International Journal of Engineering and Manufacturing @ijem
Статьи журнала - International Journal of Engineering and Manufacturing
Все статьи: 597
Mask-Aware Localized Inpainting Method for CPU-Based Inference
Статья научная
Image generation methods, including inpainting, are evolving rapidly; however, high memory requirements continue to limit their practical deployment. As a result, the efficient utilization of Latent Diffusion Models on edge devices has become increasingly important. This work explores techniques for reducing memory usage in Latent Diffusion Models while preserving their generative capabilities. We propose a resource-efficient inpainting method optimized for CPU-based inference, based on a combination of VAE tiling, attention slicing, and dynamic region-of-interest slicing. Experimental results demonstrate that the model's memory footprint can be significantly reduced while maintaining output quality, without substantial increases in computation time, enabling execution on systems with as little as 4 GB of memory and only two processing cores. While the introduced optimizations, particularly those based on localized image processing, introduce an inherent trade-off between memory usage and computational cost, resulting in longer inference times compared to GPU-accelerated solutions, they demonstrate strong potential for deployment in memory-limited environments. Additionally, we provide analysis of key deployment bottlenecks, including model compilation for cold-start overhead mitigation, proper runtime configuration and scheduler selection. These findings confirm the feasibility of effectively deploying Latent Diffusion Models for inpainting tasks on CPU-only, resource-constrained platforms, thereby broadening their applicability to edge computing scenarios.
Бесплатно
Mathematical Model of Subpopulation Dynamics in Case of Different Niches for Subpopulations
Статья научная
The article presents a model of dynamic processes occurring in non-isolated populations that differ in their habitat and mode of nutrition. The results of theoretical studies carried out on the basis of this model show the decisive influence of the ratio of the coefficients of inter-subpopulation competition on qualitative changes in the behavior of the system and individual subpopulations. This ratio is also the main factor influencing the formation of the dominant subpopulation in the system. It has been shown that the system-wide dynamics of subpopulation processes significantly depends on the reproductive potential of all subpopulations and on the mass fraction of individuals that, according to their phenotypic properties, are related to the parents. In this case, the mass fractions of individuals (transition coefficients) must correspond to the condition of closed system and be in specified intervals. It has been established that subpopulations in real life can exchange descendants, which, in turn, can significantly affect the numerical and qualitative aspects of the dynamics. Using the example of a two-dimensional system, the relationship between the sum of the main elements of the transition coefficient matrix and the mutual dependence of subpopulations, as well as their transition to qualitatively different levels, is shown. The bifurcation properties of the model of subpopulation dynamics with a Lotka–Voltaire type function in basic quality have been studied. An approximate justification of possible bifurcations of the system allows us to evaluate the factors that qualitatively influence the dynamics of the system and develop a number of recommendations to prevent the occurrence of catastrophes and collapses in the system.
Бесплатно
Medical Image Synthesis Using Variational Autoencoder and Generative Adversarial Networks
Статья научная
Nowadays, image synthesis has become essential in the medical field for lever- aging deep learning technique to improve decision- making. Our proposed research work combines Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs) to synthesize medical im- ages, enhancing diagnostics, medical training, and image analysis. The model presented combines a Discriminator, and a Variational Autoencoder to capitalize on the strengths of both VAEs and GANs. The Decoder is tasked with generating synthetic medical images, the Discriminator evaluates their distinguishing factor, and the VAE learns a probabilistic mapping from input to latent space, ensuring a structured representation of underlying medical features. The training process involves a decoder creating realistic medical images, a discriminator distinguishing real from synthetic ones, and a VAE capturing meaningful data variations in the latent space. Verified on the dataset sourced from the Kaggle. The model refines its parameters iteratively using a training loop, resulting in enhanced quality and variety of generated medical images. The proposed VAE- GAN model demonstrates its efficacy by generating diverse and realistic medical images. The structured latent space contributes to interpretability, making the images suitable for purposes like data augmentation, anomaly detection, and machine learning model training.
Бесплатно
Memory controller and its interface using AMBA 2.0
Статья научная
This paper elaborates the AMBA bus interface bridge between memory controller and other supporting peripheral. The work claims the integration with FIFO, RAM and ROM with slave interface and the master of AHB bus. The AHB master initiates the operation and generates the necessary control signal. Memory controller is implemented with finite state machine considering with all the peripheral works in synchronous mode. Despite these shortcomings of the work performed study and development that followed has led the development of a memory controller on AMBA-AHB bus at a very advanced stage and next to prototyping. VHDL code is utilized to develop the design and it is synthesized in Xilinx Virtex 6 device (XC6VCX75T). The design claims a minor area overhead with improvement in speed 185.134 MHz.
Бесплатно
Mining Associated Factors about Emotional Disease Bases on FP-Tree Growing Algorithm
Статья научная
The objective of this paper is to mine the useful information from anger and anger-in life events questionnaire, Eysenck Personality Questionnaire (EPQ) , State Trait Anger eXpression Inquiry (STAXI) Scale, Trait Coping Style Questionnair (TCSQ), Perceived Social Support Scale (PSSS) , anger and anger-in predisposition questionnaire, anger and anger-in Physiological State Questionnaire (PSQ) and a number of test indicator data, Look for associated factors, fumble rule, guide people to do early prevention and treatment. In this paper the forming process of FP-tree of the Emotional database is analyzed, the algorithm of structuring frequent model FP-tree and mining frequent itemsets are designed, the database information scanned is recorded by using FP-Tree growing algorithm through state-trees, frequent itemsets meet minimum support required are generated through reducing the search space of project sets and scanning database only one. The mine of all factors associated with emotional disease is actualized. The experiment shows strong factors associated with emotional disease can be mined from database system by the mining algorithm bases on FP-Tree frequent itemsets. The mining results can provide scientific basis for the analysis, prevention and treatment of symptoms.
Бесплатно
Model Based Approach for Identification of Relevant Images from Ancient Paintings
Статья научная
In this paper an attempt is made to retrieve the relevant paintings based on the approach of the artist using Generalized Bivariate Laplacian Mixture Model (GBLMM). This article helps in understanding the outline of assorted artists and help as a means to categorize a scrupulous painting based on the style or the text ingrained within the images. To profile the artist style GBLMM is used. The projected model helps to discriminate the strokes of the artists and lend a hand in the classification of paintings. The proposed model is implemented using high resolution Chinese painting images.
Бесплатно
Modeling Aspects with AODML: Extended UML approach for AOD
Статья научная
Aspect Oriented Software Development (AOSD) has been considered one of the most promising abstractions to make software structure more maintainable and configurable. It also helps to overcome two big issues of current object oriented programming principles, to reduce the problem of code tangling and code scattering. Aspect Oriented Programming (AOP) has been focused largely in the implementation/coding phase. But nowadays the AOP has been matured enough to turn into AOSD, as it the main objective of separation of concerns right through the process of software development. In this paper we deal with the impact of aspect in development of software especially in designing aspect with Unified Modelling Language (UML). We propose visual models to incorporate aspect and aspectual constructs as an UML metamodel approach and new extensions to UML. The proposed language aspect oriented design modelling language (AODML) is an extension for aspect modelling into existing UML specifications. This paper allows designers to specify and realize aspects in the design and implementation phase explicitly. The proposed visual models, supports Aspect, aspectual components and its association with base components i.e. classes to be incorporated into UML. AODML motivates designer to get benefited to develop the system using AOSD paradigm. It allows to model aspects in design diagrams so that it can be implemented in any AOP language effectively.
Бесплатно
Modeling Changing Graphical Structure
Статья научная
We introduce the graphical models to describe the changing dependency structure between multivariate time series and design the algorithm by the markov chain monte carlo method. The model is applied to the stock market of Shanghai in China to study the changing correlation of five segments of the market, empirical results show that there is stronger dependency structure in the bear market and weaker correlation in the bull market.
Бесплатно
Modeling and Real-Time simulation of large hydropower plant
Статья научная
In this paper, modeling and simulation of large hydropower plant in real-time platform named Real-Time Laboratory (RT-LAB) is carried out. First, a hydropower plant model consisting of nonlinear hydro turbine with PID governor and synchronous generator (SG) with DC1A excitation system and connected to grid is developed in MATLAB/Simulink environment. This model is then simulated in RT-LAB after the modification of MATLAB/Simulink model required for suitable operation in RT-LAB environment. Finally, the real-time simulation of hydropower plant when subjected to disturbances of load addition, reduction of load and short circuit fault analysis is presented and discussed.
Бесплатно
Modeling and Simulation of an Indirect Natural Convection Solar Dryer with Thermal Storage Bed
Статья научная
The intermittent nature of solar energy limits a 24 hour operation and the effectiveness of solar thermal devices. Affordable and environmentally friendly materials for storing solar energy are currently in search. A natural convection solar cabinet dryer coupled with thermal energy storage bed (gravels) is modeled and simulated for space heating application (tomatoes drying) using TRNSYS 16 software. Performance of the solar thermal system (solar cabinet dryer) with a thermal storage bed will serve as a guide in developing a gravel-pit (GP) and or water-gravel pit storage system (WGPS) on a medium to large scale to facilitate solar thermal storage of heat for space and water heating applications in homes, health care and educational facilities. Thermal storage volume and thickness of gravel bed were determined and an optimized solar collector area obtained using TRNSYS 16 software for drying 6kg of tomatoes slices. A computer program was written to predict the product drying temperature, mass of moisture removed, moisture content and drying rate at two different trays including solar collector efficiency, heat storage bed temperature profile using meteorological data input of dryer location, gravel properties, solar collector parameters and solar cabinet dryer chamber variables. The month of August was used as the design month bearing in mind that it has the least solar radiation in Bauchi and thus, predicted the least drying performance while, the month of March with the most solar radiation predicted the optimum drying performance. The maximum predicted gravel bed temperatures were 44 and 59.3°C for the months of August and March respectively. Predicted performance of the solar cabinet dryer was compared to a similar cabinet dryer without thermal storage bed. Predicted maximum product drying temperatures of 48 and 69°C were obtained for solar cabinet dryer with thermal storage bed as against 46 and 66°C for solar cabinet dryer without thermal storage bed in the month of August and March corresponding to solar intensity value of 575.4 and 1049.2W/m2 respectively. To attain 4.5% moisture content for 3kg of tomatoes slices placed on each tray containing 94% of moisture, requires 37 (20 hours of sunshine and 7 hours of supplementary heat stored) and 53 (26 hours of sunshine and 6 hours of supplementary heat stored) hours of drying for solar cabinet dryer with thermal storage bed and, 52 (25 hours of sunshine) and 75 (34 hours of sunshine) hours under same weather condition for similar solar cabinet dryer without thermal storage bed for the month of March and August respectively. The average moisture extraction rate is 0.0759 and 0.0531kg per hour in the month of March for solar cabinet dryer with and without thermal storage bed and, 0.0540 and 0.0374kg per hour the month of August respectively. Predicted maximum solar collector efficiency for cabinet dryer with thermal storage bed is 50.12 and 43.85% for the month of March and August whereas, it was 45.83 and 37.66% for cabinet dryer without thermal storage bed respectively. The performance prediction of the solar cabinet dryer with thermal storage bed indicates clearly good potential for storing solar thermal heat collected during the day and effectively utilizing the stored heat during off-sunshine hours for heating applications. It is recommended that a gravel-pit (GP) and or water-gravel pit storage system (WGPS) should be developed and adequately studied for a range of operating parameters based on temperature distribution, thermal energy stored, available energy stored in the bed, energy consumption by blower (for active bed), and thermal efficiency of the collector to give clear guidelines for using the gravels for large scale solar thermal energy storage for space and water heating applications.
Бесплатно
Статья научная
In this paper, a novel method based on online monitoring of Instantaneous Exciting Current Space Phasor is presented in order to detect inter-turn faults on the transformer windings. This approach detects winding faults by comparison between presentation of the Instantaneous Exciting Current Space Phasor under healthy and faulty condition. In this work, the angular speed of Instantaneous Exciting Current Space Phasor has been introduced as one of the fault detection tools that has good sensitivity for detection of minor inter-turn faults. Firstly, a typical transformer is simulated based on Finite Element Analysis (FEA) to investigate the transformer behavior under different conditions. Then, the accuracy and performance of proposed diagnosis technique are studied by applying it to the simulated transformer.
Бесплатно
Modelling of Air Standard Thermodynamic Cycles Using CyclePad
Статья научная
The paper aims to explore the application of CyclePad in modelling air standard thermodynamic cycles. CyclePad is a powerful software tool designed for the simulation and analysis of various thermodynamic cycles. This paper provides an in-depth investigation into its capabilities and effectiveness in modelling air standard cycles, including the analysis of performance parameters such as efficiency, work output, and heat transfer. To explore the potential of CyclePad, Carnot, Otto, Stirling, Ericsson, Diesel, and Dual cycles were explored first thermodynamically and then modelled using the software. These cycles were tested against practical numerical problems, and it has been observed that the results obtained from the CyclePad are in agreement with the existing literature. Moreover, to understand the impact of input parameters on the performance of cycle output and efficiency sensitivity analysis was performed and reported. The results obtained are very encouraging and stem from the fact the CyclePad can be used effectively to understand and analysis any thermodynamic cycle (both open and close) having any level of complexity.
Бесплатно
Modular Approach based Backbone Construction Using STP with CDS
Статья научная
In a dense environment, wireless sensor network (WSN) requires more energy to work in an effective and efficient manner. Hence, energy conservation is the main objective. In the paper, we have proposed a methodology to construct a backbone using modular antennas in combination with spanning tree protocol (STP), graph sampling, and Connecting Dominating Set (CDS) strategy. The backbone construction is based upon the modular antenna based WSNs, where the dominating sets can avoid the intermediate connection in order to reduce the hop count and energy consumption. The dominating sets have been connected using the modular transmission range of the wireless sensor networks to construct the backbone. The dominating set selection procedure to construct the WSN backbone is based upon the degree of connections of the nodes, which enables the locally centralized behavior of the connected dominating sets. The proposed methodology has been proved effective resulting in the construction of an energy efficient backbone.
Бесплатно
Monkeypox Detection Using Support Vector Machine with a Quadratic Polynomial Kernel
Статья научная
This study looks at how well a Support Vector Machine (SVM) with a quadratic polynomial kernel works for detecting Monkeypox. The SVM method is compared to other machine learning models like Neural Networks, KNN, Logistic Regression, Random Forest, Decision Tree, and Naïve Bayes. By using features from medical images called Local Binary Patterns (LBP), the SVM model showed the best results, with 93.33% accuracy, 95.24% recall, 91.67% true negative rate, and 90.91% precision. The LBP features are used because they exhibit unique textural patterns that can distinguish Monkeypox and normal cases. The results show that the SVM with this kernel is good at telling the difference between Monkeypox and normal cases, making it a helpful tool for early detection in healthcare.
Бесплатно
Статья научная
The objective of the present paper is to optimize the machining parameters for turning of EN8steel on lathe machine using a combination of Taguchi and Grey Relational Analysis to yield minimum cutting forces and surface roughness. The process parameters such as rotational speed, feed, depth of cut and cutting fluid have been selected. In this study, the experiments were carried out as per Taguchi experimental design and L9 orthogonal array was used. Analysis of variance (ANOVA) was also used to find out the most influence of processing parameters on the responses. The regression equations were also established between the process parameters and responses. The results indicate that the depth of cut is the most significant factor affecting the cutting force and surface roughness followed by a feed, speed and cutting fluid.
Бесплатно
Multi-Sensor Approach for Monitoring Pipelines
Статья научная
Pipeline vandalization is one of Nigeria's economy killer, since Nigeria economy to a great extent relies on oil in that capacity. Therefore, a third party damage to petroleum pipelines can be cataclysmic if undetected and prevented. This act results in budgetary misfortunes, ecological contamination and incessant death and loss of properties worth millions as an aftereffect of vandalization. Consequently, it is very paramount to protect these pipelines from vandals through intelligent monitoring systems. Several efforts have been made towards providing a reliable monitoring system for oil pipeline, however, no practically implementable solution have been achieved. Therefore, a pipeline monitoring system using multi-sensors is presented herewith. The sensor array consists of a Passive Infrared (PIR), vibration and sound sensor. An uninvolved infrared (PIR) sensor was utilized to detect intruders before they get in contact with the pipeline and for affirmation of intruders, sound and vibration sensor were set up. As the PIR recognizes an on-coming human the sound and vibration sensors affirms if the human is an intruder(s). An intrusion message containing the location of the vandals is sent to the appropriate authority by the microcontroller via a connected GSM module. Results obtained proved the system as a viable solution for detecting pipeline vandals.
Бесплатно
Multiple Master Communication in AHB IP using Arbiter
Статья научная
The major disadvantage of a standard bus topology is the constraint of being able to realize only one communication at a time (the tasks may take place in parallel but the communications are only done in a sequential). As these communications are handled by the bus arbiter, a Bottleneck when the number of communications increases, but also when the bandwidth constraints of several communications become important.This arbitration plays a predominant role because it authorizes communications on the bus but it is also in charge of resolving the conflicts (several requests of communications at the same time). This arbitration implies therefore a limitation on the number of IP connected to the bus to a dozen elements. This work elaborates the AMBA bus interface with four masters interacting with single memory system, using Arbiter between memory controller and other supporting peripherals. Different module of i.e., AHB MSTER, AHB SLAVE INTERFACE AND AHB ARBITER(round robin algorithm)has been developed with VHDL. Further integration with FIFO, RAM and ROM with memory controller is done. The Four AHB master initiates the operations and generates the necessary control signals on single bus to memory controller with the help of arbiter. The proposed architecture shows the area efficient management as compared to previous researches of multiple data communication in AHB BUS system. The system model is synthesized with Xilinx XC6vx75t-2ff484 and simulated with MODELSIM.
Бесплатно
Статья научная
In this paper, we study the two-point boundary value problems for systems of nonlinear third-order dierential equations .Under some conditions, we show the existence and multiplicity of positive solutions of the above problem by applying the fixed point theorems in cones.
Бесплатно
Multiple-address Technique of Satellite Communication
Статья научная
In this paper, we have studied the operation modes of satellite communication system. They include that signal transmission mode, signal processing mode and signal exchanging mode. We have especially studied the multiplex mode, modulation mode, coding mode, multiple-address connection, and channel distribution and exchanging system. All we have studied are important to the satellite communication station’s changing and signal processing.
Бесплатно
NB-IoT based Status Measurement System for 33kV Power Distribution Networks in Smart Grids
Статья научная
In the recent decade, there has been a lot of focus on developing intelligent systems and appliances to suit the century's needs and make life easier. During the same period, the electric power industry introduced Smart-Grid, a crucial innovation to meet today's electric supply-demand and effectively use electric resources. The smart grid is an aspect of the electricity industry's evolution and reformation. An electrical power grid is a complex system consisting of generation, transmission, distribution, storage, and utilization. Coordinating these systems further increases the complexity of this interconnection of systems. The existing power distribution system available in the industry consists of monitoring equipment such as Supervisory Control and Data Acquisition(SCADA) to monitor some network parts. However, there's no automated way of monitoring power outages or load current flow in some sub-sections of the distribution line. Physical inspection is not convenient as it's more time-consuming.Moreover, these sub-sections may have up to ten distribution transformers or even could be more. In this work, A novel IoT-based power line monitoring system has been introduced to overcome those issues. Narrow Band Internet of Things(NB-IoT) is used in this system as the primary wireless technology. A current sensor measures electrical line currents, and sensor values are pushed to a remote IoT cloud. Implemented system tested in several 33kV power lines and result and performances of the system is presented.
Бесплатно
