Search Results (17,463)

This paper presents a novel methodology for short-term load forecasting in the context of significant shifts in the daily load curve due to the rapid and extensive adoption of Distributed Energy Resources (DERs). The proposed solution, built upon the Similar Days Method (SDM) and Artificial Neural Network (ANN), introduces several novelties: (1) selection of similar days based on hidden representations of day data using Autoencoder (AE); (2) enhancement of model generalization by utilizing a broader set of training examples; (3) incorporating the relative importance of training examples derived from the similarity measure during training; and (4) mitigation of the influence of outliers by applying an ensemble of ANN models trained with different data splits. The presented AE configuration and procedure for selecting similar days generated a higher-quality training dataset, which led to more robust predictions by the ANN model for days with unexpected deviations. Experiments were conducted on actual load data from a Serbian electrical power system, and the results were compared to predictions obtained by the field-proven STLF tool. The experiments demonstrated an improved performance of the presented solution on test days when the existing STLF tool had poor predictions over the past year. Full article

The dissolved oxygen concentration (DOC) is an important indicator of water quality. Accurate DOC predictions can provide a scientific basis for water environment management and pollution prevention. This study proposes a hybrid DOC forecasting framework combined with Variational Mode Decomposition (VMD), a convolutional neural network (CNN), a Gated Recurrent Unit (GRU), and the Beluga Whale Optimization (BWO) algorithm. Specifically, the original DOC sequences were decomposed using VMD. Then, CNN-GRU combined with an attention mechanism was utilized to extract the key features and local dependency of the decomposed sequences. Introducing the BWO algorithm solved the correction coefficients of the proposed system, with the aim of improving prediction accuracy. This study used 4-h monitoring China urban water quality data from November 2020 to November 2023. Taking Lianyungang as an example, the empirical findings exhibited noteworthy enhancements in performance metrics such as MSE, RMSE, MAE, and MAPE within the VMD-BWO-CNN-GRU-AM, with reductions of 0.2859, 0.3301, 0.2539, and 0.0406 compared to a GRU. These results affirmed the superior precision and diminished prediction errors of the proposed hybrid model, facilitating more precise DOC predictions. This proposed DOC forecasting system is pivotal for sustainably monitoring and regulating water quality, particularly in terms of addressing pollution concerns. Full article

In order to solve the problem of complex coupling characteristics between multivariate load sequences and the difficulty in accurate multiple load forecasting for integrated renewable energy systems (IRESs), which include low-carbon emission renewable energy sources, in this paper, the TCN-FECAM-Informer multivariate load forecasting model is proposed. First, the maximum information coefficient (MIC) is used to correlate the multivariate loads with the weather factors to filter the appropriate features. Then, effective information of the screened features is extracted and the frequency sequence is constructed using the frequency-enhanced channel attention mechanism (FECAM)-improved temporal convolutional network (TCN). Finally, the processed feature sequences are sent to the Informer network for multivariate load forecasting. Experiments are conducted with measured load data from the IRES of Arizona State University, and the experimental results show that the TCN and FECAM can greatly improve the multivariate load prediction accuracy and, at the same time, demonstrate the superiority of the Informer network, which is dominated by the attentional mechanism, compared with recurrent neural networks in multivariate load prediction. Full article

The micro-machining scale effect makes it challenging to forecast and control the process parameters of the micro-milling process, which makes the micro-flanking-milling of weak-rigidity micro-thin-walled parts prone to deformation. To determine the critical cutting parameters for chip formation in the micro-milling of curved thin-walled parts at the mesoscale, the strain-softening effect of titanium alloy during high-speed milling and the scale effect of mesoscale cutting were comprehensively considered and a finite element prediction model for curved micro-thin-wall micro-milling was established to determine the critical milling parameters for effective material removal. Based on the determined milling parameters, an experimental design of response surface optimization was carried out. Based on the response surface methodology, a data-driven quantitative model with milling process parameters as design variables and deformation amounts as response variables was established to reveal the influence mechanism of multiple milling process parameters on machining accuracy. Based on the process requirements for deformation control in the micro-milling of curved thin-walled structures, dynamic optimization of the milling process parameters was performed using an improved NSGA-III algorithm to obtain non-dominated solutions. A visual ranking and a determination of the unique solution were conducted using the entropy weight–TOPSIS method. Finally, micro-milling validation experiments were carried out using the optimal parameter combination. The optimal solution for the process parameters of the arc-shaped micro-thin-wall micro-milling of titanium alloy established by the institute provides a relevant reference and guidance for mesoscale arc-shaped thin-wall micro-milling. Full article

As urban spaces expand, changes in land use significantly affect the structure and function of urban ecosystems, particularly with challenges such as green space reduction and uneven distribution. This study focused on the central urban area of Chengdu, China, simulating and forecasting various urban development scenarios for 2035, including cultivated land protection (CP), economic development (ED), ecological priority (EP), and natural development (ND). The construction of green space ecological networks followed a systematic process, incorporating key methods such as ecological source identification, landscape resistance surface construction, and ecological corridor extraction. The connectivity of these ecological networks was assessed using the space syntax. The results indicated that: (1) Construction land expanded across all scenarios, with the ED scenario having the largest area, while the EP scenario resulted in a significant increase in green space. (2) Ecological corridors were established under every scenario, with the EP scenario featuring the most extensive and well-connected network, linking urban green patches with surrounding natural areas. (3) The EP scenario’s ecological network displayed integration, choice, connectivity, and depth values that indicate the most complete and stable network structure. This study provides a comprehensive analysis of green space ecological network changes under different urban development strategies, offering valuable insights for optimizing urban green space planning and management. Full article

The Hong Kong Observatory has been using a parametric storm surge model to forecast the rise of sea level due to the passage of tropical cyclones. This model includes an offset parameter to account for the rise in sea level due to other meteorological factors. By adding the sea level rise forecast to the astronomical tide prediction using the harmonic analysis method, coastal sea level prediction can be produced for the sites with tidal observations, which supports the high water level forecast operation and alert service for risk assessment of sea flooding in Hong Kong. The Coupled Ocean-Atmosphere-Wave-Sediment Transport (COAWST) Modelling System, which comprises the Weather Research and Forecasting (WRF) Model and Regional Ocean Modelling System (ROMS), which in itself is coupled with wave model WaveWatch III and nearshore wave model SWAN, was tested with tropical cyclone cases where there was significant water level rise in Hong Kong. This case study includes two super typhoons, namely Hato in 2017 and Mangkhut in 2018, three cases of the combined effect of tropical cyclone and northeast monsoon, including Typhoon Kompasu in 2021, Typhoon Nesat and Severe Tropical Storm Nalgae in 2022, as well as two cases of monsoon-induced sea level anomalies in February 2022 and February 2023. This study aims to evaluate the ability of the WRF-ROMS-SWAN model to downscale the meteorological fields and the performance of the coupled models in capturing the maximum sea levels under the influence of significant weather events. The results suggested that both configurations could reproduce the sea level variations with a high coefficient of determination (R²) of around 0.9. However, the WRF-ROMS-SWAN model gave better results with a reduced RMSE in the surface wind and sea level anomaly predictions. Except for some cases where the atmospheric model has introduced errors during the downscaling of the ERA5 dataset, bias in the peak sea levels could be reduced by the WRF-ROMS-SWAN coupled model. The study result serves as one of the bases for the implementation of the three-way coupled atmosphere–ocean–wave modelling system for producing an integrated forecast of storm surge or sea level anomalies due to meteorological factors, as well as meteorological and oceanographic parameters as an upgrade to the two-way coupled Operational Marine Forecasting System in the Hong Kong Observatory. Full article

This study explores the trust dynamics in financial forecasting by comparing how individuals perceive the credibility of AI and human experts during significant structural market changes. We specifically examine the impact of two types of structural breaks on trust: Additive Outliers, which represent a single yet significant anomaly, and Level Shifts, which indicate a sustained change in data patterns. Grounded in theoretical frameworks such as attribution theory, algorithm aversion, and the Technology Acceptance Model (TAM), this research investigates psychological responses to AI and human advice under uncertainty. This experiment involved 157 participants, recruited via Amazon Mechanical Turk (MTurk), who were asked to forecast stock prices under different structural break scenarios. Participants were randomly assigned to either the AI or human expert treatment group, and the experiment was conducted online. Through this controlled experiment, we find that, while initial trust levels in AI and human experts are comparable, the credibility of advice is more severely compromised following a structural break in the Level Shift condition, compared to the Additive Outlier condition. Moreover, the decline in trust is more pronounced for human experts than for AI. These findings highlight the psychological factors influencing decision making under uncertainty and offer insights into the behavioral responses to AI and human expert systems during structural market changes. Full article

Cunninghamia lanceolata (Lamb.) Hook accounts for 12% of the total forest area in southern China, second only to Masson pine forests, and is an important part of the forest landscape in this region, which has a significant impact on the overall forest structure in southern China. In this study, we used kernel density analysis, landscape index calculation, variance test, and Markov prediction to analyze and forecast the evolution trend of landscape pattern in the central area of C. lanceolata in ten years. The objective is to investigate the change trend of the spatial pattern of C. lanceolata landscape in the long time series and its possible impact on zonal vegetation, as well as the macro-succession trend of C. lanceolata under the current social and economic background, and to make a scientific and reasonable prediction of its future succession trend. The current and future forecast results show that the landscape fragmentation degree of C. lanceolata is intensified, the erosion of bamboo forest is continuously intensified, and the landscape quality is continuously low. These results provide a reference for the future development direction of C. lanceolata and emphasize the need for targeted C. lanceolata management strategies in the future development of C. lanceolata, emphasizing the strengthening of monitoring, controlling harvesting, and managing bamboo competition in order to balance wood production with landscape quality and ecosystem stability. Full article

Oil is one of the main resources used by all countries in the world. The ever-growing demand for oil and oil products forces oil companies to increase production and refining. In order to increase net profit, oil producing companies are constantly upgrading equipment, improving oil production technologies, and preparing oil for further processing. When considering the elements of primary oil refining in difficult conditions, such as hard-to-reach or in remote locations, developers face strict limitations in energy resources and dimensions. Therefore, the use of traditional systems causes a number of difficulties, significantly reducing production efficiency. In this study, the authors solve the problem of improving the characteristics of the oil separation process. In their work, the authors analyzed the separation columns of primary oil distillation, identified the shortcomings of the technological process, and searched for technological solutions. Having identified the lack of technical solutions for monitoring the state of the temperature field of the separation column, the authors developed their own hardware–software complex for monitoring the separation column (RF patents No. 2020665473, No. 2021662752 were received). The complex was tested and successfully implemented into production. The study provides an assessment of the economic efficiency of implementation for a year and a forecast of the economic effect for 10 years. Full article

Short-term load forecasting plays an important role in power system scheduling, optimization, and maintenance, but no existing typical method can consistently maintain high prediction accuracy. Hence, fusing different complementary methods is increasingly focused on. To improve forecasting accuracy and stability, these features that affect the short-term power system are firstly extracted as prior knowledge, and the advantages and disadvantages of existing methods are analyzed. Then, three typically methods are used for short-term power load forecasting, and their interaction and complementarity are studied. Finally, the Choquet integral (CI) is used to fuse the three existing complementarity methods. Different from other fusion methods, the CI can fully utilize the interactions and complementarity among different methods to achieve consistent forecasting results, and reduce the disadvantages of a single forecasting method. Essentially, a CI with n inputs is equivalent to n! constrained feedforward neural networks, leading to a strong generalization ability in the load prediction process. Consequently, the CI-based method provides an effective way for the fusion forecasting of short-term load in power systems. Full article

This study aimed to forecast trends in swimming world records (WRs) and in the age of record holders. A total of 566 individual freestyle WRs (290 for males and 276 for females) were retrieved from open access websites. The frequency of observations in WRs in each decade and event was computed for males and females. The swimmers’ chronological age was converted into decimal age at the time of breaking the world record. ARIMA forecasting models and exponential smoothing techniques were used to examine historical trends and predict future observations. The WRs improved over time, and there was a nuanced pattern in the age of world record holders. While certain events (50 m and 100 m) showed swimmers achieving records at older ages, others (e.g., 200 m, 400 m, 800 m, and 1500 m) displayed variations. Forecasting shows a continuing improvement in WRs in the upcoming years, with the age of male world record holders stabilizing in shorter events and decreasing in longer distance ones, while for females, general stabilization should be expected for the majority of competitive events. Full article

With the ubiquitous deployment of mobile and sensor technologies in modes of transportation, taxis have become a significant component of public transportation. However, vacant taxis represent an important waste of transportation resources. Forecasting taxi demand within a short time achieves a supply–demand balance and reduces oil emissions. Although earlier studies have forwarded highly developed machine learning- and deep learning-based models to forecast taxicab demands, these models often face significant computational expenses and cannot effectively utilize large-scale trajectory sensor data. To address these challenges, in this paper, we propose a hybrid deep learning-based model for taxi demand prediction. In particular, the Variational Mode Decomposition (VMD) algorithm is integrated along with a Bidirectional Long Short-Term Memory (BiLSTM) model to perform the prediction process. The VMD algorithm is applied to decompose time series-aware traffic features into multiple sub-modes of different frequencies. After that, the BiLSTM method is utilized to predict time series data fed with the relevant demand features. To overcome the limitation of high computational expenses, the designed model is performed on the Spark distributed platform. The performance of the proposed model is tested using a real-world dataset, and it surpasses existing state-of-the-art predictive models in terms of accuracy, efficiency, and distributed performance. These findings provide insights for enhancing the efficiency of passenger search and increasing the profit of taxicabs. Full article

Tourism demand is critical for the hospitality industry and is influenced by a set of continuously changing factors. The tourism and hospitality industries play a critical role in many regions and countries, supporting the local economy, providing employment, and fostering economic and social development with effects across multiple industries. This study aims to analyse the nature of tourism and hotel demand through a thematic analysis. By conducting a review of the existing literature published over the period of 2018–2023, this research identifies overarching patterns, trends, and themes characterising the current research landscape. Research results reveal significant insights into market trends and strategic industry shifts. It particularly emphasises areas such as customer demand forecasting, technology integration, and sustainability, which are crucial for understanding demand fluctuations. The findings offer insights into the theoretical foundations of tourism and hotel demand and provide practical implications for industry stakeholders aiming to strategise effectively in a dynamic market. Full article

Forecasting carbon dioxide (CO₂) emissions has become a relevant issue. International organizations have emphasized the necessity of generating a plan to gradually reduce the concentrations of this pollutant to combat climate change. Cement industries represent one of the key sectors expected to solve this problematic. The objective of this study is to predict CO₂ emissions for North American cement industries. To achieve this, a multi-objective mathematical model is developed, integrating various machine learning algorithms. The results demonstrate a considerable improvement in accuracy metrics, with a 48.13% reduction in Mean Absolute Error achieved using the Generalized Reduced Gradient method (GRG). The forecasts reveal an increment in emissions from about 0.58 MtCO₂ every year between 2020 and 2050. The proposed framework can help decision makers and policy makers focus on the technical and logistics requirements to meet net-zero emissions targets. Full article

A convolutional neural network (CNN) has been developed to forecast the occurrence of low-visibility events or hazes in the Paris area. It has been trained and validated using multi-decadal daily regional maps of many meteorological and hydrological variables alongside surface visibility observations. The strategy is to make the machine learn from available historical data to recognize various regional weather and hydrological regimes associated with low-visibility events. To better preserve the characteristic spatial information of input features in training, two branched architectures have recently been developed. These architectures process input features firstly through several branched CNNs with different kernel sizes to better preserve patterns with certain characteristic spatial scales. The outputs from the first part of the network are then processed by the second part, a deep non-branched CNN, to further deliver predictions. The CNNs with new architectures have been trained using data from 1975 to 2019 in a two-class (haze versus non-haze) classification mode as well as a regression mode that directly predicts the value of surface visibility. The predictions of regression have also been used to perform the two-class classification forecast using the same definition in the classification mode. This latter procedure is found to deliver a much better performance in making class-based forecasts than the direct classification machine does, primarily by reducing false alarm predictions. The branched architectures have improved the performance of the networks in the validation and also in an evaluation using the data from 2021 to 2023 that have not been used in the training and validation. Specifically, in the latter evaluation, branched machines captured 70% of the observed low-visibility events during the three-year period at Charles de Gaulle Airport. Among those predicted low-visibility events by the machines, 74% of them are true cases based on observation. Full article