Search

Paper, a porous-fibrous network, is made up of hydrophilic fibres, which are notably susceptible to deformations due to variations in moisture content. The response of paper sheets to full or partial wetting, based on the water-induced fibre swelling, has mostly been predicted in the past using highly idealised two-dimensional network geometries. The purpose of the present paper is to establish how precisely the two-stage process of determining effective hygro-expansivity under uniform wetting by a two-dimensional model followed by bending analysis using a continuum model predicts the curl of a fully three-dimensional network. In the present multi-scale modelling work, the moisture-induced homogenised sheet scale curl, derived from hygro-elastic deformations using non-uniform moisture profiles, is analysed based on an idealised non-woven three-dimensional fibrous network finite element model. First, the effective paper sheet swelling properties, derived computationally for the uniform wetting condition, are compared with analytically homogenised hygro-elastic properties. Whereas the analytically predicted elastic modulus has an inaccuracy of only ∼ 5%, the hygro-expansivity coefficient is over-predicted by roughly a third for a network with straight fibres. The difference is smaller if waviness of the fibres and wrap-around in the fibre bonds are taken into account. Fibre alignment turns out to have little influence. Subsequently, paper sheet curl predictions made by the analytical model are compared with that computed from the three-dimensional network model. The inaccuracy of the fully analytical model, based on the two-dimensional network description, relative to the fully detailed three-dimensional computational model with straight fibres is on the order of 40%–50%. The fidelity of the simple model is primarily constrained by the inaccuracy of the effective hygro-expansion coefficient. If the expansivity extracted from the three-dimensional simulations under uniform wetting is inserted, the inaccuracy drops to less than 10%. The continuum bending description hence performs adequately, provided it is based on a reliable estimation of the uniform expansion behaviour of the material. • Hygroscopic curl of paper sheets may be predicted using homogenised models. • We assess the accuracy of such predictions. • As a reference simulations of an idealised three-dimensional fibre network are used. • The predictions are accurate provided an accurate hygro-expansivity is used. • Two-dimensional network models may not be sufficient for this. [ABSTRACT FROM AUTHOR]

The thin-walled tubular deployable composite boom (DCB) can realize folding and deploying functions, and it has a good application prospect in space field. This paper investigates the folding behaviour of the tubular DCB by analytical modeling. Based on the Archimedes' helix, the geometrical model of the DCB was established. By combining equilibrium equation and energy principle, an analytical model to predicted the folding moment versus rotational displacement of the DCB was presented. The failure indices in the equal-sense and opposite-sense folding processes were calculated utilizing Tsai‐Hill and maximum stress criteria. Analytical results agreed well with experimental and numerical results. At last, the influence of geometric parameters (i.e., radius, central angle and thickness of cross-section) on the folding behaviour of the DCB was further studied using the analytical model. [Display omitted] • An analytical model is presented to predict the folding behaviour of the deployable composite boom (DCB). • Failure indices are calculated using two criteria in equal-sense and opposite-sense folding processes. • The analytical model is validated against experimental and numerical results, showing good prediction accuracy. • Geometric parameters (radius, central angle, and thickness) significantly affect the folding behavior of the DCB. [ABSTRACT FROM AUTHOR]

3D human motion prediction, predicting future poses from a given sequence, is an issue of great significance and challenge in computer vision and machine intelligence, which can help machines in understanding human behaviors. Due to the increasing development and understanding of Deep Neural Networks (DNNs) and the availability of large-scale human motion datasets, the human motion prediction has been remarkably advanced with a surge of interest among academia and industrial community. In this context, a comprehensive survey on 3D human motion prediction is conducted for the purpose of retrospecting and analyzing relevant works from existing released literature. In addition, a pertinent taxonomy is constructed to categorize these existing approaches for 3D human motion prediction. In this survey, relevant methods are categorized into three categories: human pose representation , network structure design , and prediction target. We systematically review all relevant journal and conference papers in the field of human motion prediction since 2015, which are presented in detail based on proposed categorizations in this survey. Furthermore, the outline for the public benchmark datasets, evaluation criteria, and performance comparisons are respectively presented in this paper. The limitations of the state-of-the-art methods are discussed as well, hoping for paving the way for future explorations. [ABSTRACT FROM AUTHOR]

In an era of increasing energy demands and environmental concerns, optimizing energy consumption within buildings is crucial. Despite the vast improvements in HVAC and lighting systems, plug loads remain an under-studied area for enhancing building energy efficiency. This paper studies smart plug active operating mode detection, plug-level load forecasting, and plug scheduling methodologies. This research leverages a unique dataset from the University of California, San Diego, consisting of readings from over 150 smart plugs in several office buildings for more than a year, notably during the post-Covid era. This dataset is made publicly available. A comprehensive literature review on plug, i.e. , appliances operating mode detection is presented. Novel unsupervised learning approaches are applied to identify plug operating modes. A pipeline integrating the detected modes with forecasting and scheduling is developed, aiming at building energy consumption reduction. Our findings offer valuable insights and promising results into smart plug management for energy-efficient buildings. • A novel year-long dataset of power readings from 169 office smart plugs is introduced. • The first review of plug load mode detection methods and terminology is presented. • The paper proposes an unsupervised method to identify plugs load operating modes. • The proposed smart plug scheduling pipeline reduces energy consumption by 20%. [ABSTRACT FROM AUTHOR]

The quantitative prediction and control of assembly stress are crucial for the accuracy of precision optical instruments during their manufacturing process. As a typical optical component, the multipoint bonded flat lens is widely used in precision optical instruments, in which the main assembly stress (i.e., radial adhesive stress) and its induced nonuniform stress field in the lens vary with environmental parameters and time. These factors ultimately determine the accuracy and stability of such precision instruments. Effective prediction of the radial adhesive stress is important for bonding parameter optimization and accuracy enhancement but remains difficult to obtain due to the lack of direct measurement devices and effective prediction methods. Here, a novel prediction method is presented to estimate the radial adhesive stress applied on a multipoint bonded flat lenses. First, a quantitative characterization method (QCM) is developed to concisely describe the nonuniform stress field in a lens using quantitative characterization parameters (QCPs). Then, based on the QCM and the theoretical model established in our previous work (Xiong et al., 2022), a prediction method is developed for the radial adhesive stress applied on a lens by analyzing the stress field in the lens. In this method, a backpropagation neural network is constructed to obtain the mapping relationship between the key QCPs of the stress field and the radial adhesive stress applied on the lens. Finally, an experimental verification of the prediction method is carried out. This paper proposes a novel method to predict the radial adhesive stress for multipoint bonded optical lenses, which can be adopted to monitor the variation characteristics of assembly stress in optical instruments during the entire life cycle and then further employed to optimize the bonding parameters during the lens manufacturing process. This paper attempts to construct a direct mapping relationship between assembly quality and assembly parameters for precision instruments, supporting the quantitative optimization of assembly parameters and quantitative control and prediction of assembly quality during the manufacturing and working process. [Display omitted] • A quantitative characterization method to describe the stress field in a lens is proposed. • Quantitative features of the stress field in multipoint bonded lenses are revealed. • The radial adhesive stress is predicted by analyzing the stress field in the lens. • A mapping relationship between assembly quality and assembly parameter is constructed. [ABSTRACT FROM AUTHOR]

The impact of air pollution on public health is substantial, and accurate long-term predictions of air quality are crucial for early warning systems to address this issue. Air quality prediction has drawn significant attention, bridging environmental science, statistics, and computer science. This paper presents a comprehensive review of the current research status and advances in air quality prediction methods. Deep learning, a novel machine learning approach, has demonstrated remarkable proficiency in identifying complex, nonlinear patterns in air quality data, yet its application in air quality prediction is still relatively nascent. This paper also conducts a systematic analysis and summarizes how cutting-edge deep learning models are applied in air quality prediction. Initially, the historical evolution of air quality prediction methods and datasets is presented. This is followed by an examination of conventional air quality prediction techniques. A thorough comparative analysis of progress made with both traditional and deep learning-based prediction methods is provided. This review particularly focuses on three aspects: temporal modeling, spatiotemporal modeling, and attention mechanisms. Finally, emerging trends in the field of air quality prediction are identified and discussed. [ABSTRACT FROM AUTHOR]

Utilization of agricultural and food wastes for biohydrogen generation is a hot research topic. In this paper, biohydrogen generation potentials from wheat, maize, barley, rice and potato production, consumption, associated losses, and household food waste generation in Turkey is estimated till 2030 using novel semi-empirical holistic models based on different scenarios. It was estimated that up to 90 thousand tonnes of biohydrogen could be produced from household food wastes in Turkey in 2030. This amount of biohydrogen would be worth 1.5 billon US$ and generate nearly 0.8 billion US$ of net return. In addition, this study clearly shows that there is a great potential for economic gain from agricultural and food waste based biohydrogen generation in Turkey. In conclusion, the methodology developed in this study could be used by researchers around the world to estimate local and national level biohydrogen generation potentials from agricultural and food wastes. [Display omitted] • This paper examines biohydrogen potentials from agricultural and food wastes. • Holistic per capita models for estimating annual biohydrogen production potentials. • Annual production and sale costs are estimated. • Annual net returns from biohydrogen generation are calculated. • The results highlight recommendations for future research directions. [ABSTRACT FROM AUTHOR]

Fault prediction is the process of using data analysis and machine learning models to anticipate potential defects or faults in the software system. Using only the base machine learning models for software fault prediction leads to limited performance, difficulty in handling non-linear relationships and imbalanced data, inadequate feature representation, and limited complexity handling. Hence, in order to overcome these challenges, this paper proposes a new technique for the selection of classifiers that forms a heterogeneous ensemble. The main goal is to remove or trim out the classifiers that show poor performance compared to the other base classifiers, which can result into a more effective ensemble and can produce better results. The algorithm proposed in this paper finds a set of classifiers that can perform better than using all the classifiers. The challenge that was faced was how to identify the poor-performing classifiers. This challenge is dealt with by performing an experiment using different threshold values to choose the trimmed set of classifiers. For evaluation of the proposed model, 8 different benchmark software fault datasets were used, which are taken from PROMISE and the Apache repository, and AUC is used as the performance measure. The results obtained after the experimental analysis demonstrate the effectiveness of our algorithm compared to the traditional approaches, which used all the base classifiers. There is a significant increase in the AUC values for 6 datasets out of 8, while using the average of probabilities and majority voting, it was seen that there is improvement in 7 out of 8 datasets used. The best-performing dataset by using the average of probabilities is ARC, where the AUC values increase from 0.6505 to 0.694, and while using majority voting, the best-performing dataset is XALAN, where the AUC values increase from 0.5455 to 0.679. From this, it can be seen that the proposed ensemble approach achieved higher AUC values for the tested datasets when compared to the base machine learning classifiers. [ABSTRACT FROM AUTHOR]

Machine learning performs well in many problems. However, the tendency to generate predictions that violate theoretical knowledge makes it difficult to apply to practical processing. To resolve this situation, this paper combines domain knowledge with a data-driven model, proposes a theory-guided machine learning framework based on a parameter transfer strategy, and applies it to the width prediction of plates after multiple passes of hot rolling. The framework applies a swarm optimization algorithm to the original theoretical model and generates numerous highly-physical consistent samples. The established deep neural network (DNN) model is trained with simulated data, and the parameters are fine-tuned using a parameter transfer strategy combined with actual data to ensure excellent adaptation to the actual environment based on adequate learning of theoretical knowledge. In tests, the proposed model had the best overall prediction performance in this paper. Meanwhile, the developed model is consistent with the existing perception of rolling theory. This allows for the quick and reliable application of machine learning models in production. [Display omitted] • A novel theory-guided machine learning framework for width prediction of rolled plates. • Pre-training of the model using the data generated by the optimized theoretical model. • The proposed model has the best accuracy in both sufficient and less actual samples. • The variable influence of the proposed model is consistent with theoretical perceptions. [ABSTRACT FROM AUTHOR]

This paper precisely estimates the looming demand for hydrogen within the transportation sector, spanning 14 pivotal G20 nations from 2022 to 2050, anchored in the global vehicle hydrogen market penetration rate. Employing a robust techno-economic modeling methodology and leveraging Geographic Information System (GIS) for enriched visualization, it unveils distinct environmental dividends across three thoughtfully devised scenarios: Business- as -Usual, Moderate, and Aggressive. Findings herald India and China as pinnacles of hydrogen demand, with South Korea and Japan tailing closely. The study forecasts the energy consumption for solar hydrogen production in G20 countries to oscillate between 3.02 and 2.89 million GWh in 2022, while production costs are anticipated to range from $8.47/kg to $10.01/kg. In an assertive scenario, the paper illuminates a pathway towards achieving a significant CO 2 reduction, unmasking environmental savings of up to a staggering 48%, equivalent to 1.5 million mtCO 2 , by 2050. [Display omitted] • Hydrogen demand of G20 countries is analyzed using GIS and techno-economic models. • The projected adoption rates of hydrogen vary among the G20 countries. • Solar energy is crucial in determining each region's annual energy demand. • China, US, and India have the highest solar potential throughout the year. • The environmental analysis shows 850–1.5 million mtCO2 total emissions by 2050. [ABSTRACT FROM AUTHOR]

Renewable energy sources, driven by the global concern for climate change and the urgent need to reduce greenhouse gas emissions, are gaining increasing importance. Among them, solar energy stands out as a crucial renewable source and accurate solar irradiance forecasting is vital for its effective utilisation. Deep learning (DL) techniques have emerged as state-of-the-art solar irradiance forecasting methods in past years. However, traditional models that are widely used have inferior accuracy in forecasting. To bridge this gap, this paper presents an improved state-of-the-art model that leverages the convolutional property to forecast the time series data. We got a reduction of 8.32% in MAE compared to the base model. The paper's findings highlight the potential of DL-based approaches to significantly improve the accuracy of solar irradiance forecasts. [ABSTRACT FROM AUTHOR]

• Treat a problem of predicting the demand for staple food and feed grain. • A novel fractional discrete dynamic multivariate grey model is proposed. • The model considers the effects of linear correction term & random disturbance term. • Sailfish optimizer is used to optimize the fractional order accumulation parameter. • Empirical analysis shows the proposed method performs better than similar methods. Increasing geopolitical conflicts, extreme weather and other emergencies are continuously affecting global agriculture and food systems, threatening national and regional food security. Therefore, it is particularly important to accurately predict national or regional grain demand. In this paper, a novel fractional discrete dynamic multivariate grey model is proposed to predict grain demand. Based on the discrete modeling idea, the model considers the impacts of the change trend of the related factor sequences, adds the linear correction term and the random disturbance term, and applies the fractional order accumulation strategy, which improves the accuracy and robustness. An algorithm based on sailfish optimizer is introduced to optimize the fractional order accumulation parameter of the model. Taking the Yangtze River Economic Belt as an example, the fitting and prediction results of the novel and the existing three models are compared. The novel model can better fit and predict the demand for staple and feed grain, which is superior compared to other models. The predictions show that demand for staple food and urban feed grain in the Yangtze River Economic Belt will increase to varying degrees, while rural feed grain demand will remain stable overall. This paper will help the government to better grasp the changes in the structure and quantity of staple food and feed grain demand in the Yangtze River Economic Belt and formulate efforts to ensure food security. [ABSTRACT FROM AUTHOR]

• Development of a numerical tool for the static and fatigue delamination simulation able to adapt the load step size to properly estimate the delaminated area. • Numerical application according to literature Double Cantilever Beam benchmark. • Assessment of the effectiveness of the developed tool through comparison against experimental results. • Comparison of the SMXB tool against standard VCCT and Cohesive Surface. This paper deals with the Smart Time XB (SMXB) procedure, a parametric 3D methodology to simulate the delamination evolution in composite structures. The SMXB method has been developed almost ten years ago and, over time, has been improved with new features to become a forefront method compared with the state of the art. The SMXB is based on the Virtual Crack Closure Technique, which is nowadays implemented in the principal and most used Finite Elements (FE) commercial software. The main peculiarity of the SMXB is its capability to overcome the limits in terms of load step size and elements size dependences typical of the VCCT approach in mimic the delamination propagation. The SMXB procedure has been validated over years considering different structures and load conditions. The last development, which make the SMXB even more complete and attractive, is the ability to simulate the fatigue-driven delamination. Indeed, the Paris Law has been introduced in the core structure of the SMXB to make it able to mimic the delamination evolution under cyclic loading conditions. This manuscript has the intent to describe the theory behind this peculiar procedure and its evolution toward fatigue simulation. An experimental literature Double Cantilever Beam benchmark has been used to prove the robustness and effectiveness of this computational methodology, highlighting its main peculiarities. Single leg bending example, representative of a mixed mode interaction, has also been presented. [ABSTRACT FROM AUTHOR]

• A novel combined interval prediction algorithm is proposed in this paper. • Hybrid feature selection strategy is designed in the proposed system. • Multi-objective optimization mechanism reflects strong search capabilities. • Four interval prediction models cover the inherent modes of sequence. Traditional point prediction approaches can not reflect the uncertainty, which brings greater risks to decision-makers. To fill this gap, this paper extends a feature selection strategy that relies solely on correlation and redundant feature judgment, proposes a novel combined interval prediction algorithm, 3-Mcip (Combined Interval Prediction Based on Maximize Relevancy, Minimize Redundancy and Maximize Synergy) system, and solves the tradeoff between prediction accuracy and interval width. This system first designs a hybrid feature selection strategy to optimally select candidate variables and reduce model input redundancy. Secondly, the structure of the four ANN models is improved to accommodate the results of feature selection, and an optimization mechanism is introduced to search for the Pareto optimal solution set. In order to measure the comprehensive performance of the 3-Mcip system, hourly power load data and related candidate variables from Pittsburgh and Washington, D.C are considered. The numerical results show that the 3-Mcip system has coverage rates of 53.3333, 90.1667, and 99.4479 for Site 1 at different levels of interval width coefficients, which not only achieves perfect prediction of power load but also analyzes uncertainty. It is also helpful for power system managers to better capture the fluctuation range of future load and improve the flexibility of smart grid dispatching. [ABSTRACT FROM AUTHOR]

• A DC-Gaussian process regression model is used to the prediction of Dst index. • DC-GPR model has the best stability among the three models among DC-GPR, DC-SVM, and DC-NN models • DC-GPR model exhibits the better forecasting performance than DC-SVM and DC-NN models in Dst forecasting. In this paper we apply a class of Bayesian machine learning model, Gaussian Process Regression, to the prediction of Dst index by using 80 intense geomagnetic storms data (Dst ⩽ - 100 nT) from 1995 to 2014. The purpose of this paper is to compare the performance of Gaussian process regression model with Support Vector Machine model combined together with Distance Correlation (DC-SVM) and Neural Network model combined together with Distance Correlation (DC-NN) (Lu et al., 2016). For comparison, we estimate the correlation coefficients (CC), the RMS errors, the absolute value of difference in minimum Dst (Δ Dst min ) and the absolute value of difference in minimum time (Δ t Dst ) between observed Dst and predicted one.In order to compare the prediction effects and the generalizability of the three models to magnetic storm events, we combined 70 intense magnetic storm events and 10 super large magnetic storm events into one group. It is shown that DC-GPR model exhibits the better forecasting performance than by DC-SVM model and DC-NN model in magnetic storm. The CC, the RMS errors, the Δ Dst min , and the Δ t Dst of GPR are 0.65, 35.45 nT , 16.12 nT and 1.16 h , respectively. [ABSTRACT FROM AUTHOR]

This paper proposes an industrial-friendly approach based on physics-informed Bayesian machine learning for probabilistic inference and refinement of stability predictions in milling. The knowledge from physics principles and theoretical models is gathered into a substructure-based framework and enables dedicated updating and exchange of models on a substructure level. The Bayesian learning algorithm utilizes this framework as a basis to simultaneously interact with multiple machine and process configurations that share substructures, leading to federated and transfer learning over a network of machine tools on a production site. Moreover, the paper further elaborates on the Bayesian perspective to anticipate the potential usefulness of given experimental data in improving the model accuracy by quantifying information gain for two primary purposes: Firstly, recursive online learning can manage computational resources by processing only the most informative data points that may be collected during arbitrary cuts. Secondly, active learning leverages the information gain to dynamically adapt a sequential data collection, leading to accelerated learning with minimal labeled data. The experimental validations confirm that the proposed approaches lead to effective learning and reliable predictions of milling stability, outperforming traditional deterministic methods. [ABSTRACT FROM AUTHOR]

In this paper, I introduce two novel treatments into learning-to-forecast experiments: an initial anchor and a limited information treatment. The initial anchor pins down the subjects' initial expectations and makes cross-session results more comparable. In the limited information treatment, subjects can only observe the most recent market outcome. Surprisingly, price dynamics do not differ between full and limited information sessions. This suggests that when making decisions, subjects disregard most information and rely primarily on a few recent observations. Furthermore, regardless of the market's feedback system, positive or negative, subjects predominantly use a single heuristic to form expectations. • Two new treatments are introduced into learning-to-forecast experiments. • An initial anchor can successfully pin down subjects' initial expectations. • In the limited-info treatment, subjects can only observe the most recent outcome. • Surprisingly, having more information does not improve subjects' predictions. • Subjects use one heuristic to make predictions, regardless of the feedback system. [ABSTRACT FROM AUTHOR]

This paper examines the effects of macroeconomic and financial uncertainty on professional economic forecasts. Using the predictions of four variables sourced from the Bloomberg survey, we find that macroeconomic uncertainty, a proxy for the complexity of the forecasting task, is associated with high disagreement and low accuracy. By contrast, the results show that financial uncertainty is associated with low disagreement and high accuracy, which suggests that financial uncertainty encourages forecasters to adhere to the consensus to avoid large forecast errors. Furthermore, we find that the forecaster rank moderates the effect of macroeconomic uncertainty, indicating that high-ability forecasters can better navigate periods of high uncertainty. This study advances the understanding of forecasting behavior under uncertainty. [ABSTRACT FROM AUTHOR]

• A new Weibull log-logistic mixture distribution accumulation generation operator is established. • A novel grey model with multiple interaction effects is established. • The robustness and stability of the proposed method is proved and validated. • A new parameter estimation method is designed. • The proposed method is superior to other seven benchmark models in four case studies. Forecasting the failure of hard disk drives is important in server operation and has attracted increasing attention. However, current disk drive warning systems suffer from high false positive rates and high resource consumption when dealing with hard disk drive overall failure. Therefore, to accurately and stably predict hard disk drive overall failure, this paper develops a new Weibull Log-Logistic grey forecasting model with multiple interaction effects. Firstly, to capture and fit the trend of hard disk failure data flexibly and reduce the volatility, a new accumulation generation operator is established by introducing the Weibull Log-Logistic mixture distribution. Secondly, the proposed model constructs a multiple interaction term to describe the nonlinear relationship of the dependent variables on the system behavior series and the multiple interaction effects between the independent variables. Then, the parameter estimation method is designed to improve the fitting accuracy of the new model, in which linear estimation method, nonlinear estimation method, and meta-heuristic optimization algorithm are used to calculate the parameter values according to their categories. Finally, four varieties of hard disk drive data sets from BackBlaze are selected as study cases to validate the effectiveness of the proposed model. The results show that the mean MAPE of the proposed method is 2.1181 %, 2.2306 %, 8.1712 %, and 4.3417 %, respectively, corresponding to (hard disk drives ST8000NM0055, ST12000NM0007, ST10000NM0086, and ST14000NM0138), and the average value of the evaluation metrics are optimal in all competing models. Furthermore, it is observed that the number of reallocated sectors has the greatest influence in causing the failure of hard disk drives. [ABSTRACT FROM AUTHOR]

High precision ionospheric Total Electron Content(TEC) prediction is of great significance for improving the accuracy of Global Navigation Satellite System(GNSS), preventing natural disasters, and ensuring wireless communication. Given the varying frequencies of TEC signals, a hybrid CEEMDAN-BiLSTM-PSO-LSSVM-FE model for predicting ionospheric TEC content is proposed in this paper. Firstly, complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is used to decompose the extracted hourly TEC sequence and calculate the fuzzy entropy (FE) of the subsequences. Then, the signal is divided into high-frequency and low-frequency parts based on the fuzzy entropy value, for the high-frequency component, Bidirectional Long Short-Term Memory network (BiLSTM) is used for prediction; for the low-frequency component, Particle Swarm Optimization-based Least Squares Support Vector Machine (PSO-LSSVM) is used for prediction. The hourly TEC values collected from six evenly distributed GPS stations in China are used as the main input variable for the proposed model, with solar and geomagnetic activity data used as auxiliary data, using the TEC data from the previous 48 h to forecast the TEC content for the next hour. The performance of the model is evaluated by comparing it with other models under different geographical locations, seasons, solar and geomagnetic activity conditions. Experimental results show that the proposed hybrid model outperforms other models, with a correlation coefficient R2 and root mean square error (RMSE) of 0.99 and 0.20 TECU, respectively. The model effectively overcomes the problem of low TEC prediction accuracy and can provide more precise ionospheric delay correction services. [ABSTRACT FROM AUTHOR]

Breast cancer is a multifactorial heterogeneous disease and the leading cause of cancer-related deaths in women; its diagnosis and treatment require clinical sensitivity and a comprehensive disciplinary research approach. The expression of different receptors on tumor cells not only provides the basis for molecular typing of breast cancer but also has a decisive role in the diagnosis, treatment, and prognosis of breast cancer. To date, immunohistochemistry (IHC), which uses invasive histological sampling, has been extensively used in clinical practice to analyze the status of receptors and to make an accurate diagnosis of breast cancer. As an invasive assay, IHC can provide important biological information on tumors at a single point in time, but cannot predict future changes (due to treatment or tumor mutations) without additional invasive procedures. These issues highlight the need to develop a non-invasive method for predicting receptor status. The emerging field of radiomics may offer a non-invasive approach to identification of receptor status without requiring biopsy. In this paper, we present a review of the latest research results in radiomics for predicting the status of breast cancer receptors, with potential important clinical applications. [ABSTRACT FROM AUTHOR]

This paper proposes a three-stage online deep learning model for time series based on the ensemble deep random vector functional link (edRVFL). The edRVFL stacks multiple randomized layers to enhance the single-layer RVFL's representation ability. Each hidden layer's representation is utilized for training an output layer, and the ensemble of all output layers forms the edRVFL's output. However, the original edRVFL is not designed for online learning, and the randomized nature of the features is harmful to extracting meaningful temporal features. In order to address the limitations and extend the edRVFL to an online learning mode, this paper proposes a dynamic edRVFL consisting of three online components, the online decomposition, the online training, and the online dynamic ensemble. First, an online decomposition is utilized as a feature engineering block for the edRVFL. Then, an online learning algorithm is designed to learn the edRVFL. Finally, an online dynamic ensemble method, which can measure the change in the distribution, is proposed for aggregating all layers' outputs. This paper evaluates and compares the proposed model with state-of-the-art methods on sixteen time series. [ABSTRACT FROM AUTHOR]

With increasing quality requirements, especially for safety-critical products such as in medical technology, manufacturing companies face the challenge, that the quality inspection of component characteristics may become a crucial factor for economic production. For increasing the ratio of correctly identified scrap without increasing the amount of physical inspection, prediction models pose a valuable tool. Prediction models in manufacturing build process models through process data collected during the manufacturing process. The prediction models can for example be used to predict the product quality from the process data, which enables the substitution of physical quality inspections through virtual inspections. Before selecting suitable prediction models, checkpoints in the manufacturing process and inspection sequence (MPIS) for integrating the prediction models profitably have to be identified. Therefore, the paper aims to provide a methodology to generate integrated quality strategies for MPIS that combine physical inspection with the prediction of quality characteristics. Furthermore, a methodology for the economic evaluation of the quality strategies as decision support is provided. Since the number of prediction models available for predicting quality based on process data is manifold, the explicit recommendation for the implementation of a certain prediction model is not the focus of this paper. The methodology presented rather focuses on the identification of suitable checkpoints in the MPIS for prediction model integration. For generating quality strategies suitable checkpoints in the manufacturing process sequence have to be identified. Therefore, three aspects are considered, the influence of the manufacturing process on the change in a characteristic and the overall importance of a characteristic for the component function as well as the ability of the characteristic and process for being predicted. A prediction recommendation is calculated that enables the derivation of quality strategies as a combination of physical and virtual inspection for different characteristics in different checkpoints. For the economic evaluation of the derived quality strategies two cases for supplementing physical inspection with prediction models to either compensate for the alpha- or beta-error rate are presented. The most reasonable quality strategy is chosen according to its economic potential. For validation, the applicability of the methodology for integrating prediction models in manufacturing process and inspection sequences to a use case of medical technology is demonstrated successfully. It is shown that the methodology supports users in identifying possible quality strategies and their economic evaluation for manufacturing process and inspection sequences as a basis for the decision regarding quality strategies to be implemented. [ABSTRACT FROM AUTHOR]

Accurate prediction of PV power is essential to ensuring the safe and economic operation of power systems with high PV penetration. The current PV power prediction scheme considering the spatio-temporal correlation characteristics is relatively simple in data processing, resulting in low prediction accuracy; at the same time, the missing data also poses a great problem to the prediction. Therefore, in order to improve the prediction accuracy and solve the problem of missing data, this paper proposes a PV power spatio-temporal prediction model considering time-shift correction and a multi-station information fusion strategy Firstly, relevant power station clusters are constructed using hierarchical clustering, and a similar daily data filtering model considering the variation characteristics of daily power characteristic curves is proposed to filter the data; Secondly, multiple BP neural network models are constructed and multiple reference power stations with high relevance are predicted using irradiance information; Thirdly, the prediction results of multiple reference power stations are input to the data processing module for time-shift analysis and spatial correlation information fusion correction, which solves the missing data problem of the target power station to be predicted. Finally, it is input to One-dimensional Convolutional Neural Network(1DCNN) to achieve the power prediction of the target power station with missing data. The simulation analysis shows that the root mean square error (RMSE) of a sunny day forecast is 3.31%; the RMSE of a non-sunny day forecast is 9.65%, which proves the accuracy of this two-layer neural network is higher compared to other model structures, so the proposed scheme has certain reliability and accuracy in the prediction of PV power with missing data. • The paper examines the current issues and challenges with PV power forecasting. • A similar daily data filter considering the various characteristics of the daily PV power output is proposed. • A BP-1DCNN network considering spatio-temporal sharing is designed to predict the PV power of the missing data plants. • A time-shift correction and spatially relevant information fusion strategy are used to connect the two networks. • The spatio-temporally corrected network prediction model has higher accuracy and solves the problem of missing data. [ABSTRACT FROM AUTHOR]

This paper presents a new methodology to obtain prediction regions of the output of a dynamical system. The proposed approach uses stored past outputs of the system and it is entirely data-based. Only two hyperparameters are necessary to apply the proposed methodology. These scalars are chosen so that the size of the obtained regions is minimized while fulfilling the desired empirical probability in a validation set. In this paper, methods to optimally estimate both hyperparameters are provided. The provided prediction regions are convex and checking if a given point belongs to a computed prediction region amounts to solving a convex optimization problem. Also, approximation methods to build ellipsoidal prediction regions are provided. These approximations are useful when explicit descriptions of the regions are necessary. Finally, some numerical examples and comparisons for the case of a non-linear uncertain kite system are provided to prove the effectiveness of the proposed methodology. • A method to compute prediction regions for multivariate systems is presented. • These regions contain future values of the outputs with a certain probability. • The proposed method does not need any knowledge of the system dynamics. • A procedure to obtain ellipsoidal approximations of the regions is also presented. • The results are illustrated by means of numerical examples. [ABSTRACT FROM AUTHOR]

With increasing multimedia content and social activities, moderation problems increase. There are different approaches to moderation and automation. However, they have limitations in terms of usage in real-time. The analysis of scientific papers revealed that most of the more common approaches solve the task of detection instead of prediction by considering the final utterance. For this reason, calls are unprotected in toxic languages, and online broadcasts can be unpredictable. In this work, a new way for automatic speech moderation in terms of dynamic word prediction was suggested. The considered task involves the analysis of the auditory and textual channels of speech. Words can have different meanings depending on the context, so in solving the problem it is planned to consider profanity, which is socially unacceptable regardless of the context. In this paper approaches for working with speech stream in the task of profanity prediction were proposed. It can be possible to have smaller latency with usage of audio features. We also suggest the pipeline for real-time (with the ability to predict the sequence with a higher duration than the latency of the processing) prediction for multimodal prediction, which compensates the latency of ASR systems. As a result, in this paper, we compared different solutions for the next color prediction task for English speech and reached the F1 score of 86.6 for 3 class prediction. [ABSTRACT FROM AUTHOR]

The accidental leakage of gaseous hydrogen from hydrogen refueling station facilities has the potential to result in a large fire and explosion catastrophe. The ability to forecast the spatiotemporal evolution of hydrogen plumes in real-time is crucial for monitoring the distribution of plume concentrations and mitigating the risk of fire and explosion. The utilization of deep learning has been extensively employed in a diverse range of spatiotemporal forecasting tasks. However, these approaches encounter limitations in accurately quantifying uncertainty when predicting spatiotemporal concentrations and plume boundaries. This research paper introduces a novel model called DPSTNN_H 2 Evolution, which is a deep probabilistic spatial-temporal neural network designed for forecasting the spatiotemporal evolution of hydrogen plumes. The benchmark dataset is constructed by the implementation of numerical simulations pertaining to the inadvertent leak of hydrogen within a hydrogen refueling station. The findings of the study indicate that incorporating quantified uncertainty information might enhance the precision of plume boundaries and the resilience of plume concentrations when predicting the spatiotemporal development of hydrogen. By employing Monte Carlo sampling with a sample size of m = 100 and a dropout rate of p = 0.1 , the model demonstrates the ability to provide real-time inference for 10 instances of plumes, while maintaining a high level of accuracy with R2 = 0.9829. In comparison to the current state-of-the-art model, the proposed model demonstrates superior accuracy and robustness in forecasting the spatiotemporal evolution of hydrogen plumes. In general, our proposed model has the potential to serve as a dependable option for the development of a digital twin for emergency management of hydrogen refueling stations. • Probabilistic spatial-temporal neural network model for hydrogen plume evolutions is proposed. • Normalized uncertainty contours and uncertainty intervals to improve accuracy and robustness. • Numerical experiment of hydrogen leakage and dispersion in refueling station are conducted. • Optimal hyper-parameters of the proposed model are determined • Comparison between the proposed model with point-estimation based model is conducted [ABSTRACT FROM AUTHOR]

This paper proposes a trend prediction analysis method based on the evolution of knowledge graph fusion for the analysis of price fluctuation trends with limited information and lacking historical data. Based on an actual business case, the article first establishes a knowledge graph and corresponding data model based on the price analysis and application scenarios in business decision-making and based on the collected data. Afterward, the evolution mechanism model and the corresponding theoretical basis are described. After that, the feasibility of the proposed method is proved theoretically and via simulation verification. This method does not require tags and historical data as necessary support, thereby reducing the data requirements of machine learning predictions, and can be used as a supplementary solution to complete trend analysis with less requirement on data curation. The method's stability and feasibility have been substantiated through experiments, showcasing distinct effects on the trend emanating from different interrelated knowledge graphs. Recognizing the potentially conflicting constraints between diverse correlations, we undertake a comprehensive evaluation of each map's influence. This evaluation is achieved through knowledge graph fusion processing, enabling a predictive assessment under the collective impact of various factors. [ABSTRACT FROM AUTHOR]

Current inflation rates have significantly been increasing construction project costs globally. Thus, it is crucial to consider potential inflationary trends when estimating the cost at completion of prolonged projects. However, the absence of literature providing guidance on analyzing inflation trends and adjusting the ongoing project cost performance factor is a gap that needs filling. This study proposes a framework for estimating ongoing project costs based on trend and seasonality analysis of the project cost performance through the Holt-Winters method. The proposed framework also incorporates the estimation of project completion time and is applied to a real case study to assess its efficacy. The results are compared with the ones obtained through pre-existing index-based methods. This paper seeks to improve the accuracy of project cost estimation and management in challenging and uncertain economic environments by providing an alternative to the traditional cost at completion forecasting approach. [ABSTRACT FROM AUTHOR]

This paper delves into the critical aspect of demand forecasting within the broader context of optimizing drop trailer management in volatile networks, with a specific focus on a large pallet manufacturer's supply chain operation. The study underscores the importance of accurate demand forecasting as a foundational element for informing subsequent optimization models. The main objective is to enhance our sponsor company's supply chain efficiency by accurately predicting future trailer requirements, which is crucial for the subsequent development of an effective inventory control and optimization model. This research utilizes forecasting methods like Gradient Boosted Trees and highlights the challenges of traditional forecasting methods in the context of our sponsor company's complex and dynamic supply chain network. The demand forecast is meant to inform the development of optimization models crucial for ensuring the effective allocation and management of trailer assets, ultimately leading to cost reductions and improved service levels within our sponsor company's network. The study contributes significantly to supply chain management literature by showcasing the application of sophisticated forecasting techniques in a real-world context and setting the stage for the development of robust optimization models in the domain of drop trailer management. [ABSTRACT FROM AUTHOR]

Human age and gender prediction from facial images has garnered significant attention due to its importance in various applications. Traditional models struggle with large-scale variations in unfiltered images. Convolutional Neural Networks (CNNs) have emerged as effective tools for facial analysis due to their robust performance. This paper presents a novel CNN approach for robust age and gender classification using unconstrained real-world images. The CNN architecture includes convolution, pooling, and fully connected layers for feature extraction, dimension reduction, and mapping to output classes. Adience and UTKFace datasets were utilized, with the best training and testing accuracies achieved using an 80% training and 20% testing data split. Robust image pre-processing and data augmentation techniques were applied to handle dataset variations. The proposed approach outperformed existing methods, achieving age prediction accuracies of 86.42% and 81.96%, and gender prediction accuracies of 97.65% and 96.32% on the Adience and UTKFace datasets, respectively. [ABSTRACT FROM AUTHOR]

This research paper focuses on comparing two Machine Learning (ML) approaches—Time Series Analysis and a Feature-Based Prediction—for predicting traffic accident counts. Spanning the years 2005 to 2016 in Paris, the study leverages a comprehensive dataset that includes attributes such as weather conditions. Following dataset preprocessing and partitioning into training and test sets, the research systematically develops and evaluates these two contrasting ML modeling approaches. Furthermore, the study extends its comparison by evaluating the performance of various ML models, emphasizing the power of the proposed Neural Network (NN) model. Metrics such as Mean Absolute Percentage Error (MAPE) are employed to assess and contrast the predictive capabilities of each model. The results not only showcase the effectiveness of the neural network model but also elucidate its superiority over alternative ML approaches, particularly in the context of traffic accident prediction. [ABSTRACT FROM AUTHOR]

The mixed data sampling (MIDAS) model has attracted increasing attention due to its outstanding performance in dealing with mixed frequency data. However, most MIDAS model extension studies are based on statistical methods or machine learning models, which suffer from insufficient prediction performance and stability in small sample environments. To solve this problem, this paper proposes a novel mixed frequency sampling discrete grey model (MDGM(1, N)), which is a coupled form of the MIDAS model and discrete grey multivariate model. By adjusting the structure parameters, the model can be adapted to different sampling frequencies data, and degenerate into several types of grey models. Then, the unbiasedness and stability of the model are proved using the mathematical analysis method and numerical random experiment. The meta-heuristic algorithm is introduced to obtain the optimal weight parameters and the maximum lag order, improving the model's fitting ability to mixed frequency data. To demonstrate the effectiveness of the new model, a model evaluation system consisting of traditional evaluation metrics and a monotonicity test is established. Taking four hard disk drive failure datasets as research cases, the performance of the proposed model is compared with seven mainstream benchmark models. The results show that the proposed model has excellent applicability and outperforms other competition models in terms of validity, stability, and robustness. Furthermore, it is observed that the reported uncorrectable errors and the command timeout have a greater impact on hard disk drive failure. Finally, the new model is employed to forecast the failure of four hard disk drives. The forecasting results indicate that in the next four time points with a cycle of 21 days beginning in April 2023, the failure of the smaller capacity hard disk drives (0055 and 0086, corresponding to 8TB and 10TB) show a decreasing trend, reaching 67.442% and 89.7683%, respectively. The failure of the other larger capacity hard disk drives (0007 and 0138, corresponding to 12TB and 14TB) has increased, with a growth rate of 17.1016% and 123.7899%. [Display omitted] • A novel mixed frequency sampling discrete grey model is proposed. • The proposed model is proved to be unbiased and stable theoretically. • A Chimpanzee Optimization algorithm is introduced to globally search for weight parameters and maximum lag order. • A new model evaluation system for evaluating the effectiveness of prediction models is constructed. • The proposed model outperforms the other seven benchmark models in four case studies. [ABSTRACT FROM AUTHOR]

The short-term forecast of photovoltaic (PV) power is crucial for the security and economics of power system operations. However, the fluctuation characteristics of the PV power, which are closely related to the meteorological factors, introduce inaccuracies in its forecast. Towards this end, the paper studies the effects of clustering analysis at long time scale and data reconstruction technique at short time scale on capturing PV power fluctuation characteristics. A short-term PV power forecasts method based on multi-scale fluctuation characteristics extraction (MFCE), which employs a path analysis to identify the relevance of meteorological factors with PV power at long time scale and a phase space reconstruction to analyze PV power fluctuation characteristics at short time scale, is proposed in this paper. The proposed MFCE methodology deploys a widely-used extreme gradient boosting (XGBoost) model to output the forecasting results. Both the effectiveness and accuracy of the proposed methodology are verified by using the real data under the conditions of sunny and cloudy days of four seasons compared to traditional methodologies. • The multi-scale fluctuation characteristics are analyzed for PV power forecast. • Identify dominant meteorological factors affecting PV power by path analysis. • The hour-level fluctuation is reduced by using path analysis-based clustering. • The phase space reconstruction is used to reduce the minute-level fluctuation. [ABSTRACT FROM AUTHOR]

A time-evolving (or dynamic) social network is a network whose structure may change over time. The observed changes may refer to entities (for example, new members may join or existing members may leave the network) or connections between them (new relationships may be established or existing relationships may be broken). A number of studies have focused on the empirical analysis of time evolving networks to examine their properties. In this paper, the problem of link prediction is considered. It aims at predicting future connections between pairs of unconnected links based on existing current connections. A social network considered in the paper is an organizational social network, meant as a network where nodes represent employees or units of the organization, and edges between nodes represent formal or informal relations or information flow between them. The structure of our network is formed by digital communication between employees. The aim of the paper is to experimentally verify to what extent, if any, the similarity-based methods based on local or global information may support the process of predicting potential new (non-existent yet) connections between employees in a time-evolving organizational social network. Ten similarity methods based on local and global information have been considered and tested. As a result, the best similarity methods (candidates to be considered as good predictors of these future connections), have been suggested. [ABSTRACT FROM AUTHOR]

To reflect the future electricity demand variations in the industrial sector of western China under the "carbon peaking" and "carbon neutral" strategies, that the traditional methods of electricity consumption forecasting are no longer effective, this paper proposes a new hierarchy of electricity demand influencing factors based on the transition path of low-carbon and CO 2 emission intensity constraints of the industrial sector. Then, the feedback equations of the influencing factors based on CO 2 emission intensity is constructed, and a long-term electricity forecast model based on system dynamics is established. Finally, taking Ningxia Province as an example, the electricity demand and carbon emissions of the industrial sector are predicted under different policy constraint scenarios. The results show: (1) In the baseline scenario, electricity demand will tend to saturate with about 140.1TWh in 2030, and the CO 2 emission level of electricity demand will peak at 72.52 million tons in 2029. (2) Compared with the system dynamics model without considering CO 2 emission intensity, the mean absolute percentage error is reduced about 3.16 % by inputting the influence factors and regression coefficients designed in this paper. (3) Compared with other forecasting methods, the model proposed in this paper has the lowest mean absolute percentage error of 2.75 %. The model improves the accuracy and provides a reference for planning departments to develop energy transition plans, where the prediction results reflect the reality of the industrial sector in western China. Correspondingly, this model has reference value for relevant institutions in forecasting the electricity demand under the low-carbon development path in industrial sectors of different regions. • Influencing factors are screened according to the low-carbon transition paths. • The distributed power installed capacity and load interaction coefficient are considered. • Multiple regression equations are developed between CO 2 emission targets and influencing factors. • A long-term electricity demand forecast model driven by CO 2 constraints is constructed. • The multi-scenario electricity demand in western China is forecasted in response to carbon peaking and carbon neutral targets. [ABSTRACT FROM AUTHOR]

Container throughput is an essential indicator for measuring the container terminal's efficiency. Gate in and gate out containers are the containers that are transported to and transported out of the terminal respectively. Containers are stacked in the container yard before they leave the terminal. Handling of these containers accounts for a major workload at the terminal. Therefore, accurate short-term forecasting of the daily gate in and gate out containers at a container terminal is critical for operational planning. While most forecasts are made for the strategical level of the overall container throughput, this paper focuses on the daily gate in and gate out container quantities with a case study of Ningbo Zhoushan Port Beilun Second Container Terminal. Traditional approaches are mainly time series prediction algorithms that purely rely on the previously observed values. This paper proposes a novel decomposition-ensemble methodology framework which first divides the daily forecasting into a group of forecasting on a set of vessels which arrive in a certain time range around the predicted day. A novel machine learning method, "Extreme Gradient Boost (XGBoost)", is used for the vessel level of forecasting, where some temporal-related features of vessels are considered in the model. We then ensemble the predicated value of all vessels in the set as the total gate in/out amount for a day. Experimental results demonstrate that our proposed method achieves notable performance gains compared to the time series-based method ARIMA. In addition, the results of this new technology can be fed into the terminal operating system (ToS) in the Ningbo Beilun container terminal for better management. Specifically, the prediction results facilitate effective real-time decision-making for operational management and policy drafting, such as workload planning, equipment scheduling, yard planning, etc. [ABSTRACT FROM AUTHOR]

The stock market is highly volatile as it depends on political, financial, environmental, and various internal and external factors along with historical stock data. Such information is available to people through microblogs and news and predicting stock price merely on historical data is hard. The high volatility emphasizes the importance to check the effect of external factors on the stock market. In this paper, a machine learning model is proposed where the financial news is used along with historical stock price data to predict upcoming stock prices. The paper has used three algorithms to calculate various sentiment scores and used them in different combinations to understand the impact of financial news on stock price as well the impact of each sentiment scoring algorithm. Experiments have been conducted on ten-year historical stock price data as well as financial news of four different companies from different sectors to predict the next day and next week's stock trends and accuracy metrics were checked for a period of 10, 30, and 100 days. The proposed model can achieve the highest accuracy of 0.90 for both trend and future trends for a period of 10 days. Experiments have also been performed to check the difficulty in predicting some stocks. It was found that Tata Motors an automobile company stock prediction has maximum MAPE and hence deviates more from actual prediction as compared to others. [ABSTRACT FROM AUTHOR]

COVID-19 is a pandemic that has resulted in numerous fatalities and infections in recent years, with a rising tendency in both the number of infections and deaths and the pace of recovery. Accurate forecasting models are important for making accurate forecasts and taking relevant actions. As a result, accurate short-term forecasting of the number of new cases that are contaminated and recovered is essential for making the best use of the resources at hand and stopping or delaying the spread of such illnesses. This paper shows the various techniques for forecasting the covid-19 cases. This paper classifies the various models according to their category and shows the merits and demerits of various fore-casting techniques. The research provides insight into potential issues that may arise during the forecasting of covid-19 instances for predicting the positive, negative, and death cases in this pandemic. In this paper, numerous forecasting techniques and their categories have been studied. The goal of this work is to aggregate the findings of several forecasting techniques to aid in the fight against the pandemic. [ABSTRACT FROM AUTHOR]

This paper provides a chronological review of publications that implement and advance the receptance coupling substructure analysis (RCSA) approach first applied to tool tip receptance (or frequency response function) prediction for milling applications in 2000. The review topics mimic the RCSA approach, where the tool, holder, and spindle-machine receptances are coupled analytically, and include: tool-holder receptance modeling; connection modeling; spindle-machine receptances; and applications. The review paper summarizes contributions from multiple, international authors (198 papers) to these topics. It also provides a comprehensive resource for those beginning an investigation into RCSA. • This paper provides a review of publications that implement and advance the receptance coupling substructure analysis (RCSA) approach. • The review topics include: tool-holder receptance modeling; connection modeling; spindle-machine receptances; and applications. • The review paper summarizes contributions from multiple, international authors to these topics. • The paper provides a comprehensive resource for those beginning an investigation into RCSA. [ABSTRACT FROM AUTHOR]

Heterogeneity in agents' preferences is generally analysed through mixed logit models, which assume taste parameters are distributed in the population according to a certain mixing distribution. As a result, if the utility function is linear in attributes, the willingness to pay is the ratio of two random parameters and is itself random. This paper proposes a technique built on the Delta method, partly analytical and partly based on simulations, to obtain the sampling distribution of the willingness to pay, accounting for both heterogeneity and sampling error. The paper contributes to the literature by: (i) redressing some imprecisions in Bliemer and Rose (2013) that produce biased results; (ii) proposing a faster estimation process, compared to the Krinsky and Robb (1986, 1990) method that, relying on simulation only, proves computationally more demanding; (iii) comparing the performance of different methods using both synthetic and real data sets. The paper shows, via a Monte Carlo study, that the method we develop and the Krinsky and Robb one produce similar results, while outperforming that proposed by Bliemer and Rose. • Uncertainty in estimating Willingness To Pay under Mixed Logit Model arises from both heterogeneity and sampling error. • Delta method can be used, in combination with normal mixtures, to estimate both sources of uncertainty. • Both confidence and prediction intervals can be computed for Willingness To Pay by means of Delta method. • Delta method, being partly analytical, proves time-saving when compared to Krinsky and Robb method. [ABSTRACT FROM AUTHOR]

In this research paper, we propose a novel hybrid deep learning approach, SSA-CNN-LSTM, for forecasting solar power generation. The approach combines Singular Spectrum Analysis (SSA), Convolutional Neural Networks (CNNs), and Long Short-Term Memory (LSTM) networks to leverage temporal and spatial dependencies in real-time greenhouse solar power generation data. Through a comprehensive comparative analysis, SSA-CNN-LSTM is compared against three established models, CNN-LSTM, SSA-CNN, and SSA-LSTM, employing real solar power generation data over a two-year period. The findings prominently demonstrate SSA-CNN-LSTM's exceptional performance, particularly in the 1-hour ahead prediction horizon. With an hour-ahead Mean Absolute Error (MAE) of 0.1202, SSA-CNN-LSTM surpasses the forecast precision of CNN-LSTM (0.6269), SSA-CNN (0.2354), and SSA-LSTM (0.2049). This excellence extends to the 2-hour-ahead forecast, where SSA-CNN-LSTM maintains its superiority with an MAE of 0.1400. In the day-ahead forecast, SSA-CNN-LSTM upholds its competitiveness, demonstrating an MAE of 0.1774. These outcomes underscore the immense potential of SSA-CNN-LSTM as a formidable tool for precise solar power forecasting. The model's effectiveness empowers greenhouse operators and energy management systems to optimize resource allocation, ultimately fostering elevated energy efficiency and overall greenhouse productivity. [ABSTRACT FROM AUTHOR]

This paper presents a comprehensive study on the application of machine learning techniques in the prediction of respiratory rate via time-series-based statistical and machine learning methods using several physiological signals. Two different models, ARIMA and LSTM, were developed. The LSTM model showed a stronger capacity for learning and capturing complicated patterns in the data compared to the ARIMA model. The findings imply that LSTM models, by incorporating many variables, have the ability to provide predictions that are more accurate, particularly in situations where respiratory rate values vary significantly. [ABSTRACT FROM AUTHOR]

Predicting stock prices is a well-known and significant problem. We can learn about market behaviour over time and identify trends that might not have been seen without an effective stock prediction model. Machine learning will be a useful approach to solving this issue with the increased processing capacity of computers. Behavioural economics also asserts that the investments made by investors depend on their emotions so psychological theories can also be applied to explain their behaviour and its impact on the market. Combining the analysis of these behavioural patterns with the use of historical financial data sets can result in an approach for accurate stock price predictions. The primary focus of the proposed model is on comparing different models to provide a comparative analysis of the results provided by models used in the literature. The paper provides insight into the improvisation of the current techniques and how different parameters and different error analysis techniques can be implemented. [ABSTRACT FROM AUTHOR]

The concept of predictions has gained much attention over the last few years. Research on prediction based on experience is error-prone. Usually, a lot of data has been available with multiple variables, but not all features are relevant. This work uses different feature selection techniques and their combinations to select the best possible feature subset. This paper uses feature selection techniques for time series data prediction with deep learning models. In this work, we have used filter, wrapper, and embedded methods for feature selection. We combine these techniques and compare their performance with single feature selection techniques. We found that the embedded method alone performs well with fewer features, but its performance deteriorates when the number of features increases. We input these reduced feature sets into deep learning models such as recurrent neural networks (RNN), long short-term memory (LSTM), and 1d Convolutional with LSTM (1d CNN-LSTM). We assess the performance of different deep-learning models with reduced features. Our results analysis found that the right feature subset may help increase the performance of deep learning models for time series prediction. Performance improvement is specific to the dataset, feature selection techniques, and the deployed DL models; performance improvement varies on average, from a few percent to fifty percent. [ABSTRACT FROM AUTHOR]

Link prediction is significant in the field of network analysis which helps to determine the future or hidden connections between the entities in a network. Traditional link prediction methods typically utilize single algorithms to make predictions, which might not be optimal for diverse and complex networks. In this research paper, a new framework is proposed which utilizes the existing link prediction algorithms with ensemble techniques. We have explored three ensemble strategies: bagging, boosting, and voting classifier and four individual link prediction algorithms: Adamic Adar, Katz, CCPA, Accurate link prediction. Methods considering ensemble techniques combine multiple individual models to improve prediction accuracy and robustness. We have experimentally compared the performance of proposed framework using six real world datasets in terms of AUC and Accuracy. The results show the superiority of the proposed framework over individual link prediction algorithms. [ABSTRACT FROM AUTHOR]

Weather forecasting is a critical tool for many different applications, from agriculture and transportation to disaster preparedness and response. While weather forecasts are not always perfect, they provide valuable information that can help people make informed decisions and take appropriate actions to protect themselves and their property from the impacts of extreme weather events. In this paper, to forecast the weather we are using the SARIMA model as RMSE is 1.24 and analyse how it forecast the weather with high accuracy. Note that the SARIMA model is a machine learning (ML) technique used to forecast time series data that has both trend and seasonal components. It is an extension of the ARIMA model, which stands for Autoregressive Integrated Moving Average. To account for periodic changes in the data, the SARIMA model augments the ARIMA model with a seasonal component. The SARIMA model is trained on historical weather data and can be used to predict future weather patterns. It is a powerful tool for weather forecasting as it can accurately predict both short-term and long-term weather trends. The quality of the input data determines how accurate the SARIMA model will be, and the model can be adjusted to perform better. In conclusion, the model known as SARIMA is a ML based approach that has been successfully used to predict weather. It is a powerful tool that can handle non-stationary data and seasonal components and can accurately predict short-term and long-term weather trends. [ABSTRACT FROM AUTHOR]

In alumina production, the evaporation as the key process uses recyclable resources and reduces environmental pollution. In fact, the quality of export product with offline and delayed, results in low precision of process control and high energy consumption. To ensure green and efficient production, in this paper, a new prediction method integrating process knowledge and data-driven spatial-temporal adaptive model is put forward. First, to preprocessed production data for ensuring modeling accuracy, data reconciliation technology is adopted. Then, based on material and heat transfer mechanism, for equipment and industrial process, the mechanism models are established. Furthermore, with time difference and moving window model, an error compensation method is utilized in terms of double locally weighted kernel PLS for estimation error in hypothesis-based mechanism modeling. Finally, the data-driven spatial-temporal adaptive model and the process knowledge-based mechanism model are integrated. To illustrate the model feasibility, an industrial sodium aluminate solution evaporation is used. It demonstrates that, for the developed model, the prediction accuracy can reach more than 90% within the ± 2% error range, and effectively estimate the actual product quality and ensure the prediction effect. • A new concentration prediction method integrating knowledge and data-driven spatial-temporal adaptive model is proposed. • Mechanism modelling based on data reconciliation and process analysis is achieved to estimate concentration. • A TD-MW-DLWKPLS model considering both sample similarity and variable correlation is adopted for error compensation. • The performances of the developed method are validated on evaporation industrial application. [ABSTRACT FROM AUTHOR]

The influence of operating parameters on the hydrogen production performance of the proton exchange membrane electrolysis cell (PEMEC) is crucial. In this paper, we first investigate the effect of a single operating parameter on the performance of PEMEC, and find the law of single operation effect. After that, the orthogonal experimental methods are used to investigate the effect of cross-coupling of combined operating parameters on the performance of PEMEC and the optimal combination of levels of operating parameters is obtained. Finally, the LSTM neural network is used to predict the PEMEC performance under the multiple operating parameters. The results showed that the weights of voltage, temperature, water flow rate, and bolt torque on PEMEC performance were 68.72 %, 20.63 %, 6.17 %, and 4.48 %, respectively. The optimal combination of operating parameters for the experiment is a torque of 3.5 N m, temperature of 70 °C, voltage of 2.0 V, and the water flow rate of 40 mL/min. It is also found that the long short term mermory (LSTM) neural network is able to predict the performance of PEMEC well, and that the experimental data-driven approach is an effective method for predicting the experimental performance. • The single operating parameter influence law is discovered. • The optimal combination of operating parameters is derived. • The order and weights of operating parameters affecting PEMEC performance are found. • An effective experimental performance prediction method is proposed. [ABSTRACT FROM AUTHOR]

Intermittent data, characterized by sporadic and irregular occurrences, have successive zero values in time series, had present unique challenges for modeling and analysis. Forecasting using intermittent data is not easy to do. This paper presents an application of the Group Method of Data Handling (GMDH) in modeling and forecasting intermittent data. GMDH models excel in capturing non-linear patterns, handling missing or sparse data, and adapting to changing dynamics. By iteratively selecting the most informative variables and estimating their coefficients, GMDH constructs a hierarchical network of interconnected models that can effectively handle intermittent data. The forecasting results show that GMDH is able to follow the pattern of actual data. The mean deviation accuracy analysis shows a value of 72.22%. The findings showcase the potential of GMDH as a valuable tool in addressing the challenges associated with intermittent data and offer insights into its application across various domains. [ABSTRACT FROM AUTHOR]