KTU Open Journal Systems (Kaunas University of technology)
Not a member yet
13395 research outputs found
Sort by
Ground Fissure Identification in Mining Areas from UAV Images Based on DN-CAMSCBNet
No full textThe development and use of mine resources have had many adverse impacts on the environment of mining areas. Among them, ground fissures are the most serious. They not only threaten the ecological protection of mining areas but also hinder the sustainable exploitation of energy. To mitigate the damage to the ecological environment caused by mining areas and ensure sustainable long-term resource exploitation, it is of particular importance to identify ground fissures in mining areas efficiently. Therefore, this paper proposes a ground fissure identification model for UAV images in mining areas named DN-CAMSCBNet. This method integrates the channel attention mechanism and the dropout mechanism on the basis of the traditional U-Net. Meanwhile, it introduces the multiscale convolution block and Nesterov-accelerated adaptive moment estimation. These are used to enhance its ability to capture complex image features, expand the receptive field of the original model, reduce the number of parameters, and reduce computational complexity. To verify the segmentation performance of the model, it is compared with U-Net, D-CAMNet, and D-MSCBNet models. The experimental results show that the accuracy and precision of the DN-CAMSCBNet model can reach 99.47 % and 92.25 %, respectively, and the F1 score is 0.7699. All these are superior to comparison models and can provide strong support for the identification of ground fissures in mining areas
Enhancing Intrusion Detection and Mitigation in Ad Hoc Networks Using an AI-Driven Deep Learning Approach
No full textAd hoc networks are increasingly deployed in critical applications due to their flexibility and scalability. However, their decentralised and dynamic nature makes them highly vulnerable to a range of sophisticated security threats. This paper aims to improve the efficiency of intrusion detection and mitigation in ad hoc networks using an AI-driven deep learning approach. A hybrid deep learning model is proposed, integrating convolutional neural networks (CNNs) for feature extraction and long short-term memory networks (LSTMs) for temporal analysis to effectively detect malicious activities. Reinforcement learning, particularly using a deep Q-network (DQN), is applied to dynamically select optimal mitigation strategies. Federated learning is also used to train the model in a distributed manner, ensuring privacy while allowing scalability across network nodes. The proposed approach shows significant improvements in intrusion detection accuracy, exceeding 90 %, and offers effective real-time mitigation strategies. These results provide a comprehensive and adaptive framework for securing ad hoc networks against evolving threats
Enhancing Fracture Assessment of Pipe Girth Welds with Root Cracks
No full textDespite the widespread use of the failure assessment diagram, accurately evaluating girth weld joints with cracks remains a significant challenge within the pipeline industry, due to the mismatched strength between the parent metal and weld metal. To Address this issue, an innovative approach was proposed by converting the heterogeneous girth welds as an equivalent homogeneous structure and deriving an equivalent stress-strain relation, thereby enabling the construction of a failure assessment curve for girth welds. This study aims at enhancing the precision of defect assessment in girth welded joints. Firstly, a finite element model is developed to characterize the crack driving force in high-grade pipeline girth welds. Furthermore, an optimized calculation method is proposed for determining the reference stress and limit load of pipeline girth welds with root cracks by introducing a correction factor. Moreover, a predictive formula for the correction factor has been established based on finite element simulation results. Through rigorous numerical simulations and methodical calculation techniques, this study aims to provide guidance for the accurate evaluation of girth weld root cracks in pipeline engineering applications
The Improvement of Detection Precision of Internal Surface Detection System of Main Cylinder
No full textIn view of the current status of the inspection of the main cylinder inner surface of the key components in the automobile brake system, an advanced inspection scheme integrating light, machine, electricity and calculation is put forward, the basic structure and the concrete implementation scheme of the detection system are given, and the geometric error elements are analyzed according to the working principle of the detection system, based on the theory of multi-body system and the principle of homogeneous coordinate transformation, the synthetic error model of the detection system is constructed, and the geometric error of the detection system is detected on-line by XL-80 dual-frequency laser interferometer, the existing geometric error data are obtained, the error compensation scheme of the detection system is put forward, and the error compensation verification of the detection system is carried out, it is proved that the proposed method can effectively improve the detection accuracy of the internal surface of the master cylinder
Current Transport Mechanisms and Electrophysical Characteristics of the 4H-SiC p-n Junctions Formed by Aluminum Diffusion
No full textIn this paper, the electrophysical characteristics of the 4H-SiC p-n junction created by low-temperature diffusion of aluminum were studied. Current-voltage (I-V) characteristics are analysed, and the current transport mechanisms in 4H-SiC p-n junctions are discussed. It is shown that at low forward bias voltages, the generation–recombination mechanism dominates, and the I-V characteristics at voltages U > 3.0 V obey the linear law. At reverse biases, the dominant mechanism of current transfer is limited by the space charge
Microstructure and Visible Light Photocatalytic Studies of CeO2 Doped ZnO Nanoparticles
No full textCerium oxide (CeO2) doped with zinc oxide (ZnO) nanoparticles were synthesised using sol-gel method. The impact of CeO2 doping on the microstructure, optical, photocatalytic, and antibacterial properties of ZnO nanoparticles (NPs) were investigated. XRD results showed the presence of both CeO2 and ZnO phases. The ZnO sample crystallizes into a hexagonal wurtzite structure and CeO2 with a cubic structure. The crystallite sizes were estimated and found to be 18 nm and 29 nm for the CeO2 and ZnO, nanoparticles respectively. FTIR spectrum reveals the formation of chemical bonds. The FESEM studies showed the formation of dense crystallites with aggregation. The EDX studies confirmed the presence of cerium, zinc, and oxygen in the sample, and no additional peaks have been detected. The UV-Visible spectroscopy analysis revealed the bandgap of 3.15 eV. The photodegradation activity of the nanoparticles with methylene blue (MB) dye was carried out using visible light irradiation and the degradation efficiency was found to be 75.81 % in 1 h duration. The antibacterial susceptibility test was conducted using the agar well diffusion method to investigate the activity of CeO2 doped ZnO nanoparticles against E. coli, Salmonella typhimurium, Bacillus cereus, and Shigella flexneri, and the results showed significant antibacterial activity as compared to the control drug penicillin
Microfluidic-Engineered Preparation of Poly-L-lactic Acid Porous Microspheres as Cell Scaffolds
No full textRegenerative medicine has emerged as a promising field to address tissue damage and organ failure, with porous microspheres playing a crucial role as cell carriers in tissue engineering applications. However, conventional fabrication methods often result in heterogeneous size distributions and poorly controlled pore structures, limiting their effectiveness. This study leverages microfluidic technology to overcome these challenges and develop uniform poly (L-lactic acid) (PLLA) porous microspheres. We comprehensively evaluate their physical and biological properties, including morphology, size distribution, pore structure, and biocompatibility. Through long-term culture experiments, we investigated the microspheres\u27 capacity to support cell growth and proliferation, demonstrating their effectiveness as cell scaffolds. Our research provides valuable insights into the potential of microfluidic-produced PLLA porous microspheres and its further translational medicine
Performance Analysis of Tunnel Water Plugging Engineering Slurry Based on Polycarboxylate Superplasticizer
No full textIn response to the low efficiency of water plugging slurry in high temperature tunnel engineering, this research explores the optimization configuration of tunnel water plugging slurry. Through the analysis of the gel time, fluidity, stability and mechanical properties of the slurry fused with polycarboxylate superplasticizer, the response surface analysis method and central composite design are used to optimize the design of the slurry for water shutoff projects. Experimental verification showed that the performance of grouting was optimal when the ratio of 12.48 % fast hardening sulfoaluminate cement 425, 0.91 % polycarboxylate superplasticizer, 3.03 ‰ calcium lignosulfonate, and 1.00 ‰ hydroxypropyl methylcellulose was formulated. The gel time of the slurry prepared at 90 ℃ was 12 min, the viscosity was 2064 mPa · s, the water separation rate was 2.87 %, and the compressive strength of the slurry prepared 3 days later was high as 21.35 mPa, meeting the requirements for tunnel water plugging construction. The results showed that fast hardening sulfoaluminate cement 425 had the greatest impact on the performance of the slurry, while polycarboxylate superplasticizer had obvious effects on the stability and flowability of the slurry. The optimal configuration scheme proposed in the study has positive application significance in high surface temperature tunnel water plugging engineering. This optimal solution provides theoretical basis and technical support for tunnel water plugging engineering, and improves the effectiveness of grouting configuration
Synthesis and Performance of Polycarboxylate Superplasticizer Based on a Novel Redox System of Manganese Dioxide/Acid
No full textThis study aims to develop a novel redox system based on manganese dioxide (MnO₂)/acid for synthesizing polycarboxylate superplasticizers (MPCE) via solution copolymerization of acrylic acid (AA) and methallyl polyoxyethylene ethers (TPEG). The performance of MPCE synthesized using MnO₂ combined with different acids (citric acid, tartaric acid, oxalic acid) was compared with two conventional redox systems: ascorbic acid (VC)/H₂O₂ and ammonium persulfate. The MnO₂/citric acid system proved most effective, attributed to its one-step reaction to generate primary free radicals. Optimal reaction conditions were determined as a molar ratio of AA to TPEG (Mn=2400) of 3.5:1.0 and [MnO₂]:[citric acid] of 1:1. The longer side chains of MPCE enhanced adsorption on cement particles and dispersing performance via steric hindrance. Cement paste containing MPCE-2400 showed superior dispersing ability, retention capacity, clay resistance, and hardened mechanical properties. The findings highlight the critical role of the MnO₂/acid redox system in optimizing MPCE synthesis
Recovery Effects of Alginate Film and Thermal Nursing in Thoracoscopic Surgery Patients
No full textThe healing process of postoperative wounds is a dynamic and complex. Thoracoscopic surgery, as a new minimally invasive technique in thoracic surgery, is of great significance in exploring the postoperative recovery quality of patients undergoing this surgery. Therefore, this study uses micro controlled flow spinning technology to prepare a calcium alginate fiber coating (CaAFC) film using materials such as sodium alginate (SA). Combined with thermal insulation nursing during the recovery period, the effect of this film was explored in patients undergoing thoracoscopic general anesthesia surgery (TGAS), and it was applied in animal experiments simulating TGAS. The results showed that the permeability of the CaAFC was 93 g/m2·24 h, the porosity was 73 %, and the water absorption rate was 21.5 g/g, demonstrating good breathability and water absorption. The relative growth rate (RGR) of HaCaT cells under CaAFC was 1.52, and the stress fiber staining intensity (SFSI) was 1.31, both of which were higher than the other two dressings. These findings indicated that the material was non-toxic to cells and could significantly improve cell activity. The wound healing rate of mice treated with CaAFC on the 15th day after surgery was 96.3 %, which was significantly higher than that of mice treated with traditional alginate medical dressings (TAMD) and medical gauze (MG). These results indicated that CaAFC could promote wound healing. The above results indicate that the CaAFC has good breathability, water absorption, and antibacterial properties. Moreover, it can significantly promote wound healing, increase cell activity, reduce postoperative infection risk, and provide support for improving postoperative care for patients undergoing thoracoscopy surgery in clinical practice