The increasing environmental concerns and depletion of fossil fuel reserves have accelerated the transition toward renewable energy sources, with wind power emerging as a viable and sustainable solution. Efficient maximum power point tracking (MPPT) is essential for optimizing energy extraction, particularly in low and fluctuating urban wind conditions. This paper presents a Deep Q-learning (DQL)-based MPPT algorithm for a permanent magnet synchronous generator (PMSG)-driven hybrid vertical axis wind turbine (VAWT) designed for rooftop applications. The proposed approach eliminates the reliance on mechanical wind speed sensors, enhancing system reliability and responsiveness. Simulation results demonstrate that the DQL-based MPPT method outperforms conventional techniques in terms of accuracy, adaptability, and dynamic response, effectively tracking the maximum power point under varying wind conditions. These findings confirm the effectiveness of the proposed algorithm in improving energy capture for urban wind energy systems.

Vehicle location is crucial for transportation and computer vision. Bounding boxes distinguish vehicles, which is crucial for real-time movement estimation applications requiring precise area data. This study presents an adaptive approach for accurate vehicle detection and tracking in challenging scenarios such as heavy traffic, poor visibility, and adverse weather conditions. The proposed method integrates fuzzy subtraction and gradient partial equation (FGPE) techniques for background subtraction, overcoming fluctuations and shadows. It uses energy and histogram-oriented gradient features, chosen through recursive feature elimination, to improve discrimination capability. Further, a normalisation-based attention module (NAM) is integrated into the Enhanced YOLOv5 model for vehicle detection. The Multi-Object-based DeepSORT algorithm for vehicle tracking enhances the feature extractor. Deployment on edge devices achieves a traffic flow detection accuracy of 0.98%. Evaluation metrics including multiple-object tracking algorithm (MOTA) and multiple-object tracking precision (MOTP) validate the effectiveness of the proposed model for real-world traffic surveillance systems.

This study explores the feasibility of recovering energy from underground mine ventilation fan exhaust while optimising fan performance. Typically, exhaust air is released without energy utilisation. Here, a 700 W hybrid Darrieus-Savonius vertical axis wind turbines (VAWT) is integrated with a 120 HP mine ventilation fan, operating at a tip-speed ratio of 3. The VAWT extracts kinetic energy while generating a localised low-pressure zone, reducing backpressure and enhancing ventilation efficiency. The study includes a literature review, system conceptualisation, aerodynamic modelling and Computational fluid dynamics (CFD) simulations to analyse airflow characteristics, turbine performance and energy recovery potential. Pressure distribution and fan performance assessments highlight the impact of turbine integration on airflow dynamics. Experimental validation through field testing confirms the concept's practicality. Results indicate that 4-5% of the fan's energy consumption can be reduced, and the electrical power generated by the hybrid VAWT can be reintegrated into the grid or stored in batteries. This approach not only reduces energy consumption and operational costs but also supports sustainable mining operations.

The present study explores the wake dynamics and synergistic power generation potential of a hybrid Savonius–Darrieus vertical axis wind turbine (VAWT) positioned in front of a mine ventilation fan. By leveraging the combined airflow from the ventilation system and ambient wind, the feasibility of clustering-based power generation is assessed. Experimental investigations with a single turbine, coupled with computational fluid dynamics (CFD) simulations of a multi-turbine configuration, reveal that the hybrid VAWT significantly alters wake structure, enhancing turbulence and airflow mixing. The proposed turbine cluster configuration demonstrates improved energy extraction efficiency while maintaining stable ventilation performance without imposing excessive aerodynamic loads. The proposed wind power generation strategy utilizing clustering improves power output, boosts the cluster power coefficient (Cp) by 1.1 times, and enhances ventilation efficiency. These findings provide a novel framework for integrating renewable energy solutions into industrial ventilation systems, offering a pathway toward enhanced sustainability and energy efficiency.

In light of India's current energy crisis, it has been determined that the coal mining industry may be used to implement Demand Side Management (DSM) applications. This study explores the possibility of energy savings and reduction in energy costs associated with the installation of (Variable Speed Drives) VSDs on primary ventilation fans in below-ground coal mines. According to Ventilation-on-Demand (VOD), where air volume is changed according to demand at various time intervals, a nonlinear constraint optimization model is created to reduce energy costs and achieve better energy efficiency. To achieve accurate results at the fan's operating point, this model is also constructed to adhere to the affinity law of the fan. Tellegen's theorem and Kirchhoff's law are applied to model the ventilation system of the underground mine. The ventilation network was used to investigate energy and financial savings with various airflow scenarios. Load management is accomplished by figuring out the ideal start time for the mining schedule in accordance with the Time of Use (TOU) tariffs. A critical analysis of the Tandsi underground coal mine is presented to explain the impacts of the optimization model. The study found that implementing DSM solutions could result in a total annual energy savings of 118625 kWh, or a reduction in energy costs of USD 22995.

The market share of Electric Vehicles (EVs) is steadily increasing, benefiting the environment and energy crisis. However, the widespread adoption of EVs can negatively impact the smart grid, leading to stress, frequency swings, and power variations. To address this, a multi-polynomial regression-based model predictive controller using the XGBOOST + Relief + AOA technique is developed. It enhances the performance of dispersion frameworks while managing Energy and Voltage Control (EVC) for EV coordination. Initial information outlines input data, including harmonic distortion, unbalanced grids, transformer lifespan, voltage swings, and power loss. Data processing resolves issues, such as improper scaling, missing data, and imbalance. An oversampling technique based on aggregative clustering handles imbalanced data, and independent features are standardized using the concordance correlation coefficient-based power transform (CCC–PT). The proposed EV charging model utilizes an optimization objective function, incorporating XGBOOST, Relief, and AOA techniques. The resulting optimization maintains a balance between exploration and exploitation, minimizing errors and time complexity. The multipolynomial regression-based model predictive controller focuses on voltage and energy control to ensure charge balance. Overall, this implementation optimizes distributed state functioning by considering the impact of EV coordination on EVC. The effectiveness and applicability of this methodology are evaluated across three separate scenarios.

The underground coal mine (UCM) dynamic and complex environment impose various hazards that significantly affect the mining production and safety of the personnel. Flammable and poisonous gases significantly contribute to many fatal accidents. This study proposes a real-time-based reliable gas hazard monitoring system using multisensor data fusion. A hybrid of CNN-LSTM-based deep neural network (HCLM) is developed to serve the purpose. Due to the challenging environment of the UCM, sensor malfunctioning is inevitable and severely affects the performance of HCLM. A novel front-end filter (FEF) is developed based on Damper Shafer’s (DS) theory and belief divergence-based weighted credibility metric to overcome the drawback of HCLM. In the laboratory trial, it is observed that the hazard classification accuracy of HCLM for the faulty node scenarios is 85%. In contrast, the accuracy of the HCLM integrated with FEF is maintained at 98%, even for multiple faulty node cases. Another novelty of this study is the tinyML implementation of the proposed model. Due to UCM’s inherent complexities and challenges, traditional wireless communications face operational difficulties. Hence, a cloud-based machine learning operation is not a feasible option in UCM. Hence, using the concept of tinyML, the proposed model is directly deployed on a microcontroller near the data sources, thereby reducing network latency and security issues.

AbstractThis article explores the potential of regeneration of power from unnatural wind sources with the help of vertical axis wind turbines (VAWT). Researchers are searching for urban as well as industrial wind sources, which can be useful as the natural wind source to obtain electrical energy and utilize it. The wind sources considered here are the exhaust of the cooling or ventilation fans used in buildings, industries, and other places. The Darrieus VAWT extracts wind power from the exhaust air and reduces the power consumption of the concerned electrical drives. These uncommon energy sources have an inherent capability to recover a significant amount of energy without polluting the environment with less payback period. Recovered energy can be suitably converted into electrical energy and may be fed back to the power grid without any pollution, thus minimizing the total energy consumption of the building and reducing the cost operations. A detailed study of experimental models with different types of assembly and turbine models is conducted here with power outputs and operational behaviors on the existing systems. Various parameters and factors for existing and new applications are in detail that show the system has no negative impact on regular operation.

In this letter, we introduce a novel cluster head selection algorithm namely mixed grey wolf and improved sunflower optimization algorithm (MGWISFO). This algorithm leverages both energy requirements and inter-node distances to select cluster heads (CH). Within this algorithm, the Grey Wolf Optimizer facilitates exploration, offering a broader search, while the improved Sunflower Optimization focuses on exploitation, delivering a narrower search. This balance between exploration and exploitation leads to the identification of the optimal CH node, thereby enhancing network performance. To validate its effectiveness, the proposed algorithm is benchmarked against existing strategies such as particle swarm optimization (PSO), genetic algorithm (GA), grey wolf optimization (GWO), and sunflower optimization (SFO) across various performance parameters including throughput, the number of live and dead nodes, and residual energy. Simulation results unequivocally establish the unparalleled performance of our proposed algorithm, surpassing the capabilities of existing algorithms.

Small-scale wind energy conversion systems (WECSs) present a promising solution for rural areas and urban settings where the installation of sizable wind turbine is impractical due to space constraints and environmental concerns. The paper suggests an adaptable step size MPPT approach utilizing only one load current sensor. By utilizing load current data, the method can track maximum power without relying on rotor speed details or turbine parameter understanding. The MPPT system employs a straightforward and economical uncontrolled rectifier setup along with DC-DC boost converters for the PMSG-based Variable-Speed WECS and a load bank. This setup is then simulated. Following this, a wind speed profile is generated with step variations to evaluate the effectiveness of the MPPT algorithm. MPPT is realized by modifying the duty signal of the BC, facilitating adjustments in generator speed corresponding to varying wind speeds. Results demonstrate successful power extraction from the WECS, with detailed outcomes presented.

Low voltage ride-through (LVRT) is one of the essential aspects of grid codes for integrating doubly-fed induction generators (DFIG) to achieve reliable and uninterrupted electrical power generation. The detection time of the voltage sag is one of the crucial aspects for the LVRT improvement of the DFIG. This paper introduces a second-order generalized integrator (SOGI), quadrature signal generator (QSG) with a frequency locked loop (FLL) based algorithm to detect the symmetrical voltage sag fast and accurately. The proposed SOGI-QSG-FLL-based voltage sag detection algorithm (VSDA) works under a second-order band-pass filter with an alfa-beta stationary reference frame. The proposed VSDA is integrated with the novel feed-forward transient current compensation (FFTCC) scheme with an advanced interval type-2 fuzzy logic-proportional integral (IT2-FLC-PI) controller to enhance the LVRT capability of the DFIG. This FFTCC scheme is activated after the detection of voltage sag by using the proposed VSDA. The proposed FFTCC feeds the uninterrupted active and reactive power to the grid and reduces the torque ripple of the DFIG. The detection time of voltage sag by using the proposed VSDA is compared with existing algorithms. The experimental results are also presented to validate the proposed algorithm.

The underground coal mines (UCM) exhibit many life-threatening hazards for mining workers. In contrast, gas hazards are among the most critical challenges to handle. This study presents a comparative study of the sensor fusion methodologies related to UCM gas hazard prediction and classification. The study provides a brief theoretical background of the existing methodologies and their usage to mitigate the gas hazard issues in UCM. A brief comparison report emphasising the advantages and disadvantages of the existing models related to the UCM gas hazard monitoring is presented. Additionally, a separate comparison is also drawn, considering only neural network models based on their prediction accuracy and other performance metrics. This study attempts to observe and compare the Neural network models with the conventional method in the field of UCM gas hazard prediction, which is not explored in this fraternity.

Underground coal gasification (UCG) is a complex process as it depends on many factors such as geological and geo-hydrological analysis of strata, physical and chemical properties of coal, operational process parameters, seismic events, analysis of produced gases in real-time. This paper mainly identifies the process parameter variations due to underground coal gasification. This scientific study could help us to choose appropriate sensors, and design and develop suitable monitoring units for the control, monitoring and precaution during coal gasification. In this article, one experimental work is carried out in laboratory scale with the help of developed system. This paper also described individual control and monitoring techniques of different operational process parameters required for successful UCG operation. This is helpful for successful operation in the actual field and reducing its complexity.

Building a reliable communication network is a challenging task using the standard Ethernet as the medium for control and data acquisition networks. Such sensor-actuator networks could be distributed across multiple locations across a building. Packet retransmissions and losses could be critical when Ethernet is used for closed-loop controls and measurements. There could be multiple design challenges to build such networks such as total-traffic, compact node design, bandwidth limitations, packet retransmissions, delays, drops etc. In this paper, we report how we implemented a reliable http-based automation network of large interconnected microcontroller-based nodes with a careful design of hardware, firmware components and a new application-layer faulty-node-filter-algorithm. It analysed some of the top causes of poor network performance within such control networks using a popular network protocol-analyser “Wireshark”. Though Ethernet seemed the most future promising, it has the single drawback of nondeterministic packet transfers. Additional research is carried out to prove that IEEE 802.3 standard Ethernet could be used in applications like our slow-cryogenics-control systems of linear Accelerators to build an error-free, near deterministic control-application. This approach allowed the continued use of conventional Ethernet networking components. even for the control networks and the same communication medium. The tested system is distributed within a private LAN of more than 50 embedded servers each of which is built out of on-chip TCP/IP stacks implemented on ARM processors as nodes, for a successfully implemented Cryogenic control network of Superconducting linear Accelerator at Inter University Accelerator Centre, New Delhi, India.

The importance of the quality of life of rotating machinery increases the Bearing fault diagnosis. Deep learning models (DL)-based databases become increasingly smart in the field of fault diagnostics, the latest research has widely used CNNs (convolutional neural networks). This paper proposes a new way to diagnose bearing failures with CNN with Bilinear LSTM. Traditional CNNs are however not easy to detect defects due to the fixed geometry of complex fault diagnosis with different working conditions. Our primary and secondary classifiers at specified layers replace primitive shape convolutions with reconfigurable convolutions, resulting in classification results with stringent feature time-frequency incompatibility and a larger receptive field. To acquire more adaptive knowledge and insight into the proposed approach, we employ the CWRU (Case Western Reserve University) opensource dataset to compare classification accuracy. The bearing dataset has been subjected to comprehensive experiments and evaluations in order to confirm the efficacy of the suggested technique's diagnostic performance in a variety of settings. By comparing multiple perspectives on the same dataset with related tasks, the proposed method's superiority is proved. To limit the effect of noise and avoid temporal oscillations, degraded index sequences are matched with a CNN. Current and previous inspection data are fed into a new CNN-BiLSTM model, which is then used to predict the useful time and compatible power values of bearing RULs. When it comes to output, go with the lifetime percentage. The proposed method has been tested by accelerating bearing operation to failure, and the results show that the method has advantages in predicting RUL more accurately. The results of the experiments suggest that the proposed core distance measurement method is a viable new tool for intelligent rolling bearing diagnosis. The BiLSTM technique is more diagnostic than some generic models, according to experimental results using the 48 K and 12 K CWRU datasets, with overall accuracy of 99.80% and 98.3%, respectively.

In this paper, a new approach is presented that utilizes a least mean square root of exponential (LMSRE)-based adaptive filtering algorithm to enhance the transient stability of doubly fed induction generator (DFIG) based wind turbines through the static synchronous compensator (STATCOM). The control strategy of the STATCOM relies on the voltage source converter (VSC). The proposed method involves the online adjustment of the gain of all the proportional-integral (PI) controllers in the VSC's cascaded structure using the LMSRE algorithm, which ensures rapid convergence. The selection of the LMSRE algorithm in this study is motivated by its quick convergence rate, minimum computational intricacy, and straightforward structure. To assess the effectiveness of the proposed LMSRE-PI based adaptive controlled STATCOM, its performance is compared with a PI controller based on the least mean square (LMS) adaptive algorithm under symmetrical voltage sag conditions. The effectiveness of the proposed adaptive controlled STATCOM is verified through the utilization of MATLAB/Simulink software, thereby validating its performance.

The illumination systems in underground coal mines in India, generally operating in unfavourable, hazardous, and restricted environments, are badly in need of updating and revamping as they play a vital role in mine safety and productivity. Moreover, person-to-person communication in underground coal mines is very inadequate. In view of limited resources, the provision of illumination and communication using the same infrastructure forms an attractive proposition. As wireless communication is not a viable alternative because of the high attenuation profile there, wired personal communication employing the proposed DC distribution system along with LED illumination proves to be a very efficient and economical approach. The proposed technique establishes the design and implementation of the very low voltage DC (LVDC) power distribution system especially for mine illumination in consort with power line voice communication for underground coal mines. The environment of an Indian underground coal mine is simulated inside a sub-way tunnel and a suitable model designed. An experimental study is made to validate the technical aspects. The major contribution of the present work is a novel solution comprising of a low-cost mine-safe illumination system along with essential voice communication by the same underground infrastructure.

A novel application of adaptive fuzzy logic controller (AFLC) is proposed in this paper for low voltage ridethrough (LVRT) enhancement of the grid-connected doubly fed induction generator (DFIG)-based wind energy conversion system (WECS). A cascaded adaptive fuzzy logic control methodology is applied to regulate the rotor side converter (RSC) and grid side converter (GSC) to improve the performance of the DFIG-WECS. A new application of the generalized variable step size continuous mixed p-norm algorithm (GVSS-CMPN) is proposed as an adaptive filtering algorithm (AFA) that modifies online the calibrating factors of the fuzzy logic controllers (FLCs) at a fast convergence speed along with low normalized misalignment error (NME). The GVSS-CMPN algorithm is proposed in this work due to its fast convergence speed, low computational burden, low computational complexity, and high accuracy. The proposed GVSS-CMPN-based FLC feeds the smooth active and reactive power to the grid during severe disturbances. The transient rotor current, inrush stator current, torque ripple, and dc-link over voltage during symmetrical voltage sag are reduced by using the proposed GVSS-CMPNbased FLC. The novel GVSS-CMPN-based FLC is experimentally compared with the CMPN-FLC under severe grid disturbances. Experimental results have been presented to validate the proposed GVSS-CMPN-based adaptive control scheme.

This paper describes the novel mathematical modeling of the 3-phase induction motor, various operations and study of its dynamic behavior and regenerative braking based on the model. This model uses v/f scheme to apply braking operation with the energy flow to the supply system instead of wasting the energy in braking resistor. It discusses the theory of induction motors, which is explored through both equations and computer simulation model using SIMULINK. The model also helps to study the behavior of the motor during its starting and load variation. The results from the analysis prove the demonstration of regenerative braking and the dynamic behavior and can give an opportunity to learn the different characteristics during these conditions.

This research presents a novel utilization of the fractional-order least mean p-norm (FOLMP) adaptive filtering algorithm (AFA) to enhance the capability of a grid-connected doubly fed induction generator (DFIG)-based wind energy conversion system (WECS) to withstand low voltage ride-through (LVRT) situations. The proposed method includes dynamic modifications of the scaling parameters of the adaptive fuzzy logic controllers (AFLCs) to achieve fast convergence. Within a cascaded structure, the rotor side converter (RSC) and grid side converter (GSC) utilize an adaptive fuzzy logic control topology. The RSC focuses on minimizing active power loss, while the GSC ensures a constant dc-link voltage. To assess its effectiveness during severe grid disturbances, the proposed adaptive FOLMP-FLC is compared to the least mean p-norm (LMP) based FLC and the variable step size normalized least mean p-norm (VSS-NLMP) based FLC. MATLAB simulation results demonstrate that the proposed FOLMP-FLC surpasses the LMP and VSS-NLMP approaches in terms of convergence speed, normalized misalignment error (NME), and computational load, thereby enhancing the performance of the DFIG.