04 December 2024, Volume 23 Issue 5

  

• Select all
  |

• Select
  Comprehensive review on sea surface wind field retrieval using spaceborne microwave scatterometer

  2024, 23(5): 1-0.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  The spaceborne microwave scatterometer, characterized by its high spatiotemporal resolution and all-weather observational capability, serves as a pivotal tool for observing the sea surface wind fields with multiple bands, polarizations, and viewing angles. It has emerged as a crucial means to acquire high-resolution sea surface wind field data over expansive regions, playing a central role as the primary satellite sensor for global sea surface wind field observations. Consequently, investigating its intricacies holds significant academic significance. This paper offers a comprehensive and systematic examination of the advancements made by scholars globally in the domain of sea surface wind field inversion methods. Emphasizing the evolution of geophysical model functions, the optimization of fuzzy solution removal schemes, and the application of neural network algorithms and deep learning in ocean remote sensing, it provides a nuanced exploration of this interdisciplinary field. By delving into these topics, this paper not only furnishes valuable insights for advancing sea surface wind field retrieval technology but also presents novel perspectives for future research and applications within the realm of ocean remote sensing.
• Select
  A review on modeling net primary productivity of vegetation

  2024, 23(5): 13-12.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Net primary productivity ( NPP) of vegetation is an important component of the surface carbon cycle, and the accurate assessment of NPP is of great significance to the correct understanding of ecosystem energy transformation and the evaluation of ecosystem health. While existing studies have reviewed NPP assessment from perspectives such as light use efficiency, ecosystem process simulation, and remote sensing data-driven approaches, there remains a need for further refinement in reviewing specific drug molecules, and the relationship trend between different descriptors and the three performance indicators were preliminarily explored. Secondly, six machine learning models including Random Forest, Extreme Gradient Boosting ( XGB) , Gradient Boosting Decision Tree, Light Gradient Boosting Machine, Backpropagation Neural Network, and Support Vector Regression six machine learning algorithms with eight descriptors and three performance evaluation criteria ( Adsorption selectivity SS-IBU/ N2 ,Adsorption capacity NS-IBU/ N2 and Tradeoff value TSN) for big data training and mining, were used to establish quantitative relationships. The results show that the prediction accuracy of the six ML algorithms is N > TSN > S . For S , XGB showed the best prediction ( R2 = 0.83) . Subsequently, based on the XGB model, the SHaple Additive explanation ( SHAP) method was used to explain and analyze the importance of MOF descriptors to performance indicators. The total energy generated during MOF adsorption is considered to be the key influencing factor, and it shows a positive correlation trend with both TSN and NS-IBU . Finally, combined with toxicological analysis, a series of high-performance MOF materials were recommended and designed. This work, from molecular level, high-throughput computing to big data mining, systematically studied the adsorption and delivery mechanism of ibuprofen drug molecules in MOF, which provides theoretical guidance for drug delivery materials. calculation methods. In this study, typical NPP assessment models are systematically reviewed according to the classification of climate productivity model, physiological and ecological process model, and light energy use model. Focusing on the structure and driving parameters of each model, this paper discusses and analyzes the characteristics and applicability of each model, and makes an overview of the key issues of the models-development, pointing out that future research needs to integrate the perspectives of multiple disciplines, give full play to the advantages of the new earth observation technology, and further deepen the scale conversion related to NPP assessment.
• Select
  A machine Learning-Big Data Study on High-Energy Metal-Organic frameworks for chiral drug Ibuprofen resolution and transportation

  2024, 23(5): 25-24.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Drug separation / drug loading materials have become one of the significant research objects in controlled drug release and drug preparation technology. In order to select candidates for efficient drug loading from a large number of existing materials and explore their loading mechanisms, 1 000 MOFs materials were extracted from the CoRE-MOF 2019 database for this study, and their adsorptive loading performance the drug ibuprofen was explored by high-throughput calculations. Firstly, the eight structure / energy descriptors of MOFs were analyzed by univariate analysis with the adsorption selectivity ( SSIBU/ N2 ) , adsorption capacity ( NSIBU ) and tradeoff value ( TSN) of MOFs for ibuprofen drug molecules, and the relationship trend between different descriptors and the three performance indicators were preliminarily explored. Secondly, six machine learning models including Random For est, Extreme Gradient Boosting ( XGB) , Gradient Boosting Decision Tree, Light Gradient Boosting Machine, Backpropagation Neural Network, and Support Vector Regression six machine learning algorithms with eight descriptors and three performance evaluation criteria ( Adsorption selectivity SSIBU/ N2 ,Adsorption capacity NSIBU/ N2 and Tradeoff value TSN) for big data training and mining, were used to establish quantitative relationships. The results show that the prediction accuracy of the six ML algorithms is N > TSN > S . For S , XGB showed the best prediction ( R2 = 0。83) . Subsequently, based on the XGB model, the SHaple Additive explanation ( SHAP) method was used to ex plain and analyze the importance of MOF descriptors to performance indicators. The total energy generated during MOF adsorption is considered to be the key influencing factor, and it shows a positive correlation trend with both TSN and NSIBU . Finally, combined with toxicological analysis, a series of high-performance MOF materials were recommended and designed. This work, from molecular level, high-throughput computing to big data mining, systematically studied the adsorption and delivery mechanism of ibuprofen drug molecules in MOF, which provides theoretical guidance for drug delivery materials.
• Select
  Process simulation and energy saving analysis of isopropanol production by transesterification method

  2024, 23(5): 37-36.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Based on the thermodynamic method, NRTLRK, Aspen plus software was used to simulate isopropyl alcohol production process by transesterification and optimized design of energy saving, which is catalyzed by sodium methanol solid catalyst. The project used 50 000 tons isopropyl alcohol production per year, while the factors of reaction equilibrium, separation difficulty and cost of raw materials are considered for reactor design. The reactive distillation column is set as isopropyl alcohol synthesis reactor. The bottoms flow is 99.99 wt% of isopropyl alcohol. The pressure swing distillation system design is based on the thermal characteristics of flow from the top of the reactive distillation column, which can achieve the goals of recycle the byproducts, methyl acetate, and the raw material, methanol. Heat pump distillation and double effect distillation technologies are adopted in reactive distillation and pressure swing distillation systems, respectively. While the energy consumption values decrease 60.2% and 43.9% respectively, the heat exchange network energy was further integrated, and the final energy consumption was reduced by 59.2% .
• Select
  Local R-optimal design for a mixture exponential decay model

  2024, 23(5): 48-47.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  The mixture exponential decay model plays an important role in pharmacokinetics and chemical kinetics. Different from the previous research method which was transformed into a linear model according to Taylor′s development, based on the theory of R-optimal design, this paper uses a nonlinear optimization method, Shengjin formula and interior point method to deduce R-optimal design under the model when the two decay parameters are equal. Meanwhile, the R-optimal design algorithm is given when the two decay parameters are unequal. It is verified by the equivalence theorem that the derived design is R-optimal.
• Select
  A portmanteau test for quasi maximum exponential likehood estimation of GARCH model based on high frequency data

  2024, 23(5): 54-53.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Previous studies have shown that the quasi maximum exponential likelihood estimation based on high frequency data can improve the estimation accuracy of GARCH model, but few studies have derived the corresponding test statistic for this estimator. In this paper, a portmanteau Q test statistic is proposed based on the asymptotic property of quasi maximum exponential likelihood estimation of GARCH model based on high-frequency data. The theoretical correctness of the test statistic is vali dated through simulation in this paper, and specific applications are provided by using the data of the CSI 300, CSI 500, and SSE 50 indices. The results show that when the model is adequate, the distribution of the test statistic proposed in this paper more closely follows the theoretically derived distribution, which is better than the results of the test statistic based on low-frequency data. Moreover, the statistic is able to capture high-frequency residual autocorrelation due to the inclusion of high-frequency information. While for low-frequency residual autocorrelation, the statistic can also identify model non-sufficiency when the correlation is stronger, which is useful for order identification in GARCH model. Empirical research also indicates that the test statistic can identify the effective utilization of high-frequency information by the models based on high-frequency data, demonstrating a certain degree of practicality.
• Select
  Spatial average and its estimation in the case of missing data

  2024, 23(5): 69-68.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Spatial averages have been widely used in scientific research and practical applications since they were proposed. How to estimate the weight coefficient in the spatial average has been a problem studied by many scholars. In the case of missing data, the spatial average is the ratio of the two random variables R = ∑ βi si ri as well as S = ∑ βi si . In this paper, we use the “ Delta method”to derive approximate formulas for the squared bias, variance, and mean squared error of the estimatorr used to estimate the true spatial average, assuming that the weight βi is known. The bias of the estimator r and the source of variance are analyzed, and the spatial average estimated weight to minimize the bias is given. Finally, the obtained results are applied to the global relative abundance data of ammoniating archaea.
• Select
  Defense method for membership inference attacks based on diffusion model and mixed samples

  2024, 23(5): 76-75.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  Membership inference attacks in deep learning refer to inferring whether a given sample belongs to the training dataset of a target model. Due to the presence of privacy-sensitive information in the training dataset, defending against membership inference attacks is crucial for privacy protection. This paper begins by defining membership inference attacks and elucidating the underlying reasons causing such attacks. Subsequently, existing defense algorithms are comprehensively reviewed. Finally, a novel defense mechanism is proposed, delineating the defensive approach adopted in this paper. Compared to state-of-the-art defenses against membership inference attacks, this method offers superior trade-offs between preserving member privacy and maintaining model utility. Detailed explanations of the employed techniques are provided to facilitate a better understanding of membership inference attacks and their defenses, thereby furnishing valuable insights for mitigating privacy risks in training datasets and striking a balance between model utility and privacy security.
• Select
  The network characteristics of urban agglomerations from the perspective of multiflow: A case study of 19 urban agglomerations in China

  2024, 23(5): 85-84.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  The coordinated development of urban agglomerations is an essential requirement for achieving high-quality regional economic growth. Examining the network characteristics of Chinese urban agglomerations from a multi-dimensional perspective holds significant importance in advancing the co ordinated development of these agglomerations. This article establishes urban agglomeration information, transportation, and population networks based on data from information search indices, train schedules, and Baidu migration indices. By employing social network analysis methods, the spatial network connectivity characteristics of 19 urban agglomerations in China are examined. The research findings are as follows: ①Scale Characteristics: The ranking of network sizes among the 19 urban ag glomerations does not vary significantly across different factor flows. Higher levels of economic development correspond to larger network sizes of urban agglomerations. ②Element Connections: Urban agglomerations exhibit hierarchical and differential characteristics in their information, transportation, and population networks. The information network is primarily characterized by strong connections, while the transportation and population networks are mainly characterized by moderate to weak connections. Spatially, the information network displays a “ diamond” shaped pattern, while the transportation and population networks exhibit a “ two horizontal and three vertical” pattern. ③Cluster Characteristics: Information connections between urban agglomerations are the closest, followed by transportation and population connections. The impact of geographical proximity on different element networks varies, with the overall pattern showing a higher impact on the population network followed by the transportation network and then the information network. ④Urban Agglomeration Types: Utilizing net work indicators, the identified types of urban agglomerations are classified into four categories: radiative, siphon, balanced, and peripheral. These types demonstrate heterogeneous characteristics across different element networks. In the information network, the urban agglomerations in the eastern, central, and western regions exhibit distinct network characteristics, with a trend of “ radiation type dominance by eastern urban agglomerations and siphon type dominance by western cities” .
• Select
  The resilience network in college students and comparison between first-generation and non-first-generation college students

  2024, 23(5): 95-102.

  Abstract ( ) Download PDF ( )   Knowledge map   Save

  To examine the core features of resilience among Chinese students and to compare the differences in psychological resilience networks between first-generation and non-first-generation college students, a survey was conducted on 3 017 college students ( consisting of 2 166 first-generation college students and 851 non-first-generation college students) using the ConnorDavidson Resilience Scale. Network analysis methods were used for network model construction, centrality analysis, and network comparative analysis. The results showed that: ①The item with the highest centrality of resilience network structure was “ I can achieve my goals” ; ②There was no significant difference in the network structure and network connection strength of resilience between first-generation college students and non-first-generation college students; ③Among first-generation college students, “ I will do my best regardless of the outcome” was the item with the highest centrality, whereas among non-first-generation college students, “ I consider myself a strong person” was the item with the highest centrality. Therefore, in the future, the resilience intervention of college students can be classified and accurately implemented, and the items with the highest centrality will be used as the target for the intervention of first-generation college students and non-first-generation college students.