Search

Realizing an accurate laying rate prediction based on environmental factors plays a vital role in livestock and poultry breeding. In this paper, multiple environmental factors were considered to improve the accuracy of egg production rate prediction. A method was proposed by combining the Random Forest (RF) and Long Short-Term Memory (LSTM) to analyze the impact of the external environmental factors on the laying rate. Firstly, using RF, feature importance selection was implemented on environmental factors affecting laying rate. Secondly, the extreme Gradient Boosting (XGBoost) was introduced as a comparison to evaluate the accuracy and reliability of the RF feature importance selection. Finally, by discarding the features with low importance one by one, the multi-variable RF-LSTM laying rate prediction was conducted. Experiment results showed that the proposed RF-LSTM method significantly improved the prediction accuracy on laying rate.

A facile and low-cost two components spray deposition technique has been developed to fabricate highly conductive silver films on carbon fiber fabrics (CFFs) surface for electromagnetic interference (EMI) shielding applications. Microstructures and components of the samples were characterized by X-ray diffraction (XRD), scanning electron microscope (SEM), X-ray photoelectron spectroscopy (XPS), and energy-dispersive X-ray spectroscopy (EDS), respectively. As measured by four-point probe instrument, silver maintains its unique conductivity as a film on CFFs surface. Additionally, the vector network analyzer (VNA) data illustrated that the incident electromagnetic waves were efficiently attenuated by the reflection on the surface of the network structure and then absorbed in the composites, which primarily integrates the high conductivity of silver film. Compared with the pristine CFFs, the EMI shielding effectiveness (SE) of Ag-coated CFFs with 100 spraying passes could reach to 80.82 dB. Therefore, it is expected that this work will open up a convenient and compatible strategy to exploiting fabrics materials with high-performance EMI shielding.

Background: Magnesium (Mg) is essential to many plant physiological and biochemical processes; however, a quantitative understanding of how Mg nutrition affects the production, partitioning and utilization of photoassimilates is still lacking, especially for soilless culture system. We focused on the roles of Mg in yield formation and interactions with potassium (K) nutrition of cherry tomato. Cherry tomato yield, leaf Mg concentration, photosynthetic parameters, dry matter weight and K, Mg and calcium (Ca) uptake were investigated in two soilless experiments with seven Mg levels and five K levels.Results: Low (4 mM) Mg supply limited cherry tomato yield by decreasing 22.6-78.1% dry matter accumulation and 13.9-40.7% harvest index. The critical leaf Mg concentrations required for adequate photosynthate production in the first and second harvest periods were 4.67 and 5.52 g kg –1 , respectively. But over-supply of Mg disturbed leaf K and Ca concentrations, limiting the plant K and Ca content. Moreover, adjusting K concentrations in solution is a crucial factor influencing plant Mg functions, and therefore cherry tomato growth.Conclusion: Balanced Mg and K application increased Mg, K, and Ca uptake by cherry tomato, as well as Mg concentrations in leaves which could maintain a sustainable photosynthetic rate and plant dry matter formation.

Real-time smoke detection is of great significance for early warning of fire, which can avoid the serious loss caused by fire. Detecting smoke in actual scenes is still a challenging task due to large variance of smoke color, texture, and shapes. Moreover, the smoke detection in the actual scene is faced with the difficulties in data collection and insufficient smoke datasets, and the smoke morphology is susceptible to environmental influences. To improve the performance of smoke detection and solve the problem of too few datasets in real scenes, this paper proposes a model that combines a deep convolutional generative adversarial network and a convolutional neural network (DCG-CNN) to extract smoke features and detection. The vibe algorithm was used to collect smoke and nonsmoke images in the dynamic scene and deep convolutional generative adversarial network (DCGAN) used these images to generate images that are as realistic as possible. Besides, we designed an improved convolutional neural network (CNN) model for extracting smoke features and smoke detection. The experimental results show that the method has a good detection performance on the smoke generated in the actual scenes and effectively reduces the false alarm rate.

In this study, spray-assisted layer by layer chemical deposition was used for the preparation of a thin Cu layer with micro-nano structures on carbon fiber fabrics (CFFs). By adjusting the spraying passes, flocculent layered coarse structures composed of Cu nanoparticles have uniformly been constructed successfully. These copper-coated carbon fiber fabrics (Cu@CFFs) showed excellent electromagnetic interference (EMI) shielding effectiveness (SE), and the total SE of Cu@CFFs-100 was about 20 dB higher than the original carbon fiber fabrics. In addition to the reflection and adsorption effectiveness results, the high SE mainly depends on the outstanding conductivity of the carbon fiber fabric endowed by the deposited nanosized copper and enhanced the reflection loss of the incident electromagnetic wave and the multiple reflections of the internal network structure. Moreover, the stability and multilayer overlay of Cu@CFFs results show the possibility of applying spray-assisted layer by layer chemical deposition to produce EMI shielding materials.

This article presents a numerical method for determining tortuosity in porous beds consisting of randomly packed spherical particles. The calculation of tortuosity is car- ried out in two steps. In the first step, the spacial arrangement of particles in the porous bed is determined by using the discrete element method (DEM). Specifically, a commercially available discrete element package (PFC 3D ) was used to simulate the spacial structure of the porous bed. In the second step, a numerical algorithm was developed to construct the microscopic (pore scale) flow paths within the simulated spacial structure of the porous bed to calculate the lowest geometric tortuosity (LGT), which was defined as the ratio of the shortest flow path to the total bed depth. The numerical algorithm treats a porous bed as a series of four-particle tetrahedron units. When air enters a tetrahedron unit through one face (the base triangle), it is assumed to leave from another face triangle whose centroid is the highest of the four face triangles associated with the tetrahedron, and this face triangle will then be used as the base triangle for the next tetrahedron. This process is repeated to establish a series of tetrahedrons from the bottom to the top surface of the porous bed. The shortest flow path is then constructed geometrically by connecting the centroids of base triangles of consecutivetetrahedrons.Thetortuosityvaluescalculatedbytheproposednumericalmethod comparedfavourablywiththevaluesobtainedfromaCTimagepublishedintheliteraturefor a bed of grain (peas). The proposed model predicted a tortuosity of 1.15, while the tortuosity estimated from the CT image was 1.14.

A three dimensional discrete element model is developed by using the Particle Flow Code (PFC3D) to simulate vibration-induced pressures in model bins. Grain kernels are simulated as elastic spheres and interactions between grain kernels are modeled by contact stresses in PFC3D. The bin size that can be simulated is limited by the computer capacity. However, the simulation results may be scaled up by proportionally increasing the grain density in the simulation model. The simulated pressures were in reasonably good agreement (within 10% near the bin bottom) with the reported experimental data and the Janssen theory for the static condition (before vibration) for a 1.5 m (height) by 1.0 m (diameter) model bin filled with wheat. The simulated vibration-induced pressure (static pressure after vibration) was also within 10% of the measured values in the lower portion of the bin. The simulated dynamic pressure varied with the vibration cycle, between 50% and 150% of the static pressure.

A discrete element model (Particle Flow Code, PFC3D) was used to simulate vibration of bulk solids in storage bins. The vibration frequencies and amplitudes of individual particles were determined from simulated particle velocities and displacements at different excitation frequencies and amplitudes. A model bin test system was used to conduct experiments to validate the PFC model. The 0.28 (height) by 0.15 m (diameter) model bin was made of Plexiglas and filled with soybeans. The bin was vibrated at frequencies from 5 to 30 Hz and amplitudes from 0.5 to 5.0 mm. A high speed digital imaging system was used to record particle movement along the bin wall during vibration. Simulated vibration of particles was in good agreement with the experimental data. Simulations revealed that particle vibration was simple harmonic and had the same frequency and amplitude as the excitation at low excitation frequencies and amplitudes, whereas, particles vibrated at their own frequencies and amplitudes when the excitation frequency and amplitude were high.