Proceeding of the Electrical Engineering Computer Science and Informatics
Not a member yet
649 research outputs found
Sort by
Analysis of Waveform of Partial Discharge in Air Insulation Measured by RC Detector
Get PDFThis study discusses the measurement of Partial Discharge (PD) in air insulation. Partial Discharge Measurement is very important to know the condition of electrical equipment. The cause of partial discharge is not only old equipment, but also from set-up errors and insulation problems. In this research partial discharge measurement was performed by using electrical methods. Electrical method use RC Detector. The modeling of partial discharge was done by using needle-plane electrode distant 1 cm in air insulation. Partial discharge measurement parameters include the measurement of Background Noise (BGN), Partial Discharge Inception Voltage (PDIV) and PD Waveform. The Partial Discharge measurement result show that Vpp of BGN ON is higher than Vpp of BGN OFF. The negative PDIV signal first appeared for the RC Detector at a voltage 3.55 KV and Positive PDIV at 4.01 KV. Negative and positive PD waveform for RC Detector at 5 KV, 5.5 KV, 6 KV, 6.5 KV and 7 KV respectively, it has been found that the fall time is greater than the rise time, and peak to peak voltage (Vpp) will be greater when the applied voltage is greater
CountNet: End to End Deep Learning for Crowd Counting
Get PDFWe approach crowd counting problem as a complex end to end deep learning process that needs both a correct recognition and counting. This paper redefines the crowd counting process to be a counting process, rather than just a recognition process as previously defined. Xception Network is used in the CountNet and layered again with fully connected layers. The Xception Network pre-trained parameter is used as transfer learning to be trained again with the fully connected layers. CountNet then achieved a better crowd counting performance by training it with augmented dataset that robust to scale and slice variations
Robust Principal Component Analysis for Feature Extraction of Fire Detection System
No full textFire detection system with deep learning-based computer vision (DLCV *) algorithm is proposed in this paper. It uses visible light sensor charged-coupled device (CCD) which can be usually found in closed circuit television camera (CCTV). The performance of this DLCV fire detection depends on how many fire image datasets are trained that might lead to the curse of dimensionality. To tackle the curse of dimensionality, Principal Component Analysis (PCA) will be used. PCA is a technique for feature extraction in which the dimensionality of such datasets is reduced significantly. This will results in increasing interpretability but at the same time minimizing information loss
A Survey on Topologies and Controls of Z-Source Matrix Converter
Get PDFThis paper describes the Z-source matrix converter (ZS-MC) topology which specifically discusses topology and control on the ZS-MC. There are two topologies on the ZS-MC, namely Z-source direct-MC (ZS-DMC) and indirect-MC (ZS-IMC). The difference of each of these topologies is in the number of switching mosfets, where ZS-DMC put on nine switches, while ZS-IMC eighteen switches. ZS-IMC topology overcomes the limitations of traditional MC voltage reinforcement and accommodates the operation of buck and boost converter by reducing the number of switches and providing high efficiency
Classification of Motor Imagery and Synchronization of Post-Stroke Patient EEG Signal
Get PDFStroke attacks often cause disability, so the need for rehabilitation to restore patient's motor skills. Electroencephalogram (EEG) is an instrument that can capture electrical activity in the brain. Some post-stroke patients have brain electrical dysfunction so that EEG signal can achieve such as amplitude decrease, and wave differences from symmetric channels. However, EEG signal analysis is not easy because it has high complexity and small amplitude. However, information from EEG signals is beneficial, including for stroke identification. This study proposes the identification of EEG signals from post-stroke patients using wavelet extraction and Backpropagation Levernberg-Marquardt. EEG signals are recorded, extracted imagery motor variables, and synchronization of symmetric channels. The results of the study provide that the accuracy for identifying post-stroke EEG signals is 100% for training data and 79.69 % for new data. Research also shows that the use of learning rates affects accuracy. The smaller the learning rate provided accuracy is better. However, it had consequences for computing time so that the optimal learning rate is 0.0001
Classification of Physiological Signals for Emotion Recognition using IoT
Get PDFEmotion recognition gains huge popularity now a days. Physiological signals provides an appropriate way to detect human emotion with the help of IoT. In this paper, a novel system is proposed which is capable of determining the emotional status using physiological parameters, including design specification and software implementation of the system. This system may have a vivid use in medicine (especially for emotionally challenged people), smart home etc. Various Physiological parameters to be measured includes, heart rate (HR), galvanic skin response (GSR), skin temperature etc. To construct the proposed system the measured physiological parameters were feed to the neural networks which further classify the data in various emotional states, mainly in anger, happy, sad, joy. This work recognized the correlation between human emotions and change in physiological parameters with respect to their emotion
Performance Evaluation of Superstate HMM with Median Filter For Appliance Energy Disaggregation
Get PDFInformation on electricity consumption is one of the essential elements in terms of regulating the distribution of electricity in smart micro grid. Besides, information on electricity consumption can help consumers carry out an evaluation process to reduce electricity bill costs, which indirectly affect overall energy efficiency. One method in the process of monitoring electricity consumption is Non-Intrusive Load Monitoring (NILM). The main problem in NILM is to determine the energy disaggregation consumed by several equipment by merely performing the retrieval of data from only one measuring point. We used the Superstate Hidden Markov Model as the tool for modelling and analysis. A median data filter to the input data is applied to improve the performance of the disaggregation process. Based on the results of tests conducted using the REDD, the lowest accuracy was 96.69% for all tests performed
Object Distance Measurement System Using Monocular Camera on Vehicle
Get PDFTo support autonomous vehicles that are currently often studied by various parties, the authors propose to make a system of predicting the distance of objects using monocular cameras on vehicles. Distance prediction uses four methods and the input parameter was obtained from images processed with MobileNets SSD. Calculations using linear regression are the simplest calculations among the four methods but have an error of 1% with a standard deviation of 1.65 meters. While using the first method, the average error value is 9% with a standard deviation of 0.43 meters. By using the second calculation, the average error resulted in 6% with a standard deviation of 0.35 meters. The experimental method had an average error of 1% with a standard deviation of 0.26 meters, so the experimental method was used
An Android-based Hoax Detection for Social Media
Get PDFHoax is defined as a try to convince any readers to believe particular deception. A news frequently spreads on social media. Hoaxes news is engineered to meet a personal purpose and caused by predefined factors. In Indonesia, one of the top discussions is about hoax news related to natural disasters spread through social media. Some people have been sentenced to jail because of making and spreading hoax news. To date, the internet community is unable to distinguish hoax news from the actual news. This is because there is no available tool to detect hoaxes. Therefore, this paper propose an initiative to have a hoax detection application in Android-based devices by using the web scraping technique to extract data downloaded from the inputted URL with a simplified interface. The retrieved data are optimized by utilizing Nazief-Adriani stemming algorithm and analyzed by using Rabin-Karp algorith
Controlled Position Navigation of Single Degree Magnetic Levitation
Get PDFA permanent magnet is levitated following the electromagnetic suspension principle using the attractive magnetic force of a wire-wound electromagnet with a hall-effect sensor for position feedback. Taking the hall-effect voltage as an analog parameter and feedback signal to the micro-controller, the strength of the electromagnet is controlled by adjusting the current using the Pulse Width Modulation technique in order to levitate the permanent magnet. The stability of the levitated magnet is enhanced by the use of PID algorithm in the embedded system. Use of Laplace transform for simplification of differential equations and Taylor series for the linearization of system function supports the mathematical computation required for the levitation. Furthermore, by making the feedback signal from hall-effect sensor dependent only on the magnetic field of levitating magnet, an advancement in levitation phenomenon is achieved that aids the levitation with a greater flexibility of changing the position of the levitating magnet along the gravitational axis within a specified range.So the paper depicts about the "Controlled Position Navigation of Single Degree Magnetic Levitation"