Automatic Detection Method of Dairy Cow Feeding Behaviour Based on YOLO Improved Model and Edge Computing.

The feeding behaviour of cows is an essential sign of their health in dairy farming. For the impression of cow health status, precise and quick assessment of cow feeding behaviour is critical. This research presents a method for monitoring dairy cow feeding behaviour utilizing edge computing and deep learning algorithms based on the characteristics of dairy cow feeding behaviour. Images of cow feeding behaviour were captured and processed in real time using an edge computing device. A DenseResNet-You Only Look Once (DRN-YOLO) deep learning method was presented to address the difficulties of existing cow feeding behaviour detection algorithms' low accuracy and sensitivity to the open farm environment. The deep learning and feature extraction enhancement of the model was improved by replacing the CSPDarknet backbone network with the self-designed DRNet backbone network based on the YOLOv4 algorithm using multiple feature scales and the Spatial Pyramid Pooling (SPP) structure to enrich the scale semantic feature interactions, finally achieving the recognition of cow feeding behaviour in the farm feeding environment. The experimental results showed that DRN-YOLO improved the accuracy, recall, and mAP by 1.70%, 1.82%, and 0.97%, respectively, compared to YOLOv4. The research results can effectively solve the problems of low recognition accuracy and insufficient feature extraction in the analysis of dairy cow feeding behaviour by traditional methods in complex breeding environments, and at the same time provide an important reference for the realization of intelligent animal husbandry and precision breeding.

Unprecedented non-hysteretic superelasticity of [001]-oriented NiCoFeGa single crystals.

Superelasticity associated with the martensitic transformation has found a broad range of engineering applications1,2. However, the intrinsic hysteresis3 and temperature sensitivity4 of the first-order phase transformation significantly hinder the usage of smart metallic components in many critical areas. Here, we report a large superelasticity up to 15.2% strain in [001]-oriented NiCoFeGa single crystals, exhibiting non-hysteretic mechanical responses, a small temperature dependence and high-energy-storage capability and cyclic stability over a wide temperature and composition range. In situ synchrotron X-ray diffraction measurements show that the superelasticity is correlated with a stress-induced continuous variation of lattice parameter accompanied by structural fluctuation. Neutron diffraction and electron microscopy observations reveal an unprecedented microstructure consisting of atomic-level entanglement of ordered and disordered crystal structures, which can be manipulated to tune the superelasticity. The discovery of the large elasticity related to the entangled structure paves the way for exploiting elastic strain engineering and development of related functional materials.

A structural modeling approach to solid solutions based on the similar atomic environment.

A solid solution is one of the important ways to enhance the structural and functional performance of materials. In this work, we develop a structural modeling approach to solid solutions based on the similar atomic environment (SAE). We propose a similarity function associated with any type of atom cluster to describe quantitatively the configurational deviation from the desired solid-solution structure that is fully disordered or contains short-range order (SRO). In this manner, the structural modeling for solid solutions is transferred to a minimization problem in the configuration space. Moreover, we strive to enhance the practicality of this approach. The approach and implementation are demonstrated by cross validations with the special quasi-random structure method. We apply the SAE method to the typical quinary CoCrFeMnNi high-entropy alloy, continuous binary Ta-W alloy, and ternary CoCrNi medium-entropy alloy with SRO as prototypes. In combination with ab initio calculations, we investigate the structural properties and compare the calculation results with experiments.

An enhancement algorithm for head characteristics of caged chickens detection based on cyclic consistent migration neural network.

The enclosed multistory poultry housing is a type of poultry enclosure widely used in industrial caged chicken breeding. Accurate identification and detection of the comb and eyes of caged chickens in poultry farms using this type of enclosure can enhance managers' understanding of the health of caged chickens. However, the accuracy of image detection of caged chickens will be affected by the enclosure's entrance, which will reduce the precision. Therefore, this paper proposes a cage-gate removal algorithm based on big data and deep learning Cyclic Consistent Migration Neural Network (CCMNN). The method achieves automatic elimination and restoration of some key information in the image through the CCMNN network. The Structural Similarity Index Measure (SSIM) between the recovered and original images on the test set is 91.14%. Peak signal-to-noise ratio (PSNR) is 25.34dB. To verify the practicability of the proposed method, the performance of the target detection algorithm is analyzed both before and after applying the CCMNN network in detecting the combs and eyes of caged chickens. Different YOLOv8 detection algorithms, including YOLOv8s, YOLOv8n, YOLOv8m, and YOLOv8x, were used to verify the algorithm proposed in this paper. The experimental results demonstrate that compared to images without CCMNN processing, the precision of comb detection of caged chickens is improved by 11, 11.3, 12.8, and 10.2%. Similarly, the precision of eye detection for caged chickens is improved by 2.4, 10.2, 6.8, and 9%. Therefore, more complete outline images of caged chickens can be obtained using this algorithm and the precision in detecting the comb and eyes of caged chickens can be enhanced. These advancements in the algorithm offer valuable insights for future poultry researchers aiming to deploy enhanced detection equipment, thereby contributing to the accurate assessment of poultry production and farm conditions.

Engineering metal-carbide hydrogen traps in steels.

Hydrogen embrittlement reduces the durability of the structural steels required for the hydrogen economy. Understanding how hydrogen interacts with the materials plays a crucial role in managing the embrittlement problems. Theoretical models have indicated that carbon vacancies in metal carbide precipitates are effective hydrogen traps in steels. Increasing the number of carbon vacancies in individual metal carbides is important since the overall hydrogen trapping capacity can be leveraged by introducing abundant metal carbides in steels. To verify this concept, we compare a reference steel containing titanium carbides (TiCs), which lack carbon vacancies, with an experimental steel added with molybdenum (Mo), which form Ti-Mo carbides comprising more carbon vacancies than TiCs. We employ theoretical and experimental techniques to examine the hydrogen trapping behavior of the carbides, demonstrating adding Mo alters the hydrogen trapping mechanism, enabling hydrogen to access carbon vacancy traps within the carbides, leading to an increase in trapping capacity.

Leveraging SOLOv2 model to detect heat stress of poultry in complex environments.

Heat stress is one of the most important environmental stressors facing poultry production. The presence of heat stress will reduce the antioxidant capacity and immunity of poultry, thereby seriously affecting the health and performance of poultry. The paper proposes an improved FPN-DenseNet-SOLO model for poultry heat stress state detection. The model uses Efficient Channel Attention (ECA) and DropBlock regularization to optimize the DenseNet-169 network to enhance the extraction of poultry heat stress features and suppress the extraction of invalid background features. The model takes the SOLOv2 model as the main frame, and uses the optimized DenseNet-169 as the backbone network to integrate the Feature Pyramid Network to detect and segment instances on the semantic branch and mask branch. In the validation phase, the performance of FPN-DenseNet-SOLO was tested with a test set consisting of 12,740 images of poultry heat stress and normal state, and it was compared with commonly used object detection models (Mask R CNN, Faster RCNN and SOLOv2 model). The results showed that when the DenseNet-169 network lacked the ECA module and the DropBlock regularization module, the original model recognition accuracy was 0.884; when the ECA module was introduced, the model's recognition accuracy improved to 0.919. Not only that, the recall, AP0.5, AP0.75 and mean average precision of the FPN-DenseNet-SOLO model on the test set were all higher than other networks. The recall is 0.954, which is 15, 8.8, and 4.2% higher than the recall of Mask R CNN, Faster R CNN and SOLOv2, respectively. Therefore, the study can achieve accurate segmentation of poultry under normal and heat stress conditions, and provide technical support for the precise breeding of poultry.

Density functional study of CO adsorbed on MnN (N = 2-8) clusters.

The geometry, electronic structure, magnetism, and adsorption properties of one CO molecule on the Mn(N) (N = 2-8) clusters have been investigated based on the density functional theory (DFT) with the spin polarized generalized gradient approximation. It is found that the CO molecule adsorbs on the atop site for N = 2, 4, 7, 8 and on the bridge site for N = 3, 5, 6. The results of the calculated second-order energy differences of bare Mn(N) cluster indicate that the Mn(3), Mn(6), and Mn(8) clusters have relatively low stability. However, their corresponding CO complexes possess high adsorption ability implied by the C-O bond length, vibrational frequency, adsorption energy, and the charge transfer between the CO molecule and the clusters. Compared with bare Mn clusters, the adsorbing of a CO molecule enhances the magnetic moments of the Mn(N) clusters for N = 4, 6-8.

Lattice distortion inducing local antiferromagnetic behaviors in FeAl alloys.

In this work, we study the local magnetic moment as a function of order degree in solid-solution FeAl alloys. Using the combination of ab initio method and similar atomic environment model, we find that the decrease of magnetic moment, even antiferromagnetic behavior, of the Fe atoms derives from the distorted local atomic clusters centered at Fe atoms on the Fe-atom sublattice sites in B2 FeAl alloys. While the local magnetic moment of Fe atoms is up to 2.2 µ B on the Al and Fe solid-solution sublattice sites. The ordering results in the decrease of Curie temperature and magnetic moment of solid-solution FeAl alloys.

Tests on the Accuracy and Scalability of the Full-Potential DFT Method Based on Multiple Scattering Theory.

We investigate a reduced scaling full-potential DFT method based on the multiple scattering theory (MST) code MuST, which is released online (https://github.com/mstsuite/MuST) very recently. First, we test the accuracy by calculating structural properties of typical body-centered cubic (BCC) metals (V, Nb, and Mo). It is shown that the calculated lattice parameters, bulk moduli, and elastic constants agree with those obtained from the VASP, WIEN2k, EMTO, and Elk codes. Second, we test the locally self-consistent multiple scattering (LSMS) mode, which achieves reduced scaling by neglecting the multiple scattering processes beyond a cut-off radius. In the case of Nb, the accuracy of 0.5 mRy/atom can be achieved with a cut-off radius of 20 Bohr, even when small deformations are imposed on the lattice. Despite that the calculation of valence states based on MST exhibits linear scaling, the whole computational procedure has an overall scaling of about O ( N 1 . 6 ) , due to the fact that the updating of Coulomb potential scales almost as O ( N 2 ) . Nevertheless, it can be still expected that MuST would provide a reliable and accessible way to large-scale first-principles simulations of metals and alloys.

Mechanism of phosphate adsorption on superparamagnetic microparticles modified with transitional elements: Experimental observation and computational modelling.

Microparticle sorbents (MPSs) containing nano-Fe3O4 core, nano-layered double hydroxides shell modified with different transitional elements (TE-MPSs) are effective for capturing phosphate from wastewater. In present article, different TE-MPSs loaded with Ce, La, Zr and Hf were synthesized and the phosphorous removal performances were compared. The molecular-level mechanism of phosphate adsorption was successfully simulated by pseudo-atom model based on virtual crystal approximation method, which was in line with the spectroscopic data. The results confirmed the inter-sphere complexation reaction between phosphate and TE-MPSs under neutral pH. And the adsorption mechanism probably changed during cycles of adsorption/regeneration. As revealed by elemental composition analysis, ion exchange mechanism probably played an important part in phosphate adsorption in Cycle 1 while the ligand complexation became dominant after regenerated with NaOH in the following cycles. This work indicates that the highly effective, easily separable and repeatedly usable TE-MPSs has the potential for phosphate sequestration and recovery from sewage wastewater.

The competition of double-, four-, and three-ring tubular B(3n) (n = 8-32) nanoclusters.

The geometry and electronic properties of three-ring tubular B(3n) clusters (n = 8-32) are studied systematically with the density functional theory. It is composed of three staggered rings with the diameter of the middle ring larger than those of the two outer rings. With the increase in boron atom numbers, the three-ring tubular clusters are energetically more stable than the double-ring and four-ring tubular clusters and the buckled sheet clusters with hexagon holes. The average binding energy tends to the finite value. The stability is further analyzed through the natural bond orbital population analysis. The highest occupied and lowest unoccupied energy gaps become small, which demonstrates a favorable metallic property.

Effect of Mo on the phase stability and elastic mechanical properties of Ti-Mo random alloys from ab initio calculations.

Ti-Mo alloys are promising materials for shape memory alloys and biomedical materials. Whereas, the appearance of metastable ω phase can cause embrittlement and destroy the shape memory effect. In order to avoid the ω phase, the effect of Mo on the temperature dependent lattice parameters, phase stability and elastic mechanical properties of ß, α, and ω Ti1-x Mo x (x = 0-2.0) random alloys was systematically investigated by using the exact muffin-tin orbitals method in combination with the coherent potential approximation. The theoretical predictions for the lattice parameters are in good agreement with the available experiments. Results show that ß Ti0.96Mo0.04 can almost transform to ω phase without lattice deformation and volume change, which suggests that the athermal ω phase is easier to precipitate and grow near 4 at.% Mo content in the ß Ti1-x Mo x alloys. The critical content of Mo for the competed stabilization of ß phase at T = 300 K is ~11.2 at.%. Its valence electron concentration of 4.224 is viewed as a necessary criterion for the competed phase stability. The calculations of formation energy are used to explain successfully why the partitioning of Mo can be found in Ti0.91Mo0.09 alloy after annealing. Through the analysis of formation energy, both Mo addition and increasing temperature can stabilize the ß phase. The calculated Cauchy pressure, Pugh's ratio, Poisson ratio, and Young's modulus suggests that ω phase is intrinsically brittle and has large Young's modulus compared with ß and α phases.

Refractive index and extinction coefficient of NH2CH = NH2PbI3 perovskite photovoltaic material.

Very recently, the NH2CH = NH2PbI3 (FAPbI3) perovskite material has attracted considerable attention in fabricating solar cells (SCs). For a photovoltaic material, its refractive index and extinction coefficient, n(λ) and k(λ), as functions of λ, are important to study its optical properties and to estimate the power conversion efficiency potential for the SCs made of it. As far as we know, to date there has been no reports of n(λ) and k(λ) for FAPbI3 material. In this article, with spectroscopic ellipsometry (SE) measurements, the n(λ) and k(λ), as well as E g = 1.45 eV for FAPbI3, are acquired. The fast deposition crystallization (FDC) procedure combined with the slowed down annealing (SDA) process is applied to fabricate smooth and uniform FAPbI3 film on quartz substrate. Several kinds of organic solvents were tried as the second solvent in the FDC procedure, and it is found that when petroleum ether is used, the smallest surface roughness and good FAPbI3 material purity of the FAPbI3 film can be acquired. The k(λ) results for FAPbI3 obtained by SE, calculated from the n(λ) using the Kramers-Kronig relationship, by absorbance, and by first-principles calculations, are compared. The n(λ) and k(λ) for FAPbI3 are also compared with those for CH3NH3PbI3, GaAs and c-Si.

Ab initio-predicted micro-mechanical performance of refractory high-entropy alloys.

Recently developed high-entropy alloys (HEAs) consisting of multiple principal elements represent a new field of metallurgy and have demonstrated appealing properties for a wide range of applications. Using ab initio alloy theory, we reveal the alloying effect on the elastic properties and the ideal tensile strength (ITS) in the [001] direction of four body-centered cubic (bcc) refractory HEAs based on Zr, V, Ti, Nb, and Hf. We find that these HEAs show high elastic anisotropy and large positive Cauchy pressure, suggesting good extrinsic ductility. Starting from ZrNbHf, it is found that the ITS decreases with equimolar Ti addition. On the other hand, if both Ti and V are added to ZrNbHf, the ITS is enhanced by about 42%. An even more captivating effect is the ITS increase by about 170%, if Ti and V are substituted for Hf. The alloying effect on the ITS is explained by the d-band filling. An intrinsic brittle-to-ductile transition is found in terms of the failure mode under uniaxial tension. These investigations suggest that intrinsically ductile HEAs with high ideal strength can be achieved by controlling the proportion of group four elements to group five elements.

Ab initio study of AlxMoNbTiV high-entropy alloys.

The Al(x)MoNbTiV (x = 0-1.5) high-entropy alloys (HEAs) adopt a single solid-solution phase, having the body centered cubic (bcc) crystal structure. Here we employ the ab initio exact muffin-tin orbitals method in combination with the coherent potential approximation to investigate the equilibrium volume, elastic constants, and polycrystalline elastic moduli of Al(x)MoNbTiV HEAs. A comparison between the ab initio and experimental equilibrium volumes demonstrates the validity and accuracy of the present approach. Our results indicate that Al addition decreases the thermodynamic stability of the bcc structure with respect to face-centered cubic and hexagonal close packed lattices. For the elastically isotropic Al(0.4)MoNbTiV HEAs, the valence electron concentration (VEC) is about 4.82, which is slightly different from VEC ∼ 4.72 obtained for the isotropic Gum metals and refractory--HEAs.

Dual-protection of a graphene-sulfur composite by a compact graphene skin and an atomic layer deposited oxide coating for a lithium-sulfur battery.

A reduced graphene oxide (rGO)-sulfur composite aerogel with a compact self-assembled rGO skin was further modified by an atomic layer deposition (ALD) of ZnO or MgO layer, and used as a free-standing electrode material of a lithium-sulfur (Li-S) battery. The rGO skin and ALD-oxide coating worked as natural and artificial barriers to constrain the polysulfides within the cathode region. As a result, the Li-S battery based on this electrode material exhibited superior cycling stability, good rate capability and high coulombic efficiency. Furthermore, ALD-ZnO coating was tested for performance improvement and found to be more effective than ALD-MgO coating. The ZnO modified G-S electrode with 55 wt% sulfur loading delivered a maximum discharge capacity of 998 mA h g(-1) at a current density of 0.2 C. A high capacity of 846 mA h g(-1) was achieved after charging/discharging for 100 cycles with a coulombic efficiency of over 92%. In the case of using LiNO3 as a shuttle inhibitor, this electrode showed an initial discharge capacity of 796 mA h g(-1) and a capacity retention of 81% after 250 cycles at a current density of 1 C with an average coulombic efficiency higher than 99.7%.

Interlayer interactions in graphites.

Based on ab initio calculations of both the ABC- and AB-stacked graphites, interlayer potentials (i.e., graphene-graphene interaction) are obtained as a function of the interlayer spacing using a modified Möbius inversion method, and are used to calculate basic physical properties of graphite. Excellent consistency is observed between the calculated and experimental phonon dispersions of AB-stacked graphite, showing the validity of the interlayer potentials. More importantly, layer-related properties for nonideal structures (e.g., the exfoliation energy, cleave energy, stacking fault energy, surface energy, etc.) can be easily predicted from the interlayer potentials, which promise to be extremely efficient and helpful in studying van der Waals structures.

A novel potential: the interlayer potential for the fcc (111) plane family.

We propose a novel interlayer potential, which is different from usual interatomic potentials. The interlayer potential represents the interaction between atomic layers in a layered material. Based on the Chen-Möbius inversion method in combination with ab initio calculations, the interlayer interactions are obtained for the face centered cubic (fcc) (111) planes. In order to check the validity of our interlayer potential, we calculate the intrinsic stacking fault energy (γ(sf)) and the surface energy (γ(s)) of five metals: Al, Ni, Cu, Ag and Au. The predicted γ(sf) and γ(s) values are compared with the theoretical results obtained from direct calculations and also with the available experimental data. Using the interlayer potentials, we also investigate the phonon dispersion and phonon density of state in the fcc (111) plane family of the considered metals.

No quenching of magnetic moment for the GenCo (n=1-13) clusters: first-principles calculations.

The authors predict that for the Ge(n)Co (n=1-13) clusters the magnetic moment does not quench, which is dark contrast to the previous results with transition-metal-doped Si(n) clusters. It may be due to the unpaired electrons of the Co atom in the clusters. For the ground state structures of the Ge(n)Co (n>or=9) clusters, the Co atom completely falls into the center of the Ge outer frame, forming metal-encapsulated Ge(n) cages. The doping of the Co atom enhances the stability of the host Ge(n) clusters. The Ge(10)Co cluster with the bicapped tetragonal antiprism structure is more stable than others, which agrees very well with the results of the experiment of the Co/Ge binary clusters by the laser vaporization.