The Industrial Internet of Things (IIoT) refers to the use of traditional Internet of Things (IoT) concepts in industrial sectors and applications. IIoT has several applications in smart homes, smart cities, smart grids, connected cars, and supply chain management. However, these systems are being more frequently targeted by cybercriminals. Deep learning and big data analytics have great potential in designing and developing robust security mechanisms for IIoT networks. In this paper, a novel hybrid deep random neural network (HDRaNN) for cyberattack detection in the IIoT is presented. The HDRaNN combines a deep random neural network and a multilayer perceptron with dropout regularization. The proposed technique is evaluated using two IIoT security-related datasets: (i) DS2OS and (ii) UNSW-NB15. The performance of the proposed scheme is analyzed through a number of performance metrics such as accuracy, precision, recall, F1 score, log loss, Region of Convergence (ROC), and Area Under the Curve (AUC). The HDRaNN classified 16 different types of cyberattacks using with higher accuracy of 98% and 99% for DS2OS and UNSW-NB15, respectively. To measure the effectiveness of the proposed scheme, the performance metrics are also compared with several state-of-the-art attack detection algorithms. The findings of HDRaNN proved its superior performance over other DL-based schemes. The deployment perspective of the proposed work is also highlighted in this work.

Introduction. Hybrid electric vehicles are offering the most economically viable choices in today's automotive industry, providing best solutions for a very high fuel economy and low rate of emissions. The rapid progress and development of this industry has prompted progress of human beings from primitive level to a very high industrial society where mobility used to be a fundamental need. However, the use of large number of automobiles is causing serious damage to our environment and human life. At present most of the vehicles are relying on burning of hydrocarbons in order to achieve power of propulsion to drive wheels. Therefore, there is a need to employ clean and efficient vehicles like hybrid electric vehicles. Unfortunately, earlier control strategies of series hybrid electric vehicle fail to include load disturbances during the vehicle operation and some of the variations of the nonlinear parameters (e.g. stator’s leakage inductance, resistance of winding etc.). The novelty of the proposed work is based on designing and implementing two robust sliding mode controllers (SMCs) on series hybrid electric vehicle to improve efficiency in terms of both speed and torque respectively. The basic idea is to let the engine operate only when necessary keeping in view the state of charge of battery. Purpose. In proposed scheme, both performance of engine and generator is being controlled, one sliding mode controllers is controlling engine speed and the other one is controlling generator torque, and results are then compared using 1-SMC and 2-SMC’s. Method. The series hybrid electric vehicle powertrain considered in this work consists of a battery bank and an engine-generator set which is referred to as the auxiliary power unit, traction motor, and power electronic circuits to drive the generator and traction motor. The general strategy is based on the operation of the engine in its optimal efficiency region by considering the battery state of charge. Results .Mathematical models of engine and generator were taken into consideration in order to design sliding mode controllers both for engine speed and generator torque control. Vehicle was being tested on standard cycle. Results proved that, instead of using only one controller for engine speed, much better results are achieved by simultaneously using two sliding mode controllers, one controlling engine speed and other controlling generator torque.

This paper highlights the design of a controller established on estimated states for one-sided Lipschitz (OSL) nonlinear systems subject to output and input delays. The controller has been devised by involving Luenberger-like estimated states. The stability of the time-delayed nonlinear system is reckoned by assuming a Lyapunov functional for delayed dynamics and for which a delay-range dependent criterion is posed with a delay ranging between known upper and lower bounds. The time derivative of the functional is further exploited with linear matrix inequality (LMI) procedures, and employing Wirtinger’s inequality for the integral terms instead of the traditional and more conservative Jensen’s condition. Moreover, a sufficient and necessary solution is derived for the proposed design by involving the tedious decoupling technique to attain controller and observer gain simultaneously. The proposed methodology validates the observer error stability between observers and states asymptotically. The solution of matrix inequalities was obtained by employing cone-complementary linearization techniques to solve the tiresome constraints through simulation tools by convex optimization. Additionally, a novel scheme of an observer-based controller for the linear counterpart is also derived for one-sided Lipschitz nonlinear systems with multiple delays. Finally, the effectualness of the presented observer-based controller under input and output delays for one-sided Lipschitz nonlinear systems is validated by considering a numerical simulation of mobile systems in Cartesian coordinates.

Bone fracture detection from radiographic images remains a challenging task due to overlapping anatomical structures, heterogeneous fracture appearances, and imaging inconsistencies across acquisition settings. This study introduces CARNFusion, a context-aware multimodal meta-learning framework that integrates convolutional and transformer-based feature extractors with ensemble reasoning for accurate and interpretable fracture diagnosis. The framework unifies DenseNet201, Vision Transformer (ViT), and a custom Context-Aware Residual Network (CARN) through a cross-attention fusion mechanism (CAFM) and a correlation-aware normalization layer (CARNorm), while a LightGBM-based meta-learner refines decision boundaries and enhances probabilistic calibration. Experiments were conducted on two benchmark datasets: D1 (10-class Bone Break Classification) and D2 (binary Bone Fracture Detection). The proposed model achieved accuracies of 98.76% and 99.63% on D1 and D2, respectively, surpassing both individual and concatenated deep models. Ablation analyses verified the contribution of each module, and robustness evaluations showed less than 1% degradation under Gaussian noise, brightness, and contrast variations. The framework achieved an average inference latency of 1.3 ms per image, supporting real-time diagnostic deployment. Overall, the results demonstrate that context-aware fusion and meta-learning can substantially improve diagnostic precision, stability, and interpretability. CARNFusion provides a scalable and computationally efficient approach for radiographic screening and presents strong potential for implementation in intelligent, clinician-assisted fracture diagnosis systems.