Asian Journal of Convergence in Technology
Not a member yet
868 research outputs found
Sort by
Gear Geometry Analysis with Asymmetric Pressure Angle
Get PDFGear design is one of the most critical components in the Mechanical Power Transmission industry. Among all the geardesign parameters pressure angle is the most critical parameter, which mainly affects the load carrying capacity of the gear.Generally gears are designed with a symmetric pressure angle for drive and coast side. It means that both flank side of gearare able to have same load carrying capacity. In some applications, such as in wind turbines, the gears experience only unidirectionalloading. In such cases, the geometry of the drive side need not be symmetric to the coast side. This allows forthe design of gears with asymmetric teeth. Therefore new gear designs are needed because of the increasing performancerequirements, such as high load capacity, high endurance, long life, and high speed. These gears provide flexibility todesigners due to their non-standard design. If they are correctly designed, they can make important contributions to theimprovement of designs in aerospace industry, automobile industry, and wind turbine industry.
In this paper we present a mathematical model of helical gear pair with an asymmetric pressure angle. We haveincreased the pressure angle of gear on the drive side to increase the load capacity and performance of the gear pair interms of noise and mesh stiffness while transmitting power. Also we have analyzed the results of bending and contactstresses generated on gear pair with the asymmetric pressure angle
SPATIAL ANALYSIS OF MONSOON RAINFALL OF MAHARASHTRA STATE USING REMOTE SENSING TRMM DATA AND QGIS
Get PDFIndia is heavily dependent on the monsoon rainfall for agriculture. In India about 46% of net sown area is irrigated land and 54% is dependent on monsoon and rains from clouds. It is therefore necessary to study distribution of rainfall which helps in water management i.e. making availability of water required for the cropping season and dependability on rainfall during season. IMD generates its rainfall data for Maharashtra based on its approximately 878 rain gauging stations spread across the state. But, these rain gauge networks are insufficient to capture high spatial and temporal variability of precipitation system accurately. With the help of IMD data we can only get graphical representation of precipitation data. However, satellite rainfall estimates provide global coverage, provide information on rainfall frequency and intensity in regions that are inaccessible to other observing systems such as rain gauges and radar and also visualize rainfall distribution on any area with the help of map but it needs area-specific calibration and validation due to the indirect nature of the radiation measurements. TRMM Precipitation Radar provides the most accurate high resolution satellite based rainfall estimates to date. The present study is carried out over the Maharashtra state to understand the spatial distribution of rainfall. All methodologies have been run in the QGIS which is open source software. To achieve the objective, it is divided into two sections. First section includes TRMM Data collection and statistical calculation of precipitation data of Maharashtra state. In the second section rainfall distribution over study area through mapping is accomplished. From this study of rainfall distribution from the year 2010 to 2016 it has been interpreted that there is heavy rainfall in Konkan and Vidarbha region and poor rainfall in Marathwada region of Maharashtra
Comparative Analysis of Digital Elevation Models: A Case Study of Kayadhu Watershedle
Get PDFIn last few years, Digital Elevation Models (DEMs) have established more popular due to their diverse utility and applications in the fields like hydrology, forestry, precision farming, geomorphology etc. DEM is used for characterizing the topography and to derive the stream network, ridge line, thereby to study the landscape within the watershed area. DEMs from satellite imageries like Cartosat -1 is becoming popular with wide applications. The resolution is allowed for comparison is the DEM of ISRO (30m) (cartosat-1). These DEMs were created using different methods and technologies, and they can differ in how they represent the topography of the same area. This study shows that the differences in these DEMs and illustrates how these differences can produce various analytical outcomes when used to study local problems. The primary objective of this study is to compare the accuracy of Cartosat -1 DEM and DEM generated from Google earth. The google earth DEM is generated with the help of ‘Triangulation’ which is SAGA (System for Automated Geoscientific Analyses) tool. For the comparison of both the DEMs, Kayadhu watershed is taken as study area. The comparative analysis of DEM is carried out on the basis of the Stream network and contours of 5m, 10m and 15m interval with their respective lengths. The counts of contours of Cartosat -1 DEM for 5 m, 10 m and 15 m interval was found to be 27794, 27954 and 18184 respectively with contour lengths at that respective interval about 30503.2 km, 12803.7 km and 8421.45 km. The counts of contours of Google Earth DEM for 5 m, 10 m and 15 m interval was found to be 1485, 776 and 492 respectively with contour lengths at that respective interval about 8308.45 km, 4112 km and 2741 km. From this study the stream counts of Cartosat-1 DEM and Google Earth DEM was found to be 34449 and 52668 with stream length about 432 km and 1134 km respectively. This study has been carried out in open source environment viz. QGIS, SAGA, GRASS GIS and Google Earth. In this study, the Cartosat -1 DEM and Google earth DEM has minimum to maximum elevation from the mean sea level was found to be 336 m to 481 m and 408.7 m to 549.3m respectively. From the study, it is observed that Cartosat-1 DEM has more accuracy than DEM generated from Google Earth. Therefore, the Cartosat -1 DEM gives clear 3D topography than DEM generated from google earth
A Review ON NEW ADVANCED SAFETY SYSTEMS FOR RESIDENTIAL ELEVATORS
Get PDFSafety of residential elevators is now a days prior requirement for any residential buildings. It causes serious injuries. This review paper is survey of accidental causes and latest advance system used to reduce and avoid accidents. Some of them are implemented and some of are the futures. Organizations such as the American Society of Mechanical Engineers (ASME) have set standards for the construction and maintenance of elevators and escalators and for their safe operation. The through study related to causes of accidents, factors creates accidents and new safety systems, needed for investigation of better safety systems for residential elevator
Brain status recognition using FPGA)
Get PDFHuman brain, the most complex and confusingcontrol system controls the entire human body and its activities justby sending electric impulses to relevant body parts. While doingthat, it processes data such as sensory information,awareness, thinking etc. The six brainwaves viz.; Infrared, Delta,Theta, Alpha, Beta, & Gamma are generated by brain are in the rangeof 0.5 Hz to 30 Hz. The main objective of this proposed projectis to acquire and process the human brain waves. Filteringand amplifying of brain signals are big issues while processingbrain waves, since the brain signals have very low amplitude andare in the low frequency. We aim to identify the brain state such asawake, normal alert consciousness, physically and mentally relaxed,awake but drowsy, deep (dreamless) sleep, loss of bodily awareness,reduced consciousness, deep meditation, dreams, light sleep, REMsleep, heightened perception etc in real time. To carry out real timeprocessing of digital signals and to maximize the efficiency,processing on a powerful signal processor viz., FPGA is selected.It supports programming for a specific task without hardware restrictions besides being faster than its embedded counterparts which are not powerful enough forreal time processing. So in this paper, we propose a system forthe brain wave capturing, filtering and classification usingartificial intelligence in MATLAB IDE. The developed programwill be converted to HDL using HDL coder of MATLAB andthen it would be dumped in FPGA for testing the functionality ofproposed system
Mood Detection through Aesthetic Assessment of Videos using Deep Learning
Get PDFrole. Monitoring and predicting varioushuman’s feelings (happy, sad, anger, fear, etc.) is a challengingtask. Human interaction and carrier of feelings amongsthumans are accomplished mainly through five senses: touch,smell, taste audio, visual. Considering Visual sense, images andvideos are important gradients in day-to-day life. It canelevate/ depress the mood of a person. Digital contents ofmultimedia are image, audio, video, text, and so on. The usageof internet is tremendously increasing, so Internet bandwidthand storage space, video data has been generated, published,and spread robustly, and becoming an important of today’s bigdata. This has encouraged the development of advancedtechniques for a wide scope of video understandingapplications including online advertising, Cinematography,video retrieval, video surveillance, video data on Social sites,etc. However, it is easy to convey a story to a viewer of video,since a video is worth of thousands worth. And this storyactually creates a mood. This work is to detect the mood ofaesthetically pleasing videos that reflect on a person’s mood
An Enhanced Approach for Tourism Recommendation System using Hybrid Filtering and Association Rule Mining
Get PDFIn the tourism recommendation system, thenumber of users and items is very large. But traditionalrecommendation system uses partial information foridentifying similar characteristics of users. Collaborativefiltering and content based filtering is the primaryapproach of any recommendation system. It provides arecommendation which is easy to understand. It is basedon similarities of user opinions like rating or likes anddislikes and content based filtering is used to provideopinion for the new users profile. So the recommendationprovided by collaborative and content cannot beconsidered as quality recommendation. Recommendationafter association rule mining is having high support andconfidence level. So that it will be considered as strongrecommendation. The hybridization of both hybridfiltering and association rule mining can produce strongand quality recommendation even when sufficient dataare not available. This paper combines recommendationfor tourism application by using a hybridization oftraditional collaborative and content filtering techniquesand data mining techniques
A comparative study of radiometric corrections on multispectral and panchromatic images
Get PDFSatellite images which are obtained by asatellite which is not in direct contact with surface butcaptures and stores the images in range ofelectromagnetic spectrum. Satellite images consists ofnumerous distortions or errors which are caused bymany external factors and internal factors. Radiometricdistortions (Spectral Anomalies) are due to errors indigital number (DN) of image. Radiometric correctionshelps in improving the quality of image by correcting theDN values which helps in comparative studies and this isthe major step in digital image processing. This study iscarried out using ERDAS imagine software. This studymainly focused on haze reduction, noise reduction andperiodic noise removal on multispectral data andpanchromatic data. Histograms and accuracy assessmentwas carried out to plot the difference between processedand unprocessed image. Haze removal techniquesdemonstrated success compared with noise removal andperiodic noise removal technique which might be due toabsence of noise in the satellite imagery
Fuzzy entropy for Feature optimization In Motor Imagery based Brain Computer Interface
Get PDFIn non-invasive Motor Imagery (MI) basedBrain Computer Interface, variation due to MI has spread notonly in time domain but also in frequency domain. Evenchannels are also occupied by this spread. Thus number offeatures belonging to all these variations is responsible forclassifying the underlying task. This paper works on featureoptimization using fuzzy entropy so as to avoid under as wellover fitting of classifier. Time-Frequency correlation of thesignal is obtained using wavelet transform. Second and thirdorder statistical features are extracted from wavelet bands.SVM and KNN with kernel variations are used forclassification. Outcome of this experimenting leads to accuracyof 93.7% for optimized features using fuzzy entropy comparedto less than 90% for features without optimization
Effect of Dust on Output Power of Conventional Solar Panels in Bangladesh
Get PDFThere is a recent upsurge in the usage of renewable energy resources. Among renewable resources solar energy is the main contributor in the power generation in our country. Efficiency of existing solar panels is reduced due to many factors due to which the rated efficiency cannot be achieved. Dust on solar panel is one of the main factors that reduces it efficiency. Dust is barrier between sunlight and solar panel. In this paper we have discussed a method to improve the efficiency of a solar panel by removing the dust from its surface. It was found that efficiency of solar panel is improved by using dust cleaning method. The experiment was carried out using two solar panels having same rating. Dust from the surface of one panel was cleaned regularly and another one was kept as it is. The measured power was found to be higher at the panel that was cleaned regularly. Experimental data and results are discussed in details