Robotic Systems and Applications
Not a member yet
20223 research outputs found
Sort by
A CNN-BiLSTM algorithm for Weibo emotion classification with attention mechanism
Get PDFWeibo short text information contains a large amount of network language, emoticons, etc., and due to the long-time span of the content, the emotions of the posts posted by people often change due to time or the occurrence of certain special events. Therefore, traditional sentiment analysis methods are not suitable for this task. This article proposes a CNNs-Bi LSTM sentiment analysis method that integrates attention mechanism. It combines convolutional neural networks and bidirectional short-term memory networks to obtain keyword information in text through attention mechanism, efficiently and accurately realizing data temporal and semantic information mining. Through experimental verification using Weibo public opinion data, the results show that this method achieves higher accuracy compared to other benchmark models and can fully utilize multidimensional matrices to capture rich text features, with certain advantages
A neural network simulator for educational purposes
Get PDFArtificial neural networks are inspired by biologic processes. Artificial neural networks are important because they can be used to deduct a function from observations, in other words artificial neural networks can learn from experience. Artificial neural network simulator to fulfill a need into the growing interest of neural network education is introduced in this study. NeuroQuick Laboratory simulator is implemented using object-oriented programming by Delphi programming and these classes can be used to create a standalone application with artificial neural networks. The NeuroQuick Laboratory Simulator is designed for a broad range of users, including beginning graduate/advanced undergraduate students, engineers, and scientists. It is particularly well-suited for use in individual student projects or as a simulation tool in one- or two-semester neural network-related courses at universities
Optimization and experimental validation of the air intake holes of the lithium-ion battery pack
Get PDFEnergy storage systems enable the storage of energy and provide access to carbon-neutral, environmentally friendly energy whenever or wherever it is needed. Lithium-ion batteries are currently the most preferred type among various battery technologies and are widely used in energy storage systems. Some of the features that make lithium-ion batteries advantageous include high energy density, long life, low maintenance requirements, and high operating voltage. The growing demand for energy throughout the day increases the need for batteries with high storage capacity. However, the increased capacity also leads to heating issues in lithium-ion batteries. The heating problem in lithium-ion batteries can result in nonhomogeneous temperature distribution, shortened lifespan, thermal runaway, increased internal resistance, and performance loss. Therefore, an effective thermal management system is essential for cooling lithium-ion batteries. This study aims to provide insight into the forced air cooling of prismatic 280 Ah LiFePo4 batteries, which have limited information in the literature and are more prone to overheating compared to lower-capacity batteries. In this study, five different battery pack case designs, each with different sizes and numbers of air intake holes, were determined and modelled using the SolidWorks program. Within the battery pack cases, 16 280 Ah lithium-ion batteries are placed, and an axial fan is used to cool these batteries. Initially, computational fluid dynamics analyses of the five different designs were performed in the SolidWorks Flow Simulation program. An experiment was then conducted on the design that provided the most efficient thermal management to validate the numerical results. The selected design, fulfilling the purpose of homogeneous temperature distribution and having the minimum temperature difference between batteries, was designated as Design 5. It exhibited a 62 % improvement in cooling performance with a 0.25 °C temperature difference, indicating successful temperature homogeneity between batteries. During a two-hour experiment with a 140 A discharge current, temperature measurements were taken from the surfaces of the batteries using thermocouples. Finally, the maximum error rate between experimental and numerical studies was determined to be 1.47 %, indicating successful validation of the numerical study. The air intake hole optimization, a novel design approach, prevents temperature distribution inhomogeneity caused by the distance of the batteries to the fan and offers an effective way to cool down high-capacity 280 Ah batteries
Failure analysis of gas turbine in first stage turbine blades in an urban power plant
Get PDFThe Combined Cycle Power Plants (CCPP) are widely known as the most efficient power plants. The concept of CCPP represented in Brayton and Rankin cycles combined to achieve the highest efficiency in power generation. During the fact finding, while performing the major outage, at 77,521 running hours some of the turbine blades at 1st stage found suffering from severe damages and sulfuric acid thick layer accumulated in the cooling airfoils. The blades were new, also known as the most essential and costly component installed after Hot Gas Path Inspection (HGPI) at 42,000 running hours. The main objective of this work is to study the surrounding parameters that directly affect the lifetime of the turbine blades. Furthermore, XRD analysis for new and used turbine blades are studied as well as the EDS analysis for the coating layers which are used to enhance the surface life of gas turbine blades. Corrective maintenance, on the other hand, has been undertaken to overcome the performance deterioration caused by wear, which is the result of any machine or plant’s operation. A gas turbine cannot be run reliably unless wear reserves are checked and, if necessary, restored during inspections and maintenance operations. These are performed to determine the machine's condition and the necessary condition-based repairs. The principal types of gas turbine inspection have been investigated and presented, including minor inspection, hot gas inspection, extended-scope hot gas path examination, and major inspection. Lifetime extension (LTE) measures are derived from these inspection types and go beyond the scope of a major inspection. Nondestructive testing is performed on the gas turbine’s key components. Furthermore, due to the high degree of contamination in the fuel, the turbine section must be cleaned at regular intervals. As a result of the examinations, replacement, and refurbishment intervals for important components, as well as outage time, could be recommended
Numerical modelling of the warping behaviour at the first layer-build plate interface in 3D-printed models produced via the fused deposition modelling process
Get PDFThe material structure of 3D-models printed via the fused deposition modelling (FDM) technique is mainly affected in the z-direction of the 3D-print as a result of the layer-by-layer approach which tend to exhibit a deformation behavior corresponding to a type of transversely orthotropic material. Moreover, uncontrolled parameters such as printing temperature and printing speed have been reported to adversely affect 3D-print quality leading to undesired effects such as distortion and warpage. The additive manufacturing process is a relatively new field in advanced manufacturing where further research and innovation are required to overcome the limited strength and structural performance observed in presently 3D-printed components. In line with the above, this study proposes the numerical investigation of the warping behavior in PLA (Polylactic acid) - based 3D printed models by considering the finite element method (FEM) software of LS-DYNA. The warping investigation was specifically centered on the cooling cycle prevailing between the layer-by-layer structures. The findings of this study showed that warpage would most likely occur in the thermal process model corresponding to abrupt change in temperature due to a buildup of strain between the bottom most layers of the 3D model and the build plate. The findings of this study, which shed light on the warping behaviour in 3D-models, has direct implications on the final quality of 3D-printed components
Research on stress through photoelastic experiment and finite element method considering sliding wear
Get PDFSliding contact on the contact interface of friction pairs is a common type of contact. The sliding wear caused by sliding contact has an obvious influence on the stress in the contact area. In this study, the photoelastic experiment and finite element method are adopted to study variation laws of stress in the contact area. The results show that the stress in the contact region is very concentrated, and the contact half-width gradually ascends with the increase of sliding wear. The stress intensity in the contact region and von Mises stress at the contact centre decrease with the increase of wear depth. In the case of a wear depth of less than 0.3 mm, the stress intensity and the contact stress decrease rapidly with the growth of wear depth. When the wear depth exceeds 0.3 mm, the influence of wear depth on the stress intensity and contact stress is small. The results of this research clarify the effect of sliding wear on the stress in the contact area, and provide a reference for studying the contact issues
Design of treatment scheme and identification method of treatment effect for soft soil subgrade
Get PDFAccording to the characteristics of the soft soil subgrade in the Haiou Island section of Nansha Bridge, a comprehensive geological survey and soil analysis were conducted. Based on the analysis results, a corresponding design scheme for the soft soil subgrade was formulated to ensure bridge safety and stability. Furthermore, various reinforcement measures such as compaction, drainage, and lightweight soil were implemented to mitigate the impact of soft soil filling on the bridge structure. To evaluate the effectiveness of our treatment method, post-construction effect testing was performed using both static pressure and vibration acceleration methods. The static pressure method provided crucial parameters related to bearing capacity and compressibility of the soft soil subgrade while vibration acceleration analysis reflected response characteristics under dynamic load conditions. By comparing and analyzing results obtained from these two methods comprehensively evaluated feasibility of our treatment approach. Finally, based on experimental study findings and data analysis outcomes, it is recommended that post-construction effect testing for Haiou Island section's soft soil subgrade should employ both static pressure method and vibration acceleration method along with targeted reinforcement measures selected specifically to minimize adverse impacts on bridge structure
Application of modal analysis to multi-objective optimization of gear box
Get PDFThe high stiffness and low mass gearbox structure design requirements were met through multi-objective optimization based on finite element modal analysis. The correlation between structural parameters and optimization objectives was identified, providing guidance for optimal design. A parametric finite element model of the box cover was established to calculate the first-order modal shape and equivalent stiffness. By using mass and equivalent stiffness as optimization objectives, a discrete data set was constructed with key dimensions of the box cover as design variables, from which a surrogate model was obtained. Standard response surface and full second-order polynomial fitted response surface function were applied after error verification. Different optimization algorithms were employed to search for the optimal value, resulting in obtaining design variables meeting the optimization objectives. The results indicated a reduction in optimized mass by approximately 6.8 %, with an error of less than 1 %
Analysis and optimization of vibrating screen structure based on modal simulation
Get PDFIn order to enhance the stability and reliability of the vibrating screen, modal and harmonic response characteristics were simulated and analyzed using the finite element method. The final assembly model was established in Creo, serving as an intermediate format for material definition and mesh accuracy verification. The analysis of natural frequency and mode revealed that the overall structural stiffness was basically satisfactory. However, harmonic response analysis indicated issues related to excessive stress and stress concentration within the structure. Subsequent optimization of the frame structure confirmed nearly unchanged low-order natural frequencies, while maximum stress was reduced by 22.15 %, holding significant implications for process cost control
Development of a rotation and swing torque detection system after bearing installation
Get PDFThe swing torque and rotational torque after the spherical bearing is installed directly affect the performance of the spherical bearing. At this stage, the friction torque detection equipment of the spherical bearing is mainly used to detect uninstalled bearings. A set of rotation and swing after the bearing is installed is designed. Torque detection system. The detection principles of rotational torque and swing torque required for flexibility detection were analyzed, the functional design requirements and main technical indicators of the detection system were clarified, and the overall design plan of the detection system was established; the host structure of the detection system was designed, including rotational torque detection system, swing torque detection system, clamping system and calibration system; completed the scheme design of the detection control system, selected the torque sensor and servo motor, designed the main electrical control circuit of the detector; conducted error analysis of the detector