Search

Hyperparameter tuning is crucial in the development of machine learning models. This study introduces the nonlinear shrinking factor and the Cauchy mutation mechanism to improve the Dujiangyan Irrigation System Optimization (DISO), proposing the improved Dujiangyan Irrigation System Optimization algorithm (IDISO) for hyperparameter tuning in machine learning. The optimization capabilities and convergence performance of IDISO were validated on 87 CEC2017 benchmark functions of varying dimensions and nine real-world engineering problems, demonstrating that it significantly outperforms DISO in terms of convergence speed and accuracy, and ranks first in overall performance among the seventeen advanced metaheuristic algorithms being compared. To construct a robust and generalizable prediction model for hydrochar element characteristics, this study utilized IDISO and DISO algorithms to fine-tune the parameters of the XGBoost model. The experimental results show that the IDISO-XGBoost model achieved an average prediction performance of 0.95, which represents a 4% improvement over the DISO-XGBoost model. These results indicate that the IDISO algorithm has significant potential and value in practical applications. [ABSTRACT FROM AUTHOR]

Effective real-time autonomous navigation for mobile robots in static and dynamic environments has become a challenging and active research topic. Although the simultaneous localization and mapping (SLAM) algorithm offers a solution, it often heavily relies on complex global and local maps, resulting in significant computational demands, slower convergence rates, and prolonged training times. In response to these challenges, this paper presents a novel algorithm called PER-n2D3QN, which integrates prioritized experience replay, a noisy network with factorized Gaussian noise, n-step learning, and a dueling structure into a double deep Q-network. This combination enhances the efficiency of experience replay, facilitates exploration, and provides more accurate Q-value estimates, thereby significantly improving the performance of autonomous navigation for mobile robots. To further bolster the stability and robustness, meaningful improvements, such as target "soft" updates and the gradient clipping mechanism, are employed. Additionally, a novel and powerful target-oriented reshaping reward function is designed to expedite learning. The proposed model is validated through extensive experiments using the robot operating system (ROS) and Gazebo simulation environment. Furthermore, to more specifically reflect the complexity of the simulation environment, this paper presents a quantitative analysis of the simulation environment. The experimental results demonstrate that PER-n2D3QN exhibits heightened accuracy, accelerated convergence rates, and enhanced robustness in both static and dynamic scenarios. [ABSTRACT FROM AUTHOR]

Interior design requires designer creativity and significant workforce investments. Meanwhile, Artificial Intelligence (AI) is crucial for enhancing the creativity and efficiency of interior design. Therefore, this study proposes an innovative method to generate multistyle interior design and videos with AI. First, this study created a new indoor dataset to train an AI that can generate a specified design style. Subsequently, video generation and super-resolution modules are integrated to establish an end-to-end workflow that generates interior design videos from texture-free 3D models. The proposed method utilizes AI to produce diverse interior design videos directly, thus replacing the tedious tasks of texture selection, lighting arrangement, and video rendering in traditional design processes. The research results indicate that the proposed method can effectively provide diverse interior design videos, thereby enriching design presentation and improving design efficiency. Additionally, the proposed workflow is versatile and scalable, thus holding significant reference value for transforming traditional design toward intelligence. [ABSTRACT FROM AUTHOR]

To identify suitable potassium fertilizers for grape (Vitis vinifera L.) production and study their mechanism of action, the effects of four potassium-containing fertilizers (complex fertilizer, potassium nitrate, potassium sulfate, and potassium dihydrogen phosphate) on sugar and organic acid metabolism in grape fruits were investigated. Potassium-containing fertilizers increased the activity of sugar and organic acid metabolism-related enzymes at all stages of grape fruit development. During the later stages of fruit development, potassium-containing fertilizers increased the total soluble solid content and the sugar content of the different sugar fractions and decreased the titratable acid content and organic acid content of the different organic acid fractions. At the ripening stage of grape fruit, compared with the control, complex fertilizer, potassium nitrate, potassium sulfate, and potassium dihydrogen phosphate increased the total soluble solid content by 1.5, 1.2, 3.5, and 3.4 percentage points, decreased the titratable acid content by 0.09, 0.06, 0.18, and 0.17 percentage points, respectively, and also increased the total potassium content in grape fruits to a certain degree. Transcriptome analysis of the differentially expressed genes (DEGs) in the berries showed that applying potassium-containing fertilizers enriched the genes in pathways involved in fruit quality, namely, carbon metabolism, carbon fixation in photosynthetic organisms, glycolysis and gluconeogenesis, and fructose and mannose metabolism. Potassium-containing fertilizers affected the expression levels of genes regulating sugar metabolism and potassium ion uptake and transport. Overall, potassium-containing fertilizers can promote sugar accumulation and reduce acid accumulation in grape fruits, and potassium sulfate and potassium dihydrogen phosphate had the best effects among the fertilizers tested. [ABSTRACT FROM AUTHOR]

The materials of the Al-Ni-Ce ternary system have promising application potential in additive manufacturing, and the systematic study of Al-Ni-Ce IMCs is an important part of the design and performance development of Al-Ni-Ce alloys. In this paper, the mechanical properties and electronic properties of seven major Al-Ni-Ce IMCs are studied using first-principles calculations, and the differences in their mechanical properties are discussed. The enthalpy of formation and cohesion energy of the Al-Ni-Ce IMCs showed that Ce4Al23Ni6 had the highest enthalpy of formation and cohesion energy, which were −49.57 kJ/mol and −4.47 kJ/mol, respectively. This demonstrated that all Al-Ni-Ce IMCs had excellent thermodynamic stability. The elastic modulus calculation of the Al-Ni-Ce IMCs showed that the Young's modulus of CeAl3Ni2 was the largest at 178.15 GPa, the bulk modulus of CeAlNi4 was the largest at 125.78 GPa, and the shear modulus of CeAlNi2 was the largest at 62.53 GPa. This proved that among the Al-Ni-Ce IMCs, CeAl3Ni2 had the greatest stiffness, CeAlNi4 had the strongest resistance to uniform deformation, and CeAl3Ni2 had the strongest resistance to plastic deformation. The electronic density of states of Al-Ni-Ce IMCs was analyzed, and the results showed that the main bonding types of Al-Ni-Ce IMCs were metallic and covalent bonds. This work can provide a theoretical basis for the material design and development of the Al-Ni-Ce ternary system. [ABSTRACT FROM AUTHOR]

Green technology innovation (GTI) is critical for economic development and environmental protection. This paper investigates the influence of the National E-commerce Demonstration Cities (NEDC) policy on GTI using a multi-period Difference-in-Difference (DID) model and data from prefecture-level Chinese cities. The findings indicate that the NEDC policy considerably facilitates GTI in China. The conclusion withstands a comprehensive set of robustness tests and remains valid even after considering potential endogeneity issues. A dynamic analysis reveals an increasing influence of the NEDC policy on GTI over time. The paper identifies producer services agglomeration, internet development, and financial support as channels through which the NEDC policy affects GTI. A heterogeneity analysis demonstrates that the NEDC policy's influence on GTI is more pronounced in larger cities with a higher degree of marketization and increased levels of human capital. Moreover, the NEDC policy exhibits spatial spillover effects, supporting GTI advancement in both local cities and neighboring regions. This study provides insights into how emerging market economies can leverage e-commerce for green development. [ABSTRACT FROM AUTHOR]

The integration of culture and tourism is the key measure for China in transforming and upgrading the tourism industry. It could promote the sustainable development by reducing the consumption of tourism resources. During the implementation of this policy, culture and tourism listed companies play a leading role, as they are the major force in the cultural and tourism industry, and their spatiotemporal characteristics shows the level of development of the cultural and tourism industry and the economy in different regions. Taking the culture and tourism companies as the research objects, this paper analyzed the spatiotemporal patterns and evolution mechanism of culture and tourism listed companies over a long time scale. The results showed that: (1) the number of newly added listed companies in the cultural, tourism, and cultural and tourism categories basically showed the same change trend, and the developing process could be divided into three periods, namely embryonic (1978–1992), exploration (1992–2006) and growth (2007–2021). (2) The spatial distribution of Chinese culture and tourism listed companies developed from "single core" to "one core and multiple stars", and gradually stabilized into a spatial pattern of "three cores". The center of gravity was located in the southeast of China, moving from south to north in the shape of a "Z". (3) Traffic service, government financial strength, financial environment and tourism resource endowment are the main factors affecting the spatial distribution of Chinese culture and tourism listed companies. Compared with the impact of a single influencing factor, the interactions between various factors are much stronger, especially the combinations of tourism resource endowment and another three factors, namely traffic service, communication development and economic development. [ABSTRACT FROM AUTHOR]

As China has experienced rapid economic growth, the study of household wealth distribution has become a pressing issue. This paper uses data from the 2012–2018 China Family Panel Studies (CFPS) to examine the impact of demographic factors and urban–rural differences on household wealth distribution. The study finds that China's household wealth Gini coefficient has increased significantly, with wealthier households accumulating wealth at a much faster rate than those at the bottom. In addition, negative wealth households, hand-to-mouth households, and indebted households have all shown an upward trend. Despite small marginal effects, demographic factors such as education, average household age, elderly numbers, and household size do not significantly affect the overall trend of the wealth distribution. Both rural and urban households at the top maintain high growth rates, but the average urban household experiences faster wealth growth than its rural counterpart. Asset allocation between rural and urban households also shows significant differences. This analysis underscores the importance of examining wealth distribution to promote equitable resource allocation and economic stability. Policymakers can use the findings of this study to reduce economic disparities and achieve the goal of common prosperity in China. [ABSTRACT FROM AUTHOR]

This paper examines the effects of political hierarchy on firms' long-term total factor productivity using evidence from the elevation of Chongqing to provincial-level government in China. Using the elevation as an exogenous shock and the Sichuan-Chongqing border as a dividing line, we utilized the spatial regression discontinuity approach to identify the causal link. We found that before Chongqing's elevation to a provincial-level municipality, the TFP of Chongqing firms was not significantly different from that of Sichuan firms, but after the elevation, the TFP of Chongqing firms near the border was significantly lower than that of Sichuan firms. The mechanism analysis shows that the land transaction price in Chongqing is significantly lower than that in Sichuan Province, which leads to an abundance of low-productivity firms and the "crowding out" of high-productivity firms. We also found that government intervention in Chongqing exacerbates the degree of resource misallocation at the firm level, which together lead to a decline in firm TFP. [ABSTRACT FROM AUTHOR]

Potassium (K) is the most important element for fruit quality improvement. This study aimed at determining the best K fertilizer type that can promote grape growth and nutrient uptake. Specifically, four types of K fertilizers (complex fertilizer, potassium nitrate, potassium sulfate, and potassium dihydrogen phosphate) were applied to grapevines grown in plastic pots, and then their effects on grape growth and nutrient uptake were explored. Results showed that the complex fertilizer and potassium nitrate treatments increased the biomass of the grapevine plants, whereas the other fertilizers had no significant effects on the biomass. Only the potassium nitrate treatment increased the contents of photosynthetic pigments in grapevine leaves. The complex fertilizer and potassium nitrate treatments increased the total N content in the grapevine plants to some extent, whereas the other fertilizer treatments decreased the total N content to some extent. It was also evident that all four K fertilizers increased the total P and K contents in the grapevine plants. Compared to the control, the complex fertilizer, potassium nitrate, potassium sulfate, and potassium dihydrogen phosphate treatments increased the scion total P content by 20.18%, 9.77%, 12.52%, and 30.81%, respectively, and increased the scion total K content by 15.37%, 8.41%, 20.15%, and 26.48%, respectively. In addition, correlation and grey relational analyses showed that the rootstock stem total N content, rootstock root biomass, and soil alkali-hydrolyzable N concentration were the top three indicators most closely associated with the scion total N content, whereas the rootstock root total P content, soil available P concentration, and rootstock stem total P content were the top three indicators most closely associated with the scion total P content. Additionally, the rootstock root total K content, soil available K concentration, and rootstock root total P content were the top three indicators most closely associated with the scion total K content. Overall, the different K fertilizers can all promote the uptake of P and K by grapevine plants, and the potassium dihydrogen phosphate fertilizer is the best choice. [ABSTRACT FROM AUTHOR]

Pedestrians who suddenly cross the street from within the blind spot of a vehicle's field of view can pose a significant threat to traffic safety. The dangerous pedestrian crossing intentions in view-obscured scenarios have not received as much attention as the prediction of pedestrian crossing intentions. In this paper, we present a method for recognizing and predicting the dangerous crossing intention of pedestrians in a view-obscured region based on the interference, pose, velocity observation–long short-term memory (IPVO-LSTM) algorithm from a road-based view. In the first step, the road-based camera captures the pedestrian's image. Then, we construct a pedestrian interference state feature module, pedestrian three-dimensional pose feature module, pedestrian velocity feature module, and pedestrian blind observation state feature module and extract the corresponding features of the studied pedestrians. Finally, the pedestrian hazard crossing intention prediction module based on a feature-fused LSTM (ff-LSTM) and attention mechanism is used to fuse and process the above features in a cell state process to recognize and predict the pedestrian hazard crossing intention in the blind visual area. Experiments are compared with current common algorithms in terms of the input parameter selection, intention recognition algorithm, and intention prediction time range, and the experimental results validate our state-of-the-art method. [ABSTRACT FROM AUTHOR]

Inadequate potassium (K) availability is a common abiotic stress that limits the growth and quality of fruit trees. Few studies have investigated the physiological and molecular responses of grapes at different potassium levels. In this study, an integrated approach was developed for grapevines grown at four different potassium fertilization levels [0 (K0-CK), 150 (K150), 300 (K300), and 450 (K450) g/plant] in combination with metabolite measurements and transcript analysis. The results showed that different K levels affected the accumulation of sugars and anthocyanins in the fruit. At 78 days after bloom (DAB), the K150, K300, and K450 treatments increased soluble sugar content by 37.39%, 31.10% and 32.59%, respectively, and anthocyanin content by 49.78%, 24.10%, and 13.06%, respectively, compared to K0. Weighted gene co-expression network analysis (WGCNA) of DEGs identified a network of 11 grapevines involved. During fruit development, potassium application promoted the accumulation of anthocyanins and sugars in fruit by regulating the up-regulation of GST, AT, UFGT and SPS, HT, PK gene expressions. These results suggest that potassium deficiency inhibits anthocyanin and sugar metabolism. In addition, it promotes the up-regulation of KUP expression, which is the main cause of K accumulation in fruits. Together, our data revealed the molecular mechanism in response to different K levels during fruit quality formation and provides the scientific foundation for the improvement of fruit quality by adding K fertilizer. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic has presented unprecedented disruptions to human society worldwide since late 2019, and lockdown policies in response to the pandemic have directly and drastically decreased human socioeconomic activities. To quantify and assess the extent of the pandemic's impact on the economy of Hebei Province, China, nighttime light (NTL) data, vegetation information, and provincial quarterly gross domestic product (GDP) data were jointly utilized to estimate the quarterly GDP for prefecture-level cities and county-level cities. Next, an autoregressive integrated moving average model (ARIMA) model was applied to predict the quarterly GDP for 2020 and 2021. Finally, economic recovery intensity (ERI) was used to assess the extent of economic recovery in Hebei Province during the pandemic. The results show that, at the provincial level, the economy of Hebei Province had not yet recovered; at the prefectural and county levels, three prefectures and forty counties were still struggling to restore their economies by the end of 2021, even though these economies, as a whole, were gradually recovering. In addition, the number of new infected cases correlated positively with the urban NTL during the pandemic period, but not during the post-pandemic period. The study results are informative for local government's strategies and policies for allocating financial resources for urban economic recovery in the short- and long-term. [ABSTRACT FROM AUTHOR]

Self-healing anti-corrosion materials are widely regarded as a promising long-term corrosion protection strategy, and this is even more significant if the damage can be monitored in real-time and consequently repaired. Inspired by the hierarchical structure of human skin, self-healing, solvent-free polyurethane/carbon nanotubes composites (SFPUHE-HTF-CNTs) with a skin-like bilayer structure were constructed. The SFPUHE-HTF-CNTs were composed of two layers, namely, a hydrophobic solvent-free polyurethane (SFPUHE-HTF) containing disulfide bonds and fluorinated polysiloxane chain segments consisting of a self-healing layer and CNTs with good electrical conductivity consisting of a corrosion protection layer, which also allowed for the real-time monitoring of damage. The results of corrosion protection experiments indicated that the SFPUHE-HTF-CNTs had a low corrosion current density (8.94 × 10−9 A·cm−2), a positive corrosion potential (−0.38 V), and a high impedance modulus (|Z| = 4.79 × 105 Ω·cm2). The impedance modulus could still reach 4.54 × 104 Ω·cm2 after self-healing, showing excellent self-healing properties for anti-corrosion protection. Synchronously, the SFPUHE-HTF-CNTs exhibited a satisfactory damage sensing performance, enabling the real-time monitoring of fractures at different sizes. This work realized the effective combination of self-healing with corrosion protection and damage detection functions through a bionic design, and revealed the green, and low-cost preparation of advanced composites, which have the advantage of scale production. [ABSTRACT FROM AUTHOR]

In multiplexed sequencing, the identification of DNA sequencing barcodes can effectively reduce the probability of sample misassignment. However, the great quantity of sequence data requires a high-throughput identification method. Therefore, based on a barcode identification scheme combining cyclic shifting with dynamic programming (DP), this paper proposes, implements and tests a hardware accelerator that can accelerate barcode identification. In the accelerator, considering that the computational complexity of the DP algorithm can be expressed as the multiplication of the lengths of both involved sequences, we design a systolic array structure with simplified processing element (PE) and a parallel circuit architecture to identify the insertion and deletion errors based on the traceback. The accelerator is implemented on a field-programmable gate array (FPGA), and its performance is compared with that of software implemented on a general-purpose computer. The experimental results indicate that, compared with the software implementation, the accelerator can achieve speedups of two orders of magnitude for longer barcodes. [ABSTRACT FROM AUTHOR]

The continuous and extensive pinewood nematode disease has seriously threatened the sustainable development of forestry in China. At present, many studies have used high-resolution remote sensing images combined with a deep semantic segmentation algorithm to identify standing dead trees in the red attack period. However, due to the complex background, closely distributed detection scenes, and unbalanced training samples, it is difficult to detect standing dead trees (SDTs) in a variety of complex scenes by using conventional segmentation models. In order to further solve the above problems and improve the recognition accuracy, we proposed a new detection method called multi-scale spatial supervision convolutional network (MSSCN) to identify SDTs in a wide range of complex scenes based on airborne remote sensing imagery. In the method, a Gaussian kernel approach was used to generate a confidence map from SDTs marked as points for training samples, and a multi-scale spatial attention block was added into fully convolutional neural networks to reduce the loss of spatial information. Further, an augmentation strategy called copy–pasting was used to overcome the lack of efficient samples in this research area. Validation at four different forest areas belonging to two forest types and two diseased outbreak intensities showed that (1) the copy–pasting method helps to augment training samples and can improve the detecting accuracy with a suitable oversampling rate, and the best oversampling rate should be carefully determined by the input training samples and image data. (2) Based on the two-dimensional spatial Gaussian kernel distribution function and the multi-scale spatial attention structure, the MSSCN model can effectively find the dead tree extent in a confidence map, and by following this with maximum location searching we can easily locate the individual dead trees. The averaged precision, recall, and F1-score across different forest types and disease-outbreak-intensity areas can achieve 0.94, 0.84, and 0.89, respectively, which is the best performance among FCN8s and U-Net. (3) In terms of forest type and outbreak intensity, the MSSCN performs best in pure pine forest type and low-outbreak-intensity areas. Compared with FCN8s and U-Net, the MSSCN can achieve the best recall accuracy in all forest types and outbreak-intensity areas. Meanwhile, the precision metric is also maintained at a high level, which means that the proposed method provides a trade-off between the precision and recall in detection accuracy. [ABSTRACT FROM AUTHOR]

A long-strip differential interferometric synthetic aperture radar (DInSAR) measurement based on multi-frame image mosaicking is currently the realizable approach to measure large-scale ground deformation. As the spatial range of the mosaicked images increases, the nonlinear variation of ground ocean tidal loading (OTL) displacements is more significant, and using plane fitting to remove the large-scale errors will produce large tidal displacement residuals in a region with a complex coastline. To conveniently evaluate the ground tidal effect on mosaic DInSAR interferograms along the west coast of the U.S., a three-dimensional ground OTL displacements grid is generated by integrating tidal constituents' estimation of the GPS reference station network and global/regional ocean tidal models. Meanwhile, a solid earth tide (SET) model based on IERS conventions is used to estimate the high-precision SET displacements. Experimental results show that the OTL and SET in a long-strip interferogram can reach 77.5 mm, which corresponds to a 19.3% displacement component. Furthermore, the traditional bilinear ramp fitting methods will cause 7.2~20.3 mm residual tidal displacement in the mosaicked interferograms, and the integrated tidal constituents displacements calculation method can accurately eliminate the tendency of tidal displacement in the long-strip interferograms. [ABSTRACT FROM AUTHOR]

This paper describes an experimental study of the cooling capabilities of microchannel and micro-pin-fin based on-chip cooling systems. The on-chip cooling systems integrated with a micro heat sink, simulated power IC (integrated circuit) and temperature sensors are fabricated by micromachining and silicon-to-silicon direct bonding. Three micro heat sink structures: a microchannel heat sink (MCHS), an inline micro-pin-fin heat sink (I-MPFHS) and a staggered micro-pin-fin heat sink (S-MPFHS) are tested in the Reynolds number range of 79.2 to 882.3. The results show that S-MPFHS is preferred if the water pump can provide enough pressure drop. However, S-MPFHS has the worst performance when the rated pressure drop of the pump is lower than 1.5 kPa because the endwall effect under a low Reynolds number suppresses the disturbance generated by the staggered micro pin fins but S-MPFHS is still preferred when the rated pressure drop of the pump is in the range of 1.5 to 20 kPa. When the rated pressure drop of the pump is higher than 20 kPa, I-MPFHS will be the best choice because of high heat transfer enhancement and low pressure drop price brought by the unsteady vortex street. [ABSTRACT FROM AUTHOR]

A large amount of research has been devoted to developing novel superhydrophobic coatings. However, it is still a great challenge to pursuean environmentally friendly method that leads to superhydrophobic coatings. Herein, we demonstrate for the first time, an environmentally friendly method for the preparation of conductive superhydrophobic coatings with sandwich-like structures by using aminoethylaminopropyl polydimethylsiloxane modified waterborne polyurethane (SiWPU) and N-octadecylamine functionalized multi-wall carbon nanotubes. These environmentally friendly coatings with the sheet resistance of 1.1 ± 0.1 kΩ/sq exhibit a high apparent contact angle of 158.1° ± 2° and a low sliding angle below 1°. The influence of the surface texture before and after heat treatment on the wetting properties is discussed. In addition, the coatings can be electrically heated by 3~113 °C with a voltage of 12~72 V, and thus, can be used for deicing. Furthermore, the resulting coatings demonstrate good performance of wear resistance and ultraviolet resistance, which will have broad application potential in harsh environments. [ABSTRACT FROM AUTHOR]

By installing FBG sensors on the geogrids, smart geogrids can both reinforce and monitor the stability for geogrid-reinforced slopes. In this paper, a geogrid-reinforced sand slope model test is conducted in the laboratory and fiber Bragg grating (FBG) sensing technology is used to measure the strain distribution of the geogrid. Based on the model test, the performance of the reinforced soil slope is simulated by finite element software Midas-GTS, and the stability of the reinforced soil slope is analyzed by strength reduction method. The relationship between the geogrid strain and the factor of safety is set up. The results indicate that the measured strain and calculated results agree very well. The geogrid strain measured by FBG sensor can be applied to evaluate the stability of geogrid-reinforced sand slopes.