Search Results (47,757)

For several years, a growing interest among numerous researchers and investors in predicting stock price movements has spurred extensive exploration into employing advanced deep learning models. These models aim to develop systems capable of comprehending the stock market’s complex nature. Despite the immense challenge posed by the diverse factors influencing stock price forecasting, there remains a notable lack of research focused on identifying the essential feature set for accurate predictions. In this study, we propose a Dynamic Feature Selection System (DFSS) to predict stock prices across the 10 major industries, as classified by the FnGuide Industry Classification Standard (FICS) in South Korea. We apply 16 feature selection algorithms from filter, wrapper, embedded, and ensemble categories. Subsequently, we adjust the settings of industry-specific index data to evaluate the model’s performance and robustness over time. Our comprehensive results identify the optimal feature sets that significantly impact stock prices within each sector at specific points in time. By analyzing the inclusion ratios and significance of the optimal feature set by category, we gain insights into the proportion of feature classes and their importance. This analysis ensures the interpretability and reliability of our model. The proposed methodology complements existing methods that do not consider changes in the types of variables significantly affecting stock prices over time by dynamically adjusting the input variables used for learning. The primary goal of this study is to enhance active investment strategies by facilitating the creation of diversified portfolios for individual stocks across various sectors, offering robust models and feature sets that consistently demonstrate high performance across industries over time. Full article

The asymptotically safe gravity is investigated in the framework of the functional renormalization group method. The low energy region of the model can account for the cosmological behavior, where it is assumed that the nonlocal effects play a crucial role. Using the Wegner–Houghton equation it is shown that the dynamically induced bilocal term modifies the infrared scaling of the model. Full article

This paper investigates the formation control problem of multi-unmanned aerial vehicle (UAV) systems with multi-operating modes. While mode switching enhances the flexibility of multi-UAV systems, it also introduces dynamic model switching behaviors in UAVs. Moreover, obtaining an accurate dynamic model for a multi-UAV system is challenging in practice. In addition, communication link failures and time-varying unknown disturbances are inevitable in multi-UAV systems. Hence, to overcome the adverse effects of the above challenges, a hybrid iterative learning formation control strategy is proposed in this paper. The proposed controller does not rely on precise modeling and exhibits its learning ability by utilizing historical input–output data to update the current control input. Furthermore, two convergence theorems are proven to guarantee the convergence of state, disturbance estimation, and formation tracking errors. Finally, three simulation examples are conducted for a multi-UAV system consisting of four quadrotor UAVs under multi-operating modes, switching topologies, and external disturbances. The results of the simulations show the strategy’s effectiveness and superiority in achieving the desired formation control objectives. Full article

This study presents an efficient vibration-based fault detection method for rotating machines utilising the poly-coherent composite spectrum (pCCS) and machine learning techniques. pCCS combines vibration measurements from multiple bearing locations into a single spectrum, retaining amplitude and phase information while reducing background noise. The use of pCCS significantly reduces the number of extracted parameters in the frequency domain compared to using individual spectra at each measurement location. This reduction in parameters is crucial, especially for large industrial rotating machines, as processing and analysing extensive datasets demand significant computational resources, increasing the time and cost of fault detection. An artificial neural network (ANN)-based machine learning model is then employed for fault detection using these reduced extracted parameters. The methodology is developed and validated on an experimental rotating machine at three different speeds: below the first critical speed, between the first and second critical speeds, and above the second critical speed. This range of speeds represents the diverse dynamic conditions commonly encountered in industrial settings. This study examines both healthy machine conditions and various simulated fault conditions, including misalignment, rotor-to-stator rub, shaft cracks, and bearing faults. By combining the pCCS technique with machine learning, this study enhances the reliability, efficiency, and practical applicability of fault detection in rotating machines under varying dynamic conditions and different machine conditions. Full article

The drain pipe wrapped in steel wire mesh serves a dual purpose of drainage and reinforcement in tailings pond projects. The self-filtering layer that develops upstream of the steel wire mesh influences the reinforcement characteristics of the drainage pipe. This study first conducts interfacial shearing experiments to explore the impact of the self-filtering layer on the shearing properties between tailings and the steel wire mesh. An exponential interface constitutive model is then proposed to delineate the shear stress–displacement relationship. Finally, through finite element simulations, the study assesses the effect of the self-filtering layer on the load-bearing behavior of the drain pipe, considering the interactive dynamics between the tailings and the steel wire mesh. The results reveal that the interfacial shear strength, across varying median particle sizes of the self-filtering layer, adheres to Mohr–Coulomb strength theory. Specifically, as the median particle size of the self-filtering layer increases, interfacial cohesion diminishes while the friction coefficient rises. The initial shear stiffness demonstrates a linear increase with the median particle size. With the presence of the self-filtering layer, the pull-out resistance of the drainage pipe can be enhanced by up to 26%. Moreover, the self-filtering layer significantly affects the distribution of negative skin friction. This research enhances the safety assessment of tailings ponds by providing crucial insights and solutions, emphasizing the influence of the self-filtering layer on the bearing behavior of the drain pipe. Full article

Amidst the global pursuit of environmental sustainability, the concurrent trends of decarbonization and intelligentization have emerged as critical strategic priorities. However, the interplay between these phenomena, particularly within the logistics sector, remains an underexplored area. This study investigates the complex dynamics between logistics intelligentization and decarbonization across 29 provinces in China from 2006 to 2019, providing a fresh perspective on a globally relevant issue. By employing Differential GMM, Systematic GMM, and instrumental variable-based GMM models, we evaluate the impact of logistics intelligence on carbon emissions while effectively addressing endogeneity issues inherent in the empirical analysis. Our findings reveal that the advancement of intelligent logistics correlates positively with increased carbon emissions, highlighting a significant incompatibility between decarbonization objectives and intelligentization efforts within the logistics sector. Additionally, we identify mediating pathways—specifically scale, structure, and technology effects—and moderating mechanisms that influence this relationship. These insights underscore the need for policymakers to establish environmental standards for intelligent technology adoption and to promote initiatives that reconcile intelligentization with sustainability goals. Ultimately, our study provides critical guidance for promoting sustainable and smart industrial growth in the logistics sector. Full article

Satellite navigation is an essential component of the national infrastructure. Space weather and ionospheric conditions are the prime sources of GNSS (global navigation satellite system) positioning, navigation, and timing (PNT) service disruptions and degradations. Protection, toughening, and augmentation (PTA) of GNSS PNT services require novel approaches in ionospheric effects mitigation. Standard global ionospheric correction models fail in the mitigation of high-dynamics and local ionospheric disturbances. Here, we demonstrate that in the case of the short-term fast-developing geomagnetic storm, a machine learning-based environment-aware GNSS ionospheric correction model for sub-equatorial regions may provide a substantial improvement over the existing global Klobuchar model, considered a benchmark. The proposed machine learning-based model utilises just the geomagnetic field density component observations as a predictor to estimate TEC/GNSS ionospheric delay as the prediction model target. Further research is needed to refine the methodology of machine learning model development selection and validation and to establish an architecture-agnostic framework for GNSS PTA development. Full article

The reducer serves as a pivotal component within the power transmission system of electric vehicles. On one hand, it bears the torque load within the power transmission system. On the other hand, it also endures the vibration load transmitted from other vehicle components. Over extended periods, these dynamic loads can cause fatigue damage to the reducer. Therefore, the reliability and durability of the reducer during use are very important for electric vehicles. In order to save time and economic costs, the durability of the reducer is often evaluated through accelerated fatigue testing. However, traditional approaches to accelerated fatigue tests typically only consider the time-domain characteristics of the load, which limits precision and reliability. In this study, an accelerated fatigue test method for electric vehicle reducers based on the SVR–FDS method is proposed to enhance the testing process and ensure the reliability of the results. By utilizing the support vector regression (SVR) model in conjunction with the fatigue damage spectrum (FDS) approach, this method offers a more accurate and efficient way to evaluate the durability of reducers. It has been proved that this method significantly reduces the testing period while maintaining the necessary level of test reliability. The accelerated fatigue test based on the SVR–FDS method represents a valuable approach for assessing the durability of electric vehicle reducers and offering insights into their long-term performance. Full article

Based on different carbon quota trading mechanisms, the price and emission reduction strategies of oligopoly manufacturers in the low-carbon market and the government carbon quota mechanism are considered. A dynamic game evolution model of the two oligopoly manufacturers with competitive relations is established. The stability of the equilibrium point of the game model, the price adjustment speed of the decision variable, the impact of carbon emission reduction investment, and the government carbon quota on the system are discussed. Through nonlinear dynamics research, it is found that the advantage of the grandfathering method is that it is conducive to maintaining market stability when the government’s carbon quota decision changes; the advantage of the benchmarking method is that when manufacturers formulate price adjustment strategies, the benchmarking method carbon quota mechanism has a stronger stability range for the market, the manufacturer’s profit price adjustment speed is positively correlated, and the government carbon quota decision and emission reduction investment are also positively correlated. Decision makers need to choose appropriate carbon quota mechanisms and manufacturers’ emission reduction strategies according to actual market changes to maintain supply chain stability. Full article

With the continuous increase in the proportion of distributed photovoltaic power stations, the demand for photovoltaic power grid connection is becoming more and more urgent, and the requirements for the accuracy of regional distributed photovoltaic power forecasting are also increasing. A distributed regional photovoltaic power prediction model based on a stacked ensemble algorithm is proposed here. This model first uses a graph attention network (GAT) to learn the structural features and relationships between sub-area photovoltaic power stations, dynamically calculating the attention weights of the photovoltaic power stations to capture the global relationships and importance between stations, and selects representative stations for each sub-area. Subsequently, the CNN-LSTM-multi-head attention parallel multi-channel (CNN-LSTM-MHA (PC)) model is used as the basic model to predict representative stations for sub-areas by integrating the advantages of both the CNN and LSTM models. The predicted results are then used as new features for the input data of the meta-model, which finally predicts the photovoltaic power of the large area. Through comparative experiments at different seasons and time scales, this distributed regional approach reduced the MAE metric by a total of 22.85 kW in spring, 17 kW in summer, 30.26 kW in autumn, and 50.62 kW in winter compared with other models. Full article

To satisfy the requirements of five-axis processing quality, this article improves and optimizes the machine tool structure design to produce improved dynamic characteristics. This study focuses on the investigation of five-axis machine tools’ static and dynamic stiffness as well as structural integrity. We also include performance optimization and experimental verification. We use the finite element approach as a structural analysis tool to evaluate and compare the individual parts of the machine created in this study, primarily the saddle, slide table, column, spindle head, and worktable. We discuss the precision of the machine tool model and relative space distortion at each location. To meet the requirements of the actual machine, we optimize the structure of the five-axis machine tool based on the parameters and boundary conditions of each component. The machine’s weight was 15% less than in the original design model, the material it was subjected to was not strained, and the area of the structure where the force was considerably deformed was strengthened. We evaluate the AM machine’s impact resistance to compare the vibrational deformation observed in real time with the analytical findings. During modal analysis, all the order of frequencies were determined to be 97.5, 110.4, 115.6, and 129.6 Hz. The modal test yielded the following orders of frequencies: 104, 118, 125, and 133 Hz. Based on the analytical results, the top three order error percentages are +6.6%, +6.8%, +8.1%, and +2.6%. In ME’scope, the findings of the modal test were compared with the modal assurance criteria (MAC) analysis. According to the static stiffness analysis’s findings, the main shaft and screw have quite substantial major deformations, with a maximum deformation of 33.2 µm. Force flow explore provides the relative deformation amount of 26.98 µm from the rotating base (C) to the tool base, when a force of 1000 N is applied in the X-axis direction, which is more than other relative deformation amounts. We also performed cutting transient analysis, cutting spectrum analysis, steady-state thermal analysis, and analysis of different locations of the machine tool. All of these improvements may effectively increase the stiffness of the machine structure as well improve the machine’s dynamic characteristics and increases its machining accuracy. The topology optimization method checks how the saddle affects the machine’s stability and accuracy. This research will boost smart manufacturing in the machine tool sector, leading to notable advantages and technical innovations. Full article

In addition to technical issues related to the instruments used, differences between soil moisture (SM) measured using ground-based methods and microwave remote sensing (RS) can be related to the main features of the study areas, which are intricately connected to hydraulic–hydrological conditions and soil properties. When long-term analysis is performed, these discrepancies are mitigated by the contribution of SM seasonality and are only evident when high-frequency variations (i.e., signal anomalies) are investigated. This study sought to examine the responsiveness of SM to seasonal variations in terrestrial ecoregions located in areas covered by the in situ Romanian Soil Moisture Network (RSMN). To achieve this aim, several remote sensing-derived retrievals were considered: (i) NASA’s Soil Moisture Active and Passive (SMAP) L4 V5 model assimilated product data; (ii) the European Space Agency’s Soil Moisture and Ocean Salinity INRA–CESBIO (SMOS-IC) V2.0 data; (iii) time-series data extracted from the H115 and H116 SM products, which are derived from the analysis of Advanced Scatterometer (ASCAT) data acquired via MetOp satellites; (iv) Copernicus Global Land Service SSM 1 km data; and (v) the “combined” European Space Agency’s Climate Change Initiative for Soil Moisture (ESA CCI SM) product v06.1. An initial assessment of the performance of these products was conducted by checking the anomaly of long-term fluctuations, quantified using the Absolute Variation of Local Change of Environment (ALICE) index, within a time frame spanning 2015 to 2020. These correlations were then compared with those based on raw data and anomalies computed using a moving window of 35 days. Prominent correlations were observed with the SMAP L4 dataset and across all ecoregions, and the Balkan mixed forests (646) exhibited strong concordance regardless of the satellite source (with a correlation coefficient R_ALICE > 0.5). In contrast, neither the Central European mixed forests (No. 654) nor the Pontic steppe (No. 735) were adequately characterized by any satellite dataset (R_ALICE < 0.5). Subsequently, the phenological seasonality and dynamic behavior of SM were computed to investigate the effects of the wetting and drying processes. Notably, the Central European mixed forests (654) underwent an extended dry phase (with an extremely low p-value of 2.20 × 10⁻¹⁶) during both the growth and dormancy phases. This finding explains why the RSMN showcases divergent behavior and underscores why no satellite dataset can effectively capture the complexities of the ecoregions covered by this in situ SM network. Full article

In this paper, a fast positioning platform (FPP) is proposed, able to meet simultaneously the requirements of large stroke and high frequency response, developed based on a PZT (piezoelectric actuator) and a quad-parallel flexible mechanism, for application in precision machining. The FPP is driven by a high-stiffness PZT and guided by a flexible hinge-based mechanism with a quad-parallel flexible hinge. The proposed quad-parallel flexible hinge mechanism can provide excellent planar motion capability with high stiffness and good guiding performance, thus guaranteeing outstanding dynamics characteristics. The mechanical model was established, the input and output characteristics of the FPP were analyzed, and the working range (output displacement and frequency) of the FPP was determined. Based on the mechanical model and the input and output characteristics of the FPP, the design method is described for of the proposed FPP, which is capable of achieving a large stroke while responding at a high frequency. The characteristics of the FPP were investigated using finite element analysis (FEA). Experiments were conducted to examine the performance of the FPP; the natural frequency of the FPP was 1315.6 Hz, while the maximum output displacement and the motion resolution of the FPP in a static state were 53.13 μm and 5 nm, respectively. Step response testing showed that under a step magnitude of 50 μm, the stabilization times for the falling and rising edges of the moving platform were 37 ms and 26 ms, respectively. The tracking errors were about ±1.96 μm and ±0.59 μm when the amplitude and frequency of the signal were 50 μm, 50 Hz and 10 μm, 200 Hz, respectively. The FPP showed excellent performance in terms of fast response and output displacement. The cutting test results indicated that compared with the uncontrolled condition, the values of surface roughness under controlled conditions decreased by 23.9% and 12.7% when the cutting depths were 5 μm and 10 μm, respectively. The developed FPP device has excellent precision machining performance. Full article

Due to the dynamic complexities of the multi-unmanned vessel target assignment problem at sea, especially when addressing moving targets, traditional optimization algorithms often fail to quickly find an adequate solution. To overcome this, we have developed a multi-agent reinforcement learning algorithm. This approach involves defining a state space, employing preferential experience replay, and integrating self-attention mechanisms, which are applied to a novel offshore unmanned vessel model designed for dynamic target allocation. We have conducted a thorough analysis of strike positions and times, establishing robust mathematical models. Additionally, we designed several experiments to test the effectiveness of the algorithm. The proposed algorithm improves the quality of the solution by at least 30% in larger scale scenarios compared to the genetic algorithm (GA), and the average solution speed is less than 10% of the GA, demonstrating the feasibility of the algorithm in solving the problem. Full article

The dynamic modulus of in-service asphalt pavements serves as a critical parameter for the computation of residual life and the design of overlays. However, its acquisition is currently limited to laboratory dynamic modulus testing using a limited number of core samples, necessitating a reassessment of its representativeness. To facilitate the prediction of dynamic modulus design parameters through Falling Weight Deflectometer (FWD) back-calculated modulus data, an integrated approach encompassing FWD testing, modulus back-calculation, core sample dynamic modulus testing, and asphalt DSR testing was employed to concurrently acquire dynamic modulus at identical locations under varying temperatures and frequencies. Dynamic modulus prediction models for in-service asphalt pavements were developed utilizing fundamental model deduction and gene expression programming (GEP) techniques. The findings indicate that GEP exhibits superior efficacy in the development of dynamic modulus prediction models. The dynamic modulus prediction model developed can enhance both the precision and representativeness of asphalt pavement’s dynamic modulus design parameters, as well as refine the accuracy of residual life estimations for in-service asphalt pavements. Concurrently, the modulus derived from FWD back-calculation can be transmuted into the dynamic modulus adhering to a uniform standard criterion, facilitating the identification of problematic segments within the asphalt structural layer. This is of paramount importance for the maintenance or reconstruction of in-service asphalt pavements. Full article