Search

Sustainable concepts for on-demand transportation, such as ridesharing or ridehailing, require advanced technologies and novel dynamic planning and prediction methods. In this paper, we consider the prediction of taxi trip durations, focusing on the problem of the estimated time of arrival (ETA). ETA can be used to compute and compare alternative taxi schedules and to provide information to drivers and passengers. To solve the underlying hard computational problem with high precision, machine learning (ML) models for ETA are the state of the art. However, these models are mostly black box neural networks. Hence, the resulting predictions are difficult to explain to users. To address this problem, the contributions of this paper are threefold. First, we propose a novel stacked two-level ensemble model combining multiple ETA models; we show that the stacked model outperforms state-of-the-art ML models. However, the complex ensemble architecture makes the resulting predictions less transparent. To alleviate this, we investigate explainable artificial intelligence (XAI) methods for explaining the first- and second-level models of the ensemble. Third, we consider and compare different ways of combining first-level and second-level explanations. This novel concept enables us to explain stacked ensembles for regression tasks. The experimental evaluation indicates that the considered ETA models correctly learn the importance of those input features driving the prediction. [ABSTRACT FROM AUTHOR]

In the realm of structural health monitoring (SHM), understanding the expected behavior of a structure is vital for the timely identification of anomalous activities. Existing methods often model only the physical quantities of monitoring data, neglecting the corresponding temporal information. To address this, this paper presents an innovative deep learning framework that synergistically combines a BiLSTM model, fortified by a temporal pattern attention (TPA) mechanism, with time-encoded temperature and traffic-induced deflection-temporal patterns. The arithmetic optimization algorithm (AOA) is employed for optimal hyperparameter tuning, and incremental learning was implemented to enable real-time updates of the model. Based on the proposed framework, an anomaly detection method was subsequently developed. This method is bidirectional: it uses quantile loss to provide expected ranges for structural behavior, identifying isolated anomalies, while the windowed normalized mutual information (WNMI) based on multivariate kernel density estimation (MKDE) helps detect trend variability caused by decreases in structural stiffness. This framework and the anomaly detection method were validated using data from an operational cable-stayed bridge. The results demonstrate that the method effectively predicts structural behavior and detects anomalies, highlighting the critical role of temporal information in SHM. [ABSTRACT FROM AUTHOR]

The security of communication networks can be compromised through both known and novel attack methods. Protection against such attacks may be achieved through the use of an intrusion detection system (IDS), which can be designed by training machine learning models to detect cyberattacks. In this paper, the KOMIG (knapsack optimization and mutual information gain) IDS was developed to detect network intrusions. The KOMIG IDS combined the strengths of optimization and machine learning together to achieve a high intrusion detection performance. Specifically, KOMIG IDS comprises a 2-stage feature selection procedure; the first was accomplished with a knapsack optimization algorithm and the second with a mutual information gain filter. In particular, we developed an optimization model for the selection of the most important features from a network intrusion dataset. Then, a new set of features was synthesized from the selected features and combined with the selected features to form a candidate features set. Next, we applied an information gain filter to the candidate features set to prune out redundant features, leaving only the features that possess the maximum information gain, which were used to train machine learning models. The proposed KOMIG IDS was applied to the UNSW-NB15 dataset, which is a well-known network intrusion evaluation dataset, and the resulting data, after optimization operation, were used to train four machine learning models, namely, logistic regression (LR), random forest (RF), decision tree (DT), and K-nearest neighbors (KNN). Simulation experiments were conducted, and the results revealed that our proposed KNN-based KOMIG IDS outperformed comparative schemes by achieving an accuracy score of 97.14%, a recall score of 99.46%, a precision score of 95.53%, and an F1 score of 97.46%. [ABSTRACT FROM AUTHOR]

This paper introduces a novel and comprehensive model for the analysis of dam deformation trends, integrating the variational mode decomposition (VMD) method, fractal theory, and the whale optimization algorithm (WOA) to re6ne the deep extreme learning machine (DELM) model. This integration allows for a meticulous denoising process through VMD, e7ectively isolating pertinent signal characteristics from noise and measurement interference. Following this, fractal theory is utilized to conduct an in-depth qualitative analysis of the denoised data, capturing intricate patterns within the deformation trends. The model further evolves with the application of WOA to optimize the DELM model, thereby facilitating an integrated approach that merges qualitative insights with quantitative analysis. The efficacy of this advanced model is demonstrated through a case study, highlighting its capability to deliver accurate and reliable predictions that are in harmony with practical engineering scenarios. This research not only o7ers a robust framework for analyzing dam deformation trends but also sets a new standard in the 6eld, providing a new solution for assessing structural integrity in hydrological engineering. [ABSTRACT FROM AUTHOR]

In the past few years, there has been a notable interest in the application of machine learning methods to enhance energy efficiency in the smart building industry. The paper discusses the use of machine learning in smart buildings to improve energy efficiency by analyzing data on energy usage, occupancy patterns, and environmental conditions. The study focuses on implementing and evaluating energy consumption prediction models using algorithms like long short-term memory (LSTM), random forest, and gradient boosting regressor. Real-life case studies on educational buildings are conducted to assess the practical applicability of these models. The data is rigorously analyzed and preprocessed, and performance metrics such as root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) are used to compare the effectiveness of the algorithms. The results highlight the importance of tailoring predictive models to the specific characteristics of each building's energy consumption. [ABSTRACT FROM AUTHOR]

In recent times, the significance of renewable energy generation has increased and photovoltaic-thermoelectric (PV-TE) technologies have emerged as a promising solution. However, the incorporation of these technologies still faces difficulties in energy storage and optimization. This review paper addresses these challenges by providing a comprehensive overview of the latest advancements in PV-TE technologies. The paper emphasizes the integration of phase change materials (PCMs) for thermal energy storage, also buttressing the use of encapsulated PCM for thermal storage and efficiency, and the use of hybrid PCM to enhance overall performance. Furthermore, reviews on the use of machine learning techniques for efficient optimization and the integration of thermoelectric modules into tandem perovskite silicon solar cells have been comprehensively analyzed. The advancements in photovoltaic-thermoelectric systems, as reviewed in this article, signify significant progress in attaining sustainable and effective energy production and storage. This review comprehensively addresses the 4Es, underlining their importance. It not only consolidates recent developments but also charts a path for future research in the field of PV-TE technologies, offering precise insights to guide upcoming studies and innovations. [ABSTRACT FROM AUTHOR]

Machine learning (ML) algorithms are increasingly used in power systems applications. One important application is the classification and localization of various types of transmission line faults. Using voltage and current measurements from phasor measurement units (PMUs), a number of useful features can be extracted, which can form the basis of a ML-based prediction of the fault type, line, and distance on the line. This paper proposes a technique to find the optimal number and placement of PMUs by performing thorough feature selection. The features are selected to maximize the accuracy of the ML classification and regression algorithms. The results show that for the IEEE 14 bus system, the use of only five PMUs is sufficient to obtain high levels of accuracy. For example, a testing accuracy of 99.0% and 97.1% can be achieved for the fault type and fault line location, respectively. As for the fault distance along the line, the testing MAE of 3.1% can be obtained along with an R 2 score of 94.4%. Adding more PMUs does not provide any additional value in terms of accuracy. [ABSTRACT FROM AUTHOR]

The respiratory disease of coronavirus disease 2019 (COVID-19) has wreaked havoc on the economy of every nation by infecting and killing millions of people. This deadly disease has taken a toll on the life of the entire human race, and an exact cure for it is still not developed. Thus, the control and cure of this disease mainly depend on restricting its transmission rate through early detection. The detection of coronavirus infection facilitates the isolation and exclusive care of infected patients. This research paper proposes a novel data mining system that combines the ensemble feature selection method and machine learning classifier for the effective identification of COVID-19 infection. Different feature selection approaches including chi-square test, recursive feature elimination (RFE), genetic algorithm (GA), particle swarm optimization (PSO), and random forest are evaluated for their effectiveness in enhancing the classification accuracy of the machine learning classifiers. The classifiers that are considered in this research work are decision tree, naïve Bayes, K -nearest neighbor (KNN), multilayer perceptron (MLP), and support vector machine (SVM). Two COVID-19 datasets were used for testing from which the best features supporting the dataset were extracted by the proposed system. The performance of the machine learning classifiers based on the ensemble feature selection methods is analyzed. [ABSTRACT FROM AUTHOR]

The discussion models the IRI-2012 TEC map over a moderate geomagnetic storm period (5 days) in February 2015 and compares the yield of the models. The models are constructed with the help of cubic Bézier curves and machine learning. In a sense, the comparison of a classical and mechanical approach with a modern and computer-based one is a considerable experience for the paper. The parametric curve approach governs models of piecewise continuous Bézier cubics, while the models employ only the TEC map. The design is separated into curve components at every five-hour curvature point, and each component is handled independently. Instead of the traditional least squares method for finding control points of cubics, it utilizes the mean of every five-hour of the piecewise curves of the TEC data. Accordingly, the prediction error can be controlled at a rate that can compete with the modern network approach. In the network model, 120 hours of the solar wind parameters and the TEC map of the storm are processed. The reliability of the network model is assessed by the (R) correlation coefficient and mean square error. In modeling the TEC map with the classical approach, the mean absolute error is 0.0901% and the correlation coefficient (R) score is 99.9%. The R score of the network model is 99.6%, and the mean square error is 0.71958 (TECU) (at epoch 47). The results agree with the literature. [ABSTRACT FROM AUTHOR]

Flight delays are a major concern for both travelers and airlines, with significant financial and reputational consequences. Accurately predicting flight delays is crucial for enhancing customer satisfaction and airline revenues. In this paper, we leverage the power of artificial intelligence and machine learning techniques to build a framework for accurately predicting flight delays. To achieve this, we collected flight information from September 2017 to April 2023, along with weather data, and performed extensive feature engineering to extract informative features to train our model. We conduct a comparative analysis of various popular machine learning architectures with distinctive characteristics, aiming to determine their efficacy in achieving optimal accuracy on our newly proposed dataset. Based on our evaluation of various architectures, our findings demonstrate that CatBoost outperformed the others by achieving the highest test accuracy and the lowest error rate in the challenging use case of Saudi Arabia. Moreover, to simulate real-world scenarios, our framework evaluates the best-performing model that has been selected for deployment in a web application, which provides users with the ability to accurately forecast flight delays and offers a user-friendly dashboard with valuable insights and analysis capabilities. [ABSTRACT FROM AUTHOR]

This paper proposed a single-piston free piston expander-linear generator (SFPE-LG) prototype applied to organic Rankine cycle systems. Two valve timing control strategies, namely, time control strategy (TCS) and position control strategy (PCS), were developed. Based on the experimental data, a back propagation neural network (BPNN) prediction model was established. The effects of structural parameters such as neural network layers, transfer function, training function, hidden layer nodes, and learning rate on the prediction accuracy of this BPNN model were discussed. The training and prediction accuracy of the BPNN model was verified using 5-fold cross-validation and Wilcoxon signed-rank test. Moreover, the BPNN model was integrated with a genetic algorithm to predict and optimize the maximum output power of the SFPE-LG. The results showed that the BPNN model used to predict the motion characteristics and output performance of the SFPE-LG exhibits strong learning ability and high prediction accuracy. Notably, the prediction accuracy of the BPNN model is significantly higher under the PCS compared to TCS. The effect of hidden layer nodes on mean square error (MSE) and correlation coefficient (R) is greater than that of the learning rate. When the number of hidden layer nodes exceeds 30, the BPNN model consistently achieves low MSE and high R. The optimization results showed that the SFPE-LG can obtain a maximum output power of 141.69 W under the TCS, when the working parameters are inlet pressure of 0.7 MPa, intake duration of 35 ms, load resistance of 67 Ω, and expansion duration of 104 ms, respectively. [ABSTRACT FROM AUTHOR]

Accurate fault detection in high-voltage direct current (HVDC) transmission lines plays a pivotal role in enhancing operational efficiency, reducing costs, and ensuring grid reliability. This research aims to develop a cost-effective and high-performance fault detection solution for HVDC systems. The primary objective is to accurately identify and localize faults within the power system. In pursuit of this goal, the paper presents a comparative analysis of current and voltage characteristics between the rectifier and inverter sides of the HVDC transmission system and their associated alternating current (AC) counterparts under various fault conditions. Voltage and current features are extracted and optimized using a metaheuristic approach, specifically Harris Hawk's optimization method. Leveraging machine learning (ML) and artificial neural networks (ANN), this technique demonstrates its effectiveness in generating a fault locator with exceptional accuracy. With a substantial volume of data employed for learning and training, the Harris Hawks optimization method exhibits faster convergence compared to other metaheuristic methods examined in this study. The research findings are applied to simulate diverse fault types and unknown fault locations at multiple system points. Evaluating the fault detection system's effectiveness, quantified through metrics such as specificity, accuracy, F1 score, and sensitivity, yields remarkable results, with percentages of 99.01%, 98.69%, 98.64%, and 98.67%, respectively. This research underscores the critical role of accurate fault detection in HVDC systems, offering valuable insights into optimizing grid performance and reliability. [ABSTRACT FROM AUTHOR]

With the emergence and vigorous development of 5G technology, there is a significant surge in network usage and traffic, resulting in heightened complexity within network and IT environments. This exponential increase in activity produces a plethora of events, making conventional systems inadequate for the efficient management of 5G networks. In comparison to 4G technology, 5G technology brings forth a host of new features, one of which is the network data analytics function (NWDAF). This function grants network operators the flexibility to either employ their own data analytics methodologies based on machine learning (ML) and deep learning (DL) into their networks. In this paper, we present a dataset named "NWDAF-NFPP" for network function performance time series prediction, collected from a laboratory at China Telecom. The dataset is carefully anonymized to ensure maximum realism and comprehensiveness, while safeguarding sensitive information. It encompasses eight categories of network functions, with data collected at five-minute intervals. The availability of this dataset provides valuable resources for researchers to conduct time series prediction research on network element performance. Following data collection, a total of six models were employed for network element performance prediction, encompassing both machine learning and deep learning approaches. This diverse set of models was carefully chosen to ensure comprehensive coverage of different techniques and algorithms. Through the comparison and analysis of these models, we aim to evaluate their predictive capabilities and identify the most effective approach for network element performance prediction. This comparative analysis will provide valuable insights into the strengths and limitations of each model, aiding in informed decision-making for network optimization and management strategies in the future. [ABSTRACT FROM AUTHOR]