickness of the inclusion. This implies that the addition of graphene can raise the critical stress for the expansion of the secondary dislocation loops and thus increase the yield strength of metals.

In-situ TEM investigation of unfaulting behavior of Frank loops in FCC Pd during H2+ & He+dual-beam irradiation

A faulted Frank dislocation loop (FDL) can transform into an unfaulted perfect dislocation loop (PDL) under plastic deformation, quenching or particle irradiation. Here, we report our direct observation of the unfaulting processes of FDLs in palladium (Pd) by in-situ ion irradiation in transmission electron microscopy. Results indicate that the FDL unfaulting is a combination of three processes involving the interaction between adjacent FDLs, the interaction between FDL and PDL, and the nucleation and growth of Shockley partial dislocations within FDL. Meanwhile, bubbles tend to nucleate within dislocation loops, and the PDLs capture more He/H2 over FDLs, which confirms for the first time that the irradiation-induced loops play an important role in the aging of Pd and the unfaulting strengthens this process.

Molecular dynamic simulations evaluating the effect of the stacking fault energy on defect formations in face-centered cubic metals subjected to high-energy particle irradiation

Austenitic stainless steels, which are used as incore structural materials in light water reactors, are characterized by an extremely low stacking fault energy (SFE) among face-centered cubic (FCC) metals. To evaluate the effects of SFE on defect formation under high-energy particle irradiation, molecular dynamics simulations were performed using the interatomic potential sets for FCC metals with different SFEs and a primary knock-on atom energy (EPKA) of 100 keV at 600 K. The results show that the number of residual defects is independent of the SFE. However, the characteristics of self-interstitial atom (SIA) clusters do depend on the SFE. For clusters smaller than a certain size, the ratio of glissile SIA clusters decreases as the SFE increases, which is similar to the trend observed at the low EPKA. However, for larger clusters, which can be detected only at a high EPKA, the ratio of glissile clusters increases. These results correspond to static energy calculations, in which the difference in the formation energy between a Frank loop and perfect loop (ΔEF-P) for the small clusters decreases as the SFE increases. In contrast, for the larger clusters, the SFE dependence of ΔEF-P changes due to the shape restrictions of stable perfect loops. At a high temperature of 600 K, large vacancy clusters with stacking faults can be detected at EPKA = 100 keV, resulting in the enhanced formation of these clusters at lower SFEs. Furthermore, several of these clusters were similar to perfect loops, with the edges split into two partial dislocations with stacking faults, although the largest clusters detected at low EPKAs were similar to stacking fault tetrahedrons.

In Case You Haven't Heard…

You have heard all the stories about cannabis and psychosis — mostly in the “which came first” category. Now a study has linked them in a perfect loop, finding that a genetic predisposition to schizophrenia may increase the risk of psychosis from cannabis use. The study, published in Translational Psychiatry by researchers from the Centre for Addiction and Mental Health (CAMH) and King's College London, found that while cannabis users had higher rates of psychotic experiences than non‐users, the difference was pronounced among those with a genetic predisposition to schizophrenia. “These results are significant because they're the first evidence we've seen that people genetically prone to psychosis might be disproportionately affected by cannabis,” said lead author Dr. Michael Wainberg, a scientist at the Krembil Centre for Neuroinformatics at CAMH. “And because genetic risk scoring is still in its early days, the true influence of genetics on the cannabis‐psychosis relationship may be even greater than what we found here.” Overall, people who had used cannabis were 50% more likely to report psychotic experiences than people who had not. However, among the fifth of participants with the highest genetic risk scores for schizophrenia, it was 60%, and among the fifth with the lowest scores, it was only 40%.

Eco-friendly green synthesis and characterizations of CoFe2-x AlxO4 nanocrystals: analysis of structural, magnetic, electrical, and dielectric properties

This work reports, thermal, structural, magnetic, electric, and dielectric investigations of the CoFe2-xAlxO4 nanocrystals synthesized through eco-friendly green sol–gel auto-ignition route. X-ray diffractograms exposed the uni-phasic geometry with Fd3m_ $${\text{O}}_{{\text{h}}}^{7}$$ space group. The standard crystallite sizes were found in between 24 and 44 nm, confirming the nano-crystalline dimension. FTIR analysis confirmed the two specific vibrational stretching bands corresponding to the spinel ferrite geometry. The surface topology of ferrite nanocrystals with size distribution was effectively observed from TEM micrographs. The magnetic study at room temperature showed lessening in saturation magnetization with rise in Al3+ concentrations, ascribed to the replacement of Fe3+ ions by paramagnetic Al3+ ions. The M–H loops showed the perfect loop of typical soft ferromagnetic behavior of the produced nano-ferrites. The saturation magnetization, coercivity, and magneto-crystalline anisotropy declines with the upsurge in Al3+ ions which is owing to the non-magnetic contamination in the magnetic matrix. The DC-electrical resistivity was enhanced with growing temperature along with Al3+ content x. The dielectric parameters boosted pointedly with Al3+ substitution and decline exponentially with a rise in frequency in compliance with ‘Koops’ phenomenological theory. All the outcomes show that Al3+ substitution influences the physical characteristics of cobalt ferrite nanocrystals which shows its suitability in various technological applications. The graphical abstract presenting the synthesis reaction, TEM micrograph (along with crystallite size distribution) and M–H plots for CoFe2-xAlxO4 nanocrystals.

Neutron irradiation-induced microstructure damage in ultra-high temperature ceramic TiC

Titanium carbide (TiC) is an ultra-high temperature ceramic with potential as a structural material candidate for advanced reactor concepts. However, the irradiation tolerance of TiC is not well understood. Here, we reveal the key irradiation damage microstructure degradation processes in TiC using mixed spectrum neutron irradiations at dose of ∼2 displacements per atom (dpa) at temperatures of ∼220, 620, and 1115 °C, combined with state-of-art microstructure characterization using transmission electron microscopy (TEM) and high-resolution TEM (HRTEM). The helium injection (∼65 atomic parts per million) produced by neutron transmutation also occurred in the sample. TiC was observed to form irradiation-induced interstitial-type dislocation loops and He-stabilized cavities. At 220 and 620 °C, the analysis of the electron diffraction patterns, rel-rod imaging and HRTEM revealed that the dislocation loops were faulted Frank loops with Burgers vector bFrank = 1/3<111> lying on {111} planes. A detailed Burgers vector identification performed by the g.b technique revealed that the dislocation loops forming at 1115 °C were unfaulted, edge-type, with Burgers vectors b = a<100> and a/2<110> with corresponding {100} and {110} habit planes. Using continuum mechanics, we estimated the critical radius at which a faulted dislocation loop transitions to a perfect loop to be 9 nm. Further, no amorphization occurred in TiC under-investigated irradiation conditions while macroscopic swelling under point defect swelling regime was observed. He-stabilized cavities were detected at 1115 °C, but not at lower temperature irradiation. These observations indicate the onset of long-range vacancy migration in TiC occurs between 620 and 1115 °C.

Bismuth-Doped Nickel Ferrite Nanoparticles for Room Temperature Memory Devices

Searching for materials and creating promising functionalities are the main driving force behind most of the studies in the area of materials science and condensed matter physics. The development of advanced multiferroic materials by utilizing the phenomenon of magnetodielectric effect provides great prominence toward the fabrication of next-generation nonvolatile memory devices. We have observed a multiferroic property at room temperature for the first time in sol–gel-derived nanocrystalline bismuth-doped nickel ferrite. Nickel ferrite, although owning traditional ferrimagnetism as observed from the magnetization study, a perfect ferroelectric polarization hysteresis loop with reasonably high polarization (10 μC/cm2), and a room temperature magnetodielectric constant of 2.8%, has been achieved at room temperature by bismuth doping. This finding is quite fascinating toward the search of multiferroic materials, which deserves further theoretical and experimental studies to pin down the exact origin of pola...

Ideal magnetic dipole resonances with metal-dielectric-metal hybridized nanodisks.

Magnetic resonances generated with nonmagnetic nanostructures have been widely used to design various functional nanophotonic devices, and it is important to realize pure magnetic dipole scattering for the unambiguous study of magnetic light-matter interactions. However, the magnetic responses often spectrally overlapping with other multipoles, which is the main obstacle to achieve ideal magnetic dipole resonances. This study proposes and theoretically demonstrates that an ideal magnetic dipole resonance can be excited with metal-dielectric-metal hybridized nanodisks. It is shown that although the generated magnetic dipole scattering around the bonding resonance of the hybridized nanodisk is spectrally overlapping with strong electric dipole and electric quadrupole contributions, an almost perfect current loop can be generated by adjusting the geometry parameters and the refractive index of the dielectric layer, thereby leading to the suppressing of the overlapping multipoles and the formation of an ideal magnetic dipole scattering. What's more important is that both electric and magnetic near-fields are enhanced simultaneously with the increasing of the refractive index of the dielectric layer, which makes the hybridized nanodisk a promising platform for enhanced magnetic light-matter interactions.

Atomistic simulations for the effects of stacking fault energy on defect formations by displacement cascades in FCC metals under Poisson’s deformation

We performed molecular dynamics simulations of displacement cascades in FCC metals under Poisson’s deformation using interatomic potentials differing in stacking fault energy (SFE), in order to investigate the effect of tensile strain on the SFE dependence of defect formation processes. There was no clear SFE dependence of the number of residual defects and the size distribution of defect clusters under both no strain and the applied strain, while the strain enhanced the defect formation to a certain extent. We also observed that the strain affected the formations of self-interstitial atom (SIA) clusters depending on their size and the Burgers vector. These results were consistent with the analysis based on the defect formation energies. Meanwhile, the number of SIA perfect loops was higher at lower SFE under both no strain and the applied strain, leading to an increase in the ratio of glissile SIA clusters with a decrease in SFE. Further, the absolute number of SIA perfect loops was increased by the applied strain, while the SFE dependence of the number of SIA perfect loops was not affected. These findings were associated with the difference in formation energy between an SIA perfect loop and an SIA Frank loop. The insights extracted from this study significantly contribute to the modeling of microstructural evolution in nuclear materials under irradiation, especially for low SFE metals such as austenitic stainless steels.

Thickness dependence of magnetic properties in submicron yttrium iron garnet films

Yttrium iron garnet (YIG) films with various thicknesses have been fabricated by liquid-phase epitaxy (LPE) on gadolinium gallium garnet (GGG) (1 1 1) substrates showing unusual magnetic properties. In this work, we studied YIG films with thicknesses between 100 nm and 1000 nm and compared them with La3+-doped YIG films (La:YIG) over a similar scale. The relaxation of tensile stress at the GGG/YIG interface, which caused a magnetic hysteresis loop, showed an easy magnetization phase for thin YIG films (⩽400 nm), while a hard magnetization phase gradually appeared with further increases to the film thickness. The stable magnetic properties of La:YIG at different thicknesses and the not so perfect magnetic hysteresis loop are caused by the intrinsic distortion of the sublattice from La3+ partially replacing Y3+. The saturation magnetization field (4πMs) obtained from broadband ferromagnetic resonance measurement of the YIG films increased with increasing Ha and stayed around the theoretical value. This work helps us to understand the properties of LPE-based YIG films and devices at the sub-microscale.

Iterative learning impedance control for rehabilitation robots driven by series elastic actuators

Existing control techniques for rehabilitation robots commonly ignore robot dynamics by assuming a perfect inner control loop or are limited to rigid-joint robots. The dynamic stability of compliantly-actuated rehabilitation robots, consisting of the dynamics of both robot and compliant actuator, is not theoretically grounded. This paper presents an iterative learning impedance controller for rehabilitation robots driven by series elastic actuators (SEAs), where the control objective is specified as a desired impedance model. The desired impedance model is achieved in an iterative manner, which suits the repeating nature of patients’ task through therapeutic process and also guarantees the transient performance of robot. The stability of the overall system is rigorously proved with Lyapunov methods by taking into account both the robot and actuator dynamics. Experimental results are presented to illustrate the performance of the proposed iterative control scheme.

Deciphering microRNAs and Their Associated Hairpin Precursors in a Non-Model Plant, Abelmoschus esculentus.

MicroRNAs (miRNAs) are crucial regulatory RNAs, originated from hairpin precursors. For the past decade, researchers have been focusing extensively on miRNA profiles in various plants. However, there have been few studies on the global profiling of precursor miRNAs (pre-miRNAs), even in model plants. Here, for the first time in a non-model plant—Abelmoschus esculentus with negligible genome information—we are reporting the global profiling to characterize the miRNAs and their associated pre-miRNAs by applying a next generation sequencing approach. Preliminarily, we performed small RNA (sRNA) sequencing with five biological replicates of leaf samples to attain 207,285,863 reads; data analysis using miRPlant revealed 128 known and 845 novel miRNA candidates. With the objective of seizing their associated hairpin precursors, we accomplished pre-miRNA sequencing to attain 83,269,844 reads. The paired end reads are merged and adaptor trimmed, and the resulting 40–241 nt (nucleotide) sequences were picked out for analysis by using perl scripts from the miRGrep tool and an in-house built shell script for Minimum Fold Energy Index (MFEI) calculation. Applying the stringent criteria of the Dicer cleavage pattern and the perfect stem loop structure, precursors for 57 known miRNAs of 15 families and 18 novel miRNAs were revealed. Quantitative Real Time (qRT) PCR was performed to determine the expression of selected miRNAs.

Multi-modal control scheme for rehabilitation robotic exoskeletons

In the past few decades, a variety of rehabilitation robotic exoskeletons have been developed for patients with stroke and traumatic brain injury, which can assist therapists and potentially improve the functional outcomes. Many robotic exoskeleton systems adopted series elastic actuators, which are known to offer a number of advantages over rigid actuators, such as high force/torque fidelity, low output impedance, intrinsic compliance, and tolerance to shocks. While several control schemes have been proposed for robotic exoskeletons driven by series elastic actuators, existing methods are limited at high level, by assuming a perfect inner control loop. Due to the nonlinearity in robot dynamics, changing interaction forces, etc., the stability of the overall system consisting of both the robot dynamics and the actuator dynamics may not be guaranteed. This paper presents a multi-modal control scheme for rehabilitation robotic exoskeletons. Three control modes are smoothly integrated into the controller, where the robot-assisted mode allows the human to exert voluntary efforts within a desired region, the robot-dominant mode is activated to correct the movement of the human when the robot is outside the region, and the safety-stop mode is to stop the robot whenever the interaction force is too large which may result in injuries. By using the regional position and force feedback, the proposed controller achieves the paradigm of “assist-as-needed” and also guarantees the safety of the human. The development of the proposed controller follows the singular perturbation approach, and the stability of the overall system is rigorously proved by using Tikhonov’s theorem. Experimental results in both upper-limb and lower-limb exoskeleton systems are presented to demonstrate the effectiveness of the proposed control method.