Search for: All records

Predicting machined surface roughness is critical for estimating a part’s performance characteristics such as susceptibility to fatigue and corrosion. Prior studies have indicated that power consumed at the tool-chip interface may represent an indicator for the surface integrity of the machining process. However, no quantita-tive association has been reported between the machining power and surface roughness due to a lack of data to develop predictive models. This paper presents a data synthesis method to address this gap. Specifically, a conditional generative adversarial network (CGAN) is developed to synthesize power signals associated with varying process parameter combinations. The quality of the synthesized signals is evaluated against experimentally measured power signals by examining the consistency in: 1) the spatial pattern of the signals induced by the cutting process as shown in the frequency domain, and 2) the temporal pattern as shown in the clustering of the synthesized and measured signals corresponding to the same parameter combination. The synthesized signals are then used to augment the measured signals and develop a convolutional neural network (CNN) for predicting the machined surface roughness. Experiments performed using H13 tool steel have shown that data augmentation by CGAN has effectively reduced the error of the surface roughness prediction from 58 %, when no synthetic data is used for CNN training, to 9.1 % when 250 synthetic samples are used. The results demonstrate the effectiveness of CGAN as a data augmentation method and CNN for mapping machining power to surface roughness. 

We propose a method to adiabatically control an atomic ensemble using a decoherence-free subspace (DFS) within a dissipative cavity. We can engineer a specific eigenstate of the system's Lindblad jump operators by injecting a field into the cavity which deconstructively interferes with the emission amplitude of the ensemble. In contrast to previous adiabatic DFS proposals, our scheme creates a DFS in the presence of collective decoherence. We therefore have the ability to engineer states that have high multi-particle entanglements which may be exploited for quantum information science or metrology. We further demonstrate a more optimized driving scheme that utilizes the knowledge of possible diabatic evolution gained from the so-called adiabatic criteria. This allows us to evolve to a desired state with exceptionally high fidelity on a time scale that does not depend on the number of atoms in the ensemble. By engineering the DFS eigenstate adiabatically, our method allows for faster state preparation than previous schemes that rely on damping into a desired state solely using dissipation. 

Cone penetration testing (CPT) is a preferred method for characterizing soil profiles for evaluating seismic liquefaction triggering potential. However, CPT has limitations in characterizing highly stratified profiles because the measured tip resistance (𝑞𝑐 ) of the cone penetrometer is influenced by the properties of the soils above and below the tip. This results in measured 𝑞𝑐 values that appear ‘‘blurred’’ at sediment layer boundaries, inhibiting our ability to characterize thinly layered strata that are potentially liquefiable. Removing this ‘‘blur’’ has been previously posed as a continuous optimization problem, but in some cases this methodology has been less efficacious than desired. Thus, we propose a new approach to determine the corrected 𝑞𝑐 values (i.e. values that would be measured in a stratum absent of thin-layer effects) from measured values. This new numerical optimization algorithm searches for soil profiles with a finite number of layers which can automatically be added or removed as needed. This algorithm is provided as open-source MATLAB software. It yields corrected 𝑞𝑐 values when applied to computer-simulated and calibration chamber CPT data. We compare two versions of the new algorithm that numerically optimize different functions, one of which uses a logarithm to refine fine-scale details, but which requires longer calculation times to yield improved corrected 𝑞𝑐 profiles. 

Abstract Background Antibiotic-producing Streptomyces bacteria are ubiquitous in nature, yet most studies of its diversity have focused on free-living strains inhabiting diverse soil environments and those in symbiotic relationship with invertebrates. Results We studied the draft genomes of 73 Streptomyces isolates sampled from the skin (wing and tail membranes) and fur surfaces of bats collected in Arizona and New Mexico. We uncovered large genomic variation and biosynthetic potential, even among closely related strains. The isolates, which were initially identified as three distinct species based on sequence variation in the 16S rRNA locus, could be distinguished as 41 different species based on genome-wide average nucleotide identity. Of the 32 biosynthetic gene cluster (BGC) classes detected, non-ribosomal peptide synthetases, siderophores, and terpenes were present in all genomes. On average, Streptomyces genomes carried 14 distinct classes of BGCs (range = 9–20). Results also revealed large inter- and intra-species variation in gene content (single nucleotide polymorphisms, accessory genes and singletons) and BGCs, further contributing to the overall genetic diversity present in bat-associated Streptomyces . Finally, we show that genome-wide recombination has partly contributed to the large genomic variation among strains of the same species. Conclusions Our study provides an initial genomic assessment of bat-associated Streptomyces that will be critical to prioritizing those strains with the greatest ability to produce novel antibiotics. It also highlights the need to recognize within-species variation as an important factor in genetic manipulation studies, diversity estimates and drug discovery efforts in Streptomyces . 

Streptomycesbacteria are known for their prolific production of secondary metabolites, many of which have been widely used in human medicine, agriculture and animal health. To guide the effective prioritization of specific biosynthetic gene clusters (BGCs) for drug development and targeting the most prolific producer strains, knowledge about phylogenetic relationships ofStreptomycesspecies, genome-wide diversity and distribution patterns of BGCs is critical. We used genomic and phylogenetic methods to elucidate the diversity of major classes of BGCs in 1,110 publicly availableStreptomycesgenomes. Genome mining ofStreptomycesreveals high diversity of BGCs and variable distribution patterns in theStreptomycesphylogeny, even among very closely related strains. The most common BGCs are non-ribosomal peptide synthetases, type 1 polyketide synthases, terpenes, and lantipeptides. We also found that numerousStreptomycesspecies harbor BGCs known to encode antitumor compounds. We observed that strains that are considered the same species can vary tremendously in the BGCs they carry, suggesting that strain-level genome sequencing can uncover high levels of BGC diversity and potentially useful derivatives of any one compound. These findings suggest that a strain-level strategy for exploring secondary metabolites for clinical use provides an alternative or complementary approach to discovering novel pharmaceutical compounds from microbes.