Search Results (45)

The increasing complexity of manufacturing processes demands accurate defect prediction and interpretable insights into the causes of quality issues. This study proposes a methodology integrating machine learning, clustering, and Explainable Artificial Intelligence (XAI) to support defect analysis and quality control in industrial environments. Using a dataset based on empirical industrial distributions, we train an XGBoost model to classify high- and low-defect scenarios from multidimensional production and quality metrics. The model demonstrates high predictive performance and is analyzed using five XAI techniques (SHAP, LIME, ELI5, PDP, and ICE) to identify the most influential variables linked to defective outcomes. In parallel, we apply Fuzzy C-Means and K-means to segment production data into latent operational profiles, which are also interpreted using XAI to uncover process-level patterns. This approach provides both global and local interpretability, revealing consistent variables across predictive and structural perspectives. After a thorough review, no prior studies have combined supervised learning, unsupervised clustering, and XAI within a unified framework for manufacturing defect analysis. The results demonstrate that this integration enables a transparent, data-driven understanding of production dynamics. The proposed hybrid approach supports the development of intelligent, explainable Industry 4.0 systems. Full article

Power systems need to meet the ever-increasing demand for higher quality and reliability of electricity in distribution systems while remaining sustainable, secure, and economical. The globe is moving toward using renewable energy sources to provide electricity. An evaluation of the influence of artificial intelligence (AI) on the accomplishment of SDG7 (affordable and clean energy) is necessary in light of AI’s development and expanding impact across numerous sectors. Microgrids are gaining popularity due to their ability to facilitate distributed energy resources (DERs) and form critical client-centered integrated energy coordination. However, it is a difficult task to integrate, coordinate, and control multiple DERs while also managing the energy transition in this environment. To achieve low operational costs and high reliability, inverter control is critical in distributed generation (DG) microgrids, and the application of artificial neural networks (ANNs) is vital. In this paper, a power management strategy (PMS) based on Inverter Control and Artificial Neural Network (ICANN) technique is proposed for the control of DC–AC microgrids with PV-Wind hybrid systems. The proposed combined control strategy aims to improve power quality enhancement. ensuring access to affordable, reliable, sustainable, and modern energy for all. Additionally, a review of the rising role and application of AI in the use of renewable energy to achieve the SDGs is performed. MATLAB/SIMULINK is used for simulations in this study. The results from the measures of the inverter control, m, VL-L, and Vph_rms, reveal that the power generated from the hybrid microgrid is reliable and its performance is capable of providing power quality enhancement in microgrids through controlling the inverter side of the system. The technique produced satisfactory results and the PV/wind hybrid microgrid system revealed stability and outstanding performance. Full article

With the continuous development of industrial intelligence, the integration of cyber–physical components creates a need for effective attack detection methods to mitigate potential DDoS threats. Although several DDoS attack detection modeling approaches have been proposed, few effectively incorporate the unique characteristics of industrial control system (ICS) architectures and traffic patterns. This paper focuses on DDoS attack detection within cloud–edge collaborative ICSs and proposes a novel detection model called FedDynST. This model combines federated learning and deep learning to construct feature graphs of traffic data. Introducing dynamic and static adjacency matrices, this work reveals the interactions between long-term industrial traffic data and short-term anomalies associated with DDoS attacks. Convolutional neural networks are utilized to capture distinctive temporal features within industrial traffic, thereby improving the detection precision. Moreover, the model enables continuous optimization of the global detection framework through a federated learning-based distributed training and aggregation mechanism, ensuring the privacy and security of industrial client data. The effectiveness of the FedDynST model was validated on the CICDDoS2019 and Edge-IIoTset datasets. The simulation results validated the superiority of the proposed approach, and thus, demonstrated significant improvements in both detection accuracy and convergence. Full article

To maximize arterial green wave bandwidth utilization, this study aims to minimize average travel delays at coordinated intersections and maximize vehicle throughput. In view of the aforementioned points, the present paper sets out a collaborative optimization method for the control of related intersection groups. The method combines multi-level speed guidance with green wave coordinated control. In an intelligent and connected environment (ICE), the driving trajectory of the initial vehicle is determined in each optimization cycle following the receipt of active speed guidance. Subsequently, the driving trajectories of subsequent vehicles are calculated, with an assessment made as to whether they can leave the intersection before the end of the green light. The subsequent step involves the calculation of a characteristic index, comprising the average speed of the arterial coordination section and its corresponding phase offset. The phase offset is then optimized with the objective of maximizing the comprehensive bandwidth of green wave coordination within the control range. The maximum average speed and the bidirectional cycle comprehensive green wave bandwidth are employed as the control objectives. Finally, a model is constructed through the combination of multi-level vehicle speed guidance with bidirectional cycle green wave coordinated control. A bi-level combinatorial optimization method is constructed through a combinatorial deep Q learning method, named Deep Q Network-Genetic Algorithm (DQNGA), with the objective of obtaining the global optimal solution. Finally, the reliability of the method is validated using traffic flow data and map sensor data on several associated road sections in a city. The results demonstrate that the proposed method reduces the average delay and number of stops by 20.76% and 44.49%, respectively, outperforming conventional traffic control strategies. This suggests that the issue of inefficient utilization of green light time in arterial coordinated signal control has been effectively addressed. Consequently, the efficiency of intersections in the intelligent and connected environment has been enhanced. Full article

The information interaction characteristics of connected vehicles are distinct from those of non-connected vehicles, thereby exerting an influence on the conventional traffic flow model. The original lane-changing model for non-connected vehicles is no longer applicable in the context of the new traffic flow environment. The modelling of the new hybrid traffic flow, comprising both connected and ordinary vehicles, is set to be a pivotal research topic in the coming years. The objective of this paper is to present a methodology for optimal mixed traffic flow dynamic modelling and cooperative control in intelligent and connected environments (ICE). The study utilizes the real-time perception and information interaction of connected vehicles for traffic information, taking into account the access characteristics of both connected and non-connected vehicles. The satisfaction-based free lane-changing and mandatory lane-changing models of connected vehicles are designed. Secondly, a mixed traffic flow lane-changing model based on influence characteristics is constructed for the influence area of connected vehicles. This model takes into account the degree of influence that connected vehicles have on non-connected vehicles, with different distances being considered respectively. Subsequently, a vehicle guidance strategy for mixed traffic flows comprising grid-connected and conventional vehicles is proposed. A variety of speed guidance scenarios are considered, with an in-depth analysis of the speed optimization of connected vehicles and the movement law of non-connected vehicles. This comprehensive analysis forms the foundation for the development of a vehicle guidance strategy for mixed traffic flows, with the overarching objective being to minimize the average delay of vehicles. In order to evaluate the effectiveness of the proposed method, the intersection of Gaota Road and Fangshui North Street in Yanqing District, Beijing, has been selected for analysis. The results of the study demonstrate that by modifying the density of the mixed traffic flow, the overall average speed of the mixed traffic flow declines as the density of vehicles increases. The findings reported in this study reflect the role of connected vehicles in enhancing road capacity, maximizing intersection capacity and mitigating the occurrence of queuing phenomena, and improving travel speed through the mixed traffic flow lane-changing model based on impact characteristics. This study also provides some guidance for future control of the mixed traffic flow formed by emergency vehicles and social vehicles and for realizing a smart city. Full article

Industrial control systems (ICSs), which are fundamental to the operation of critical infrastructure, face increasingly sophisticated security threats due to the integration of information and operational technologies. Conventional anomaly detection techniques often lack the ability to provide clear explanations for their detection, and their inherent complexity can impede practical implementation in the resource-constrained environments typical of ICSs. To address these challenges, this paper proposes a novel approach that leverages swarm intelligence algorithms for the extraction of numerical association rules, specifically designed for anomaly detection in ICS. The proposed approach is designed to effectively identify and precisely localize anomalies by analyzing the states of sensors and actuators. Experimental validation using the Secure Water Treatment (SWaT) dataset demonstrates that the proposed approach can detect over 84% of attack instances, with precise anomaly localization achievable by examining as few as two to six sensor or actuator states. This significantly improves the efficiency and accuracy of anomaly detection. Furthermore, since the method is based on the general control dynamics of ICSs, it demonstrates robust generalization, making it applicable across a wide range of industrial control systems. Full article

The optimization of photovoltaic (PV) systems is vital for enhancing efficiency and economic viability, especially under Partial Shading Conditions (PSCs). This study focuses on the development and comparison of traditional and advanced algorithms, including Perturb and Observe (P&O), Incremental Conductance (IC), Fuzzy Logic Control (FLC), Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO), and Artificial Neural Networks (ANN), for efficient Maximum Power Point Tracking (MPPT). Simulations conducted in the MATLAB/Simulink software package evaluated these algorithms’ performances under various shading scenarios. The results indicate that, while traditional methods like P&O and IC are effective under uniform conditions, advanced techniques, particularly ANN-based MPPT, exhibit superior efficiency and faster convergence under PSCs. This study concludes that integrating Artificial Intelligence (AI) and Machine Learning (ML) into MPPT algorithms significantly enhances the reliability and efficiency of PV systems, paving the way for a broader adoption of solar energy technologies in diverse environmental conditions. These findings contribute to advancing renewable energy technology and supporting green energy transition. Full article

The use of electric automobiles, or EVs, is essential to environmentally conscious transportation. Battery EVs (BEVs) are predicted to become increasingly accepted for passenger vehicle transportation within the next 10 years. Although enthusiasm for EVs for environmentally friendly transportation is on the rise, there remain significant concerns and unanswered research concerns regarding the possible future of EV power transmission. Numerous motor drive control algorithms struggle to deliver efficient management when ripples in torque minimization and improved dependability control approaches in motors are taken into account. Control techniques involving direct torque control (DTC), field orientation control (FOC), sliding mode control (SMC), intelligent control (IC), and model predictive control (MPC) are implemented in electric motor drive control algorithms to successfully deal with this problem. The present study analyses only sophisticated control strategies for frequently utilized EV motors, such as the brushless direct current (BLDC) motor, and possible solutions to reduce torque fluctuations. This study additionally explores the history of EV motors, the operational method between EM and PEC, and EV motor design techniques and development. The future prospects for EV design include a vital selection of motors and control approaches for lowering torque ripple, as well as additional research possibilities to improve EV functionality. Full article

The semiconductor industry is being radically impacted by the placing of greater emphasis on the development of hetero-devices and systems that will act as essential drivers for a wide spectrum of technological applications. The introduction of new materials and their integration with currently used materials are projected to replace integrated circuitry (IC) design and device scaling as the key enablers to the realization of improved device performance and larger density gains. Yet material selection has been constrained by existing fabrication process technology. To date, fabrication processes have dictated material selection by limiting chemical sources or precursors to those that match existing process tools associated with chemically based vapor phase processes and their variants, which in turn limits material compositions in ICs. The processing and integration of new materials compositions and structures will require the introduction of new deposition and etching processes, and manufacturing worthy tool designs and associated protocols that provide new methods for atomic-level control. To this end, a novel manufacturing paradigm is presented comprising a method and system for real-time, closed-loop monitoring and control of synthesis, supply, and consumption of precursors in process intensification techniques including chemical vapor deposition (CVD), atomic layer deposition (ALD), atomic layer etching (ALE), and other IC manufacturing processes. This intelligent automated manufacturing approach is consistent with a central component of the semiconductor industry’s recent adoption of Industry 4.0., including vertical integration of IC manufacturing through robotization, artificial intelligence, and cloud computing. Furthermore, the approach eliminates several redundant steps in the synthesis, handling, and disposal of source precursors and their byproducts for CVD, ALD, ALE and other chemically based manufacturing processes, and thus ultimately lowers the manufacturing cost for both conventional and new IC materials. Further, by eliminating the issues associated with precursor thermal, chemical, and pyrophoric instabilities, this new paradigm enables the deposition of a myriad of new thin-film materials and compositions for IC applications that are practically unattainable with existing precursors. Preliminary and planned demonstration examples for the generation and deposition of highly toxic and unstable source precursors are provided. Full article

During road construction, the accuracy of compaction work is critical for the structural stability and maintenance of the road. Although the plate load test (PLT) is commonly used for quality inspections, it is impractical to test every section due to time and cost constraints. Therefore, simpler testing methods are being extensively developed. This study compared quality inspection results using the commonly used PLT, the relatively simple dynamic cone penetrometer test (DCPT), the lightweight deflectometer (LWD) test, and an intelligent quality control system equipped with accelerometer and global positioning system (GPS) sensors in intelligent compaction (IC) rollers. The results showed a strong correlation between the conventional tests (PLT, DCPT, and LWD) and the values obtained from the intelligent quality control system. The correlation analysis between the intelligent quality control system and PLT, LWDT, and DCPT yielded R-square values of 0.69, 0.91, and 0.95, respectively, indicating significantly high correlations. The implementation of intelligent quality management systems in earthwork construction projects will facilitate a thorough verification of the compaction quality throughout all construction segments, ensuring consistent compaction across the project. By enabling real-time data acquisition and analysis, these systems differ markedly from traditional methods, reducing the frequency and necessity of manual inspections. This approach not only streamlines construction processes, but also enhances operational efficiency. As a result, integrating these intelligent systems is anticipated to significantly increase productivity by optimizing the workflow and resource utilization in earthwork construction. Full article

The objective of this research is to achieve biologically autonomous control by utilizing a whole-brain network model, drawing inspiration from biological neural networks to enhance the development of bionic intelligence. Here, we constructed a whole-brain neural network model of Caenorhabditis elegans (C. elegans), which characterizes the electrochemical processes at the level of the cellular synapses. The neural network simulation integrates computational programming and the visualization of the neurons and synapse connections of C. elegans, containing the specific controllable circuits and their dynamic characteristics. To illustrate the biological neural network (BNN)’s particular intelligent control capability, we introduced an innovative methodology for applying the BNN model to a 12-legged robot’s movement control. Two methods were designed, one involving orientation control and the other involving locomotion generation, to demonstrate the intelligent control performance of the BNN. Both the simulation and experimental results indicate that the robot exhibits more autonomy and a more intelligent movement performance under BNN control. The systematic approach of employing the whole-brain BNN for robot control provides biomimetic research with a framework that has been substantiated by innovative methodologies and validated through the observed positive outcomes. This method is established as follows: (1) two integrated dynamic models of the C. elegans’ whole-brain network and the robot moving dynamics are built, and all of the controllable circuits are discovered and verified; (2) real-time communication is achieved between the BNN model and the robot’s dynamical model, both in the simulation and the experiments, including applicable encoding and decoding algorithms, facilitating their collaborative operation; (3) the designed mechanisms using the BNN model to control the robot are shown to be effective through numerical and experimental tests, focusing on ‘foraging’ behavior control and locomotion control. Full article

Intelligent compaction (IC) has emerged as a breakthrough technology that utilizes advanced sensing, data transmission, and control systems to optimize asphalt pavement compaction quality and efficiency. However, accurate assessment of compaction status remains challenging under real construction conditions. This paper reviewed recent progress and applications of smart sensors and machine learning (ML) to address existing limitations in IC. The principles and components of various advanced sensors deployed in IC systems were introduced, including SmartRock, fiber Bragg grating, and integrated circuit piezoelectric acceleration sensors. Case studies on utilizing these sensors for particle behavior monitoring, strain measurement, and impact data collection were reviewed. Meanwhile, common ML algorithms including regression, classification, clustering, and artificial neural networks were discussed. Practical examples of applying ML to estimate mechanical properties, evaluate overall compaction quality, and predict soil firmness through supervised and unsupervised models were examined. Results indicated smart sensors have enhanced compaction monitoring capabilities but require robustness improvements. ML provides a data-driven approach to complement traditional empirical methods but necessitates extensive field validation. Potential integration with digital construction technologies such as building information modeling and augmented reality was also explored. In conclusion, leveraging emerging sensing and artificial intelligence presents opportunities to optimize the IC process and address key challenges. However, cooperation across disciplines will be vital to test and refine technologies under real-world conditions. This study serves to advance understanding and highlight priority areas for future research toward the realization of IC’s full potential. Full article

The last decade (2013–2023) was the most prolific period of organic Rankine cycle (ORC) research in history in terms of both publications and citations. This article provides a detailed review of the broad and voluminous collection of recent internal combustion engine (ICE) waste heat recovery (WHR) studies, serving as a necessary follow-on to the author’s 2013 review. Research efforts have targeted diverse applications (e.g., vehicular, stationary, and building-based), and it spans the full gamut of engine sizes and fuels. Furthermore, cycle configurations extend far beyond basic ORC and regenerative ORC, particularly with supercritical, trilateral, and multi-loop ORCs. Significant attention has been garnered by fourth-generation refrigerants like HFOs (hydrofluoroolefins), HFEs (hydrofluoroethers), natural refrigerants, and zeotropic mixtures, as research has migrated away from the popular HFC-245fa (hydrofluorocarbon). Performance-wise, the period was marked by a growing recognition of the diminished performance of physical systems under dynamic source conditions, especially compared to steady-state simulations. Through advancements in system control, especially using improved model predictive controllers, dynamics-based losses have been significantly reduced. Regarding practically minded investigations, research efforts have ameliorated working fluid flammability risks, limited thermal degradation, and pursued cost savings. State-of-the-art system designs and operational targets have emerged through increasingly sophisticated optimization efforts, with some studies leveraging “big data” and artificial intelligence. Major programs like SuperTruck II have further established the ongoing challenges of simultaneously meeting cost, size, and performance goals; however, off-the-shelf organic Rankine cycle systems are available today for engine waste heat recovery, signaling initial market penetration. Continuing forward, next-generation engines can be designed specifically as topping cycles for an organic Rankine (bottoming) cycle, with both power sources integrated into advanced hybrid drivetrains. Full article

Data scarcity is a significant obstacle for modern data science and artificial intelligence research communities. The fact that abundant data are a key element of a powerful prediction model is well known through various past studies. However, industrial control systems (ICS) are operated in a closed environment due to security and privacy issues, so collected data are generally not disclosed. In this environment, synthetic data generation can be a good alternative. However, ICS datasets have time-series characteristics and include features with short- and long-term temporal dependencies. In this paper, we propose the attention-based variational recurrent autoencoder (AVRAE) for generating time-series ICS data. We first extend the evidence lower bound of the variational inference to time-series data. Then, a recurrent neural-network-based autoencoder is designed to take this as the objective. AVRAE employs the attention mechanism to effectively learn the long-term and short-term temporal dependencies ICS data implies. Finally, we present an algorithm for generating synthetic ICS time-series data using learned AVRAE. In a comprehensive evaluation using the ICS dataset HAI and various performance indicators, AVRAE successfully generated visually and statistically plausible synthetic ICS data. Full article

Intelligent compaction (IC) is an innovative and modified technology used for quality control in subgrade digital construction. However, current intelligent compaction measurement values (ICMVs) cannot provide accurate measurement of the filling layer when the strength of the underlying layer is relatively high. Experimental field tests conducted in the cut to fill subgrade were performed to collect and analyze the variability of ICMV called the vibration modulus (E_vib). Furthermore, a new ICMV called the modulus of vibration compaction (E_vc) that could remove the interference of the underlying layer and reveal the actual compaction state of filling layer is presented based on the theoretical analysis and numerical simulation method. It was also extracted to study the variability of the filling layer’s compaction state. The results of the above research indicate the influence of the underlying layer’s stiffness on overall compaction quality is remarkable. It was found that it is more likely to achieve the variability control of the compaction state of the filling layer by a new ICMV called E_vc. The measured data, improved approaches, and associated conclusions that are presented provide valuable information for researchers and employees considering the use of IC technology. Full article