Search

Aiming at the problems of unbalanced inter-modal optimization and inadequate fusion of multimodal features in multimodal sentiment analysis, a multimodal sentiment analysis model combining dynamic gradient mechanism and multi-view co-attention mechanism (DG-MCM) is proposed, which can effectively mine single-modal representation and fully integrate multimodal information. Firstly, the model uses pre-trained model BERT (bidirectional encoder representation from transformers) and stacked long short-term memory (SLSTM) to learn the features of text, audio and video, and proposes a dynamic gradient mechanism. By monitoring the contribution difference and learning speed of each mode, the feature learning of each mode is assisted. Secondly, the features of different modes obtained are fused using the multi-view co-attention mechanism. By projecting every two modes into multiple spaces for interaction, more adequate fusion features are obtained. Finally, fusion features and single-modal features are spliced together for sentiment prediction. Experimental results on CMU-MOSI and CMU-MOSEI datasets show that this model can fully learn information between single mode and different modes, and effectively improve the accuracy of multimodal sentiment analysis.

The nuclear thermal propulsion (NTP) boasts advantages such as high specific impulse, substantial thrust, and extended operating time, giving it a clear edge in deep space exploration and orbital maneuvers. To fully harness the potential of NTP, a comprehensive system performance analysis is crucial to ensure reliability and efficiency under various operational conditions. This study provided an overview of the historical development and current status of nuclear thermal propulsion system analysis programs, both nationally and internationally. It also conducted an in-depth analysis of programs utilized in projects such as NERVA and SNTP. Building on an examination of historical international program advancements and current research requirements, system performance analysis is categorized into three primary components:steady-state performance design and optimization, steady-state off-design performance analysis, and transient performance analysis. Based on the independently developed program for analyzing nuclear engine systems (PANES) at Institute of Nuclear and New Energy Technology, Tsinghua University, an integrated NTP system analysis program framework based on "flow network-heat network" was proposed. The one-dimensional flow network model employed a concept similar to the staggered grid commonly used in computational fluid dynamics (CFD). It separately handled the momentum conservation equation in the resistance/inertia segment and the continuity and energy conservation equations in the capacitance segment. The thermal network model was solved based on the solid energy equation considering heat capacity. Coupled calculations of the thermal network and flow network were achieved by considering heat exchange in the capacitance segment of the flow network. Specifically, a point reactor model based on backward Euler (BE) or Crank-Nicolson (CN) time discretization formats was established to facilitate the calculation of power produce processes. To enhance the understanding and optimization of turbine performance, a turbine model was developed, employing iterative methods to determine mass flow based on turbine inlet parameters, pressure ratios, and characteristic curves. Additionally, to accurately simulate centrifugal pump behavior, a centrifugal pump model based on Suter curves was implemented. The integration of these models provides a comprehensive approach for considering and effectively designing and optimizing the performance of turbine and pump components within the NTP system. This work lays a crucial foundation for further research and application of NTP system design methods, paving the way for future advancements in space exploration and offering exciting prospects for upcoming space missions.

The performance of space nuclear reactors and spacecraft is significantly influenced by the size and mass of shield, therefore shielding design and optimization are crucial for the development of space nuclear power systems. In this paper, shielding design and optimization process was proposed, and the effectiveness was verified through an optimized shielding design for the Jupiter Icy Moons Orbiter (JIMO) reactor. Building upon the open lattice reactor concept of the JIMO project, the neutronic design of the JIMO reactor was supplemented. The shielding design and optimization for the reactor employed a layered combination of beryllium (Be), boron carbide (B4C), tungsten (W), and lithium hydride (LiH). Considering the radial distribution of radiation dose, the Monte Carlo method was utilized to compute the neutron flux and the photon dose at the payload. Additionally, the neutron dose at the leading edge of the LiH was taken into account. Given the computational cost associated with the Monte Carlo method, the stepwise optimization approach was proposed after the analysis of the coupled transport characteristics of neutrons and photons and the shielding design principles. The stepwise optimization and analysis revealed several key findings. Firstly, the multi-layer Be-B4C configuration, compared to the single-layer Be plus single-layer B4C arrangement, effectively reduces reflected neutrons, thereby diminishing the mass required for photon shielding material. This reduction results from a decrease in the secondary photons generated within the leading-edge structural materials. Secondly, due to the lower photon doses at the outer edge resulting from the strong penetration ability of photons, the mass of the photon shielding material can be reduced by decreasing the photon shielding radius. Thirdly, a shielding configuration with a Be to B4C thickness ratio of 7:3 demonstrates excellent shielding effectiveness while maintaining a relatively small mass. Finally, placing W at 30 cm from the shielding leading edge not only reduces the photon shielding radius but also decreases the generation of the secondary photons, leading to an optimal shielding mass. The optimized shielding demonstrates equivalent radiation attenuation capabilities to the JIMO shielding, encompassing the attenuation of neutron flux and photon dose at the payload and neutron absorption dose at the leading edge of LiH. Simultaneously, the shielding mass is reduced by 98.41 kg. This underscores the effectiveness of the shielding optimization process. The developed shield and the associated design process have the potential to serve as a valuable reference for future shielding design and optimization for space nuclear reactors.

Tea (Camellia sinensis L.) is one of the most important and traditional economic crops widely cultivated in the subtropical regions of China, which are usually developed from forestland. Soil enzyme activity is an important indicator of soil fertility and nutrient transformation. The purpose of this study is to investigate the seasonal dynamic characteristics of soil properties and enzyme activities of different planting patterns in Wuyishan city, and to provide theoretical basis for reasonable evaluation of soil ecological effects of organic tea cultivation. In this paper, three different planting patterns (forestland, conventional and organic tea gardens) were selected as the research objects. Soil samples were collected in May, August, November and February from 2021 to 2022. The soil properties and enzyme activities (urease, nitrate reductase, polyphenol oxidase, catalase, invertase and acid phosphatase) were determined in different seasons, and the dynamic changes with seasons were also investigated. The results show that: comparing with the forestland, the contents of soil ammonium nitrogen, total phosphorus, available phosphorus and available potassium increased significantly in the conventional tea garden, while the total potassium and pH decreased significantly. Compared with the conventional tea garden, the soil organic matter and total nitrogen contents increased significantly in the organic tea garden. The soil total phosphorus, available phosphorus, total potassium and available potassium contents decreased significantly. The soil pH also increased, and the proportion of soil nutrients was more coordinated. The effects of planting pattern and season and their interactions on urease and peroxidase activities were significant. Compared with the forestland, the soil urease, polyphenol oxidase, catalase and acid phosphatase activities decreased by 12.05% to 63.55% in the conventional tea garden, while urease activities significantly increased by 324.95% in the organic tea garden, and the soil nitrate reductase activities were not changed by planting mode. In general, the soil urease, polyphenol oxidase, invertase and acid phosphatase activities were significantly higher in summer and autumn (May and August) than those in winter and spring (November and February). The highest soil nitrate reductase and catalase activities were found in spring (February). The results of permutational multivariate analysis of variance show that the effect of planting pattern on the overall soil physical and chemical properties was much greater than that of seasonal changes. Redundancy analysis shows that soil environmental factors explained 77.03% of the variation in soil enzyme activity, and the soil organic matter, total nitrogen, ammonium nitrogen, total phosphorus, soil available phosphorus, total potassium, available potassium and pH were the main driving factors of soil enzymes. In summary, the conversion of forestland into tea gardens has a significant impact on soil properties and enzyme activities. Conventional planting leads to the accumulation of available phosphorus and potassium in tea garden soil and the decrease of soil enzyme activity, while organic planting improves soil enzyme activity and enhances soil carbon and nitrogen nutrient supply capacity, and thus is beneficial for maintaining a sustainable ecosystem in tea garden soil. [ABSTRACT FROM AUTHOR]

In order to develop the best formula for shrimp slide with Hypsizygus marmoreus, according to the main raw materials of shrimp slide with Hypsizygus marmoreus, the colour difference, texture, and sensory scores were used as the quality indicators. Then, the effects of different granularities, different additions of Hypsizygus marmoreus, starch, and egg white on the gelatin of minced shrimp with Hypsizygus marmoreus and shrimp slide were explored, in order to produce the high-quality shrimp slide with Hypsizygus marmoreus. The results showed that Hypsizygus marmoreus of homogenate particles and shrimp slide could be better fused, resulting in the higher whiteness of shrimp slide. Additionally, the maximum values for the gel of minced shrimp and the a* value, gel strength, and rupture force of shrimp slide were achieved with the addition of 7% Hypsizygus marmoreus. The maximum whiteness of shrimp slide was achieved with the addition of 5% starch, but when the amount of starch added was too much, the shrimp slide had obvious powder feeling, and the sensory score decreased sharply. On the other hand, the addition of 3% egg white resulted in the maximum whiteness, a* and gel strength of shrimp slide. An orthogonal test was used to optimize the best formulation of shrimp slide with Hypsizygus marmoreus, resulting in 7% homogenate of Hypsizygus marmoreus, 7.5% starch, and 3% egg white. According to the analysis of nutritional components, the contents of reducing sugar, dietary fibre, and sodium in shrimp slide with Hypsizygus marmoreus were significantly improved (P<0.05)under this formula compared with the shrimp slide without the addition of Hypsizygus marmoreus. [ABSTRACT FROM AUTHOR]

Objectives: To investigate the value of single-phase gonadotropin releasing hormone (GnRH) stimulation test in the diagnosis of central precocious puberty (CPP) in girls with different levels of body mass index (BMI)., Methods: A retrospective analysis was performed for the data of 760 girls with breast development before 7.5 years of age who attended the Third Affiliated Hospital of Zhengzhou University from January 2017 to August 2023. According to the results of GnRH stimulation test and clinical manifestations, they were divided into a CPP group (297 girls) and a non-CPP group (463 girls). According to the values of BMI, the girls were divided into a normal weight group (540 girls), an overweight group (116 girls), and an obese group (104 girls). The receiver operating characteristic (ROC) curve was used to investigate the value of single-phase GnRH stimulation test in the diagnosis of CPP in girls with different levels of BMI., Results: Luteinizing hormone (LH)/follicle-stimulating hormone at 30 minutes after GnRH stimulation had an area under the curve (AUC) of 0.985 in the diagnosis of CPP, which was higher than the AUC at 0, 60, and 90 minutes ( P <0.05). LH at 30 minutes had a similar diagnostic value to LH at 60 minutes ( P >0.05). LH at 30 minutes was negatively correlated with BMI and BMI-Z value ( P <0.05).The AUC for diagnosing CPP in normal weight, overweight, and obese girls at 30 minutes LH was 0.952, 0.965, and 0.954, respectively ( P <0.05)., Conclusions: The 30-minute GnRH stimulation test has a good value in the diagnosis of CPP in girls with different levels of BMI and is expected to replace the traditional GnRH stimulation test, but the influence of BMI on LH level should be taken seriously.

Important forensic diagnostic indicators of sudden death in coronary atherosclerotic heart disease, such as acute or chronic myocardial ischemic changes, sometimes make it difficult to locate the ischemic site due to the short death process, the lack of tissue reaction time. In some cases, the deceased died of sudden death on the first-episode, resulting in difficulty for medical examiners to make an accurate diagnosis. However, clinical studies on coronary instability plaque revealed the key role of coronary spasm and thrombosis caused by their lesions in sudden coronary death process. This paper mainly summarizes the pathological characteristics of unstable coronary plaque based on clinical medical research, including plaque rupture, plaque erosion and calcified nodules, as well as the influencing factors leading to plaque instability, and briefly describes the research progress and technique of the atherosclerotic plaques, in order to improve the study on the mechanism of sudden coronary death and improve the accuracy of the forensic diagnosis of sudden coronary death by diagnosing different pathologic states of coronary atherosclerotic plaques.