Search

Lead-based reactor coolant has high thermal conductivity, good thermal expansion and strong natural circulation ability, which has outstanding advantages in safety, economy and miniaturization. At the same time, lead-bismuth reactors have potential applications in small nuclear power sources, marine reactors and so on. Difference from land-based nuclear power plants, marine nuclear power plants will produce swaying, heaving, tilting and other movements due to their long-term operation in the ocean environment. In order to study the thermal-hydraulic characteristics of the lead-bismuth reactor system under marine conditions, an experimental study was conducted by building a lead-bismuth experimental circuit based on a heaving motion platform. Under the influence of heaving motion, the flow rate of lead-bismuth circuit fluctuates and the flow resistance and heat transfer characteristics change. In this paper, the flow resistance and heat transfer characteristics of liquid lead-bismuth eutectic in a vertical circular tube under forced circulation under heaving conditions are investigated experimentally. The range of Reynolds number (Re) in the flow resistance experiment is 23 500-261 280, and the range of Peclet number (Pe) in the flow heat transfer experiment is 524.4 450. The results show that: The peak value of instantaneous additional pressure drop caused by heaving motion is lower than the theoretical value; The heaving motion causes periodic fluctuations in the flow rate of the circuit, and the maximum fluctuation is up to 16% in the low Re region; In the low Re region, the heaving condition makes the time-averaged frictional resistance coefficient of the pipe increase slightly; The relationship between the fluctuation of loop flow caused by undulating motion and the undulating acceleration and Re was obtained by fitting, which is in good agreement with the experimental results. This method can be used to predict the fluctuation of flow rate in the reactor at different coolant flow rates by heaving motion. In the scope of this experimental working condition, although the fluctuation causes the fluctuation of the flow rate, the instantaneous temperature difference between the inlet and outlet of the experimental section and the instantaneous heat transfer coefficient h of the experimental section do not fluctuate significantly, and the fluctuation has little impact on the instantaneous heat transfer characteristics. In the forced circulation, the heaving motion does not affect the heat transfer characteristics of the circuit. The results of this study can provide data reference for the engineering design of lead-bismuth reactor.

Lead-cooled fast reactor (LFR) is known as a typical reliable new type reactor by the US Department of Energy (DOE), and the potential accidents of LFR are necessary to assess. Steam generator tube rupture (SGTR) accident can cause high-pressure hyperbaric steam-water jetting into liquid lead bismuth eutectic (LBE) pool, and leads to a rapid pressure rising in the reactor primary system together with a potential pressure wave crashing the structures. This paper focuses on the computational fluid dynamics (CFD) analyses of SGTR experimental research on the LIFUS5/MOD2 experiment facility. A two-dimensional symmetrical slice with 1 million 570 thousand structured mesh was established for the experimental reaction vessel, and each side except the injection nozzle were set as adiabatic no slip boundary in order to achieve the non-decompression process. In the aspect of multiphase model settings, the Eulerian-Eulerian two-phase flow model was adopted to track the steam-LBE interface in the transient simulation, and the CSF surface tension polynomial together with Lee substituted evaporation model were considered. In addition, the injection of high-pressure water in each time step was separated to two parts: latent heat and injected steam, so as to calculate both endothermic and jet of flash evaporation simultaneously. Among them, the latent heat was realized by user-defined functions (UDF), and the injected steam was adopted as compressible real gas with R-K gas state equation. Based on the above settings, three established thermal-hydraulics phenomena including pressure rising of LBE environment together with pressure wave spreading, level fluctuation of liquid environment with argon entrainment and level rising of liquid environment and diffusion of injected steam were analyzed. From the results a conclusion can be reached that CFD numerical simulation has little error on calculating pressure rising with pressure wave spreading. In addition, the peak value of pressure wave rises with back pressure of water increasing, and the pressure of partial steam cavity is lower than that of LBE pool around it, in turn inhibits the occurrence of its explosion. Meanwhile, LBE in the middle firstly washes up and impacts the structures, and then entrains the argon when it falls back, in turn causes the combination of argon and steam cavity, or the secondary segmentation to small argon bubbles. Moreover, the diffusion of the steam cavity along the radial direction becomes more intense with the increase of the axial height, accompanies by the steam cavity divided from the center to the edge along the radius.

Generative Adversarial Networks (GANs) possess remarkable fitting capabilities and play a crucial role in the field of computer vision. Super-resolution restoration is the process of converting low-resolution images into high-resolution ones, providing more detail and information. This is of paramount importance for monitoring and managing forest resources, enabling the surveillance of vegetation, wildlife, and potential disruptive factors in forest ecosystems. In this study, we propose an image super-resolution model based on Generative Adversarial Networks. We incorporate Multi-Scale Residual Blocks (MSRB) as the core feature extraction component to obtain image features at different scales, enhancing feature extraction capabilities. We introduce a novel attention mechanism, GAM Attention, which is added to the VGG network to capture more accurate feature dependencies in both spatial and channel domains. We also employ the adaptive activation function Meta ACONC and Ghost convolution to optimize training efficiency and reduce network parameters. Our model is trained on the DIV2K and LOVEDA datasets, and experimental results indicate improvements in evaluation metrics compared to SRGAN, with a PSNR increase of 0.709/2.213 dB, SSIM increase of 0.032/0.142, and LPIPS reduction of 0.03/0.013. The model performs on par with Real-ESRGAN but offers significantly improved speed. Our model efficiently restores single-frame remote sensing images of forests while achieving results comparable to state-of-the-art methods. It overcomes issues related to image distortion and texture details, producing forest remote sensing images that closely resemble high-resolution real images and align more closely with human perception. This research has significant implications on a global scale for ecological conservation, resource management, climate change research, risk management, and decision-making processes. [ABSTRACT FROM AUTHOR]

Simultaneous Localization and tracking was studied to solute the problem of tracking a moving target in a sensor network while simultaneously localizing and calibrating the nodes of the network. RSSI is used for measuring the distance between the nodes pairs, multidimensional scaling techniques are used for the initial position of wireless sensor network based on the distance matrix. Then the compression Kalman filter is used to estimate and update the sensor node position and the target position. Simulation results show that the algorithm has high accuracy and real-time performance under low network, especially for long distance tracking.

Plant leaves can be used to effectively detect plant diseases. However, the number of images of unhealthy leaves collected from various plants is usually unbalanced. It is difficult to detect diseases using such an unbalanced dataset. We used DoubleGAN (a double generative adversarial network) to generate images of unhealthy plant leaves to balance such datasets. We proposed using DoubleGAN to generate high-resolution images of unhealthy leaves using fewer samples. DoubleGAN is divided into two stages. In stage 1, we used healthy leaves and unhealthy leaves as inputs. First, the healthy leaf images were used as inputs for the WGAN (Wasserstein generative adversarial network) to obtain the pretrained model. Then, unhealthy leaves were used for the pretrained model to generate 64*64 pixel images of unhealthy leaves. In stage 2, a superresolution generative adversarial network (SRGAN) was used to obtain corresponding 256*256 pixel images to expand the unbalanced dataset. Finally, compared with images generated by DCGAN (Deep convolution generative adversarial network). The dataset expanded with DoubleGAN, the generated images are clearer than DCGAN, and the accuracy of plant species and disease recognition reached 99.80 and 99.53 percent, respectively. The recognition results are better than those from the original dataset. [ABSTRACT FROM AUTHOR]