Claim Missing Document
Check
Articles

Found 6 Documents
Search

Optimizing Smart Power Grid Stability Based on the Prediction of a Deep Learning Model Hamad Khaleefah, Shihab; A. Mostafa, Salama; Gunasekaran, Saraswathy Shamini; Farooq Khattak, Umar; Ahmed Jubair, Mohammed; Afyenni, Rita
JOIV : International Journal on Informatics Visualization Vol 8, No 3 (2024)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.8.2.2758

Abstract

A smart grid is an electricity transmission system that uses digital technology to control getting and dispatching electricity from all generation sources to satisfy end users' fluctuating electricity demands. It achieves this through deploying technologies such as technology and smart grids, which are pivotal in increasing the power supply's efficiency, reliability, and sustainability to the public. Decentralized Smart Grid Control (DSGC) is a system where the control and decision-making functions are distributed to different grid points instead of in one central place. This paradigm is critical for the fault resistance and efficiency of the grid because it enables the local regions to carry on by themselves, manage electric power flows, respond to changes, and integrate many kinds of energy sources successfully. The grid frequency is monitored via the DSGC to ensure dynamic grid stability estimation. All parties, from users to energy producers, may take advantage of the price of power tied to grid frequency. The DSGC, a vital component of this research, gathered information about clients' consumption and used several assumptions to predict the behavior of the consumers. It establishes a method to assess against current supply circumstances and the resultant recommended pricing information. This research proposes a long short-term memory (LSTM) model to analyze data gathered regarding smart grid characteristics and predict grid stability. The results show a strong capacity for the LSTM model, achieving an accuracy of 96.73% with a loss of just 7.44%. The model also achieves a precision of 96.70%, recall of 98.18%, and F1-score of 97.43%.
Augmented Reality Applications for Cultural Heritage Using Kinect Sensor Afyenni, Rita; Erianda, Aldo; Wahyuni, Putri; Firosha, Ardian; Gusman, Taufik; Sumema
International Journal of Advanced Science Computing and Engineering Vol. 6 No. 3 (2024)
Publisher : SOTVI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/ijasce.6.3.221

Abstract

There are quite a number of advantages gained from Microsoft Kinect which is a motion sensing device that also includes AR applications to aid in the preservation and promotion of cultural assets. In this work, a batik model was developed with the help of Microsoft Kinect by utilizing its core components consisting of a depth sensor, an RGB camera and a microphone array to allow motion tracking, gesture recognition, and voice command functionalities of the device.The depth sensing features of the device, in particular the structural light employed, transition to Time-of-Flight technology has improved the efficiency and applicability of the device towards AR. Kinect has the ability to create an immersive AR experience by accurately mapping virtual objects to the user’s environment, leveraging body joint tracking to create a 3D skeletal model. It is required to take into consideration the kinnect positioning technique which has importance in allowing 3D objects and models to align. Therefore, the conclusion addresses Kinect as it relates to human-computer interaction, in particular, concerning its potential to change body interaction in real time and how it sets the stage for future systems that require an active interaction across a number of fields including culture and heritage.
Analysis of Realistic Game Graphics Development Using Photogrammetry Method Isyrin Apridal, Fauzi; Hidayat, Rahmat; Afyenni, Rita
Jurnal EECCIS (Electrics, Electronics, Communications, Controls, Informatics, Systems) Vol. 18 No. 2 (2024)
Publisher : Faculty of Engineering, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21776/jeeccis.v18i2.1706

Abstract

The development of realistic video game graphics has increasingly become a primary focus in the gaming industry. Among the technologies enabling this advancement, photogrammetry stands out for its ability to create highly accurate and realistic 3D models from real-world objects or environments. This study utilized RealityCapture software to process a dataset of flower pot images into a detailed 3D model. The photogrammetry process encompasses several critical stages: importing the image dataset, calibrating the camera to correct for lens distortions, reconstructing the 3D model from aligned images, refining the model to correct any inaccuracies, and finally integrating the model into a game engine for use. The results illustrate that photogrammetry can significantly enhance the realism of video game graphics, offering a level of detail and authenticity that traditional modeling techniques may struggle to achieve. Despite some technical challenges, such as high hardware requirements and large data sizes, photogrammetry reduces the time and cost of creating graphic assets while delivering an immersive experience for players.
Perbandingan Akurasi EfficientNetV2 dan MobileNetV2 pada Klasifikasi Makanan Tradisional Indonesia Rayhan, Ahmad; Hidayat, Rahmat; Afyenni, Rita
Journal Cerita: Creative Education of Research in Information Technology and Artificial Informatics Vol 10 No 2 (2024): Journal CERITA : Creative Education of Research in Information Technology and Ar
Publisher : UNIVERSITAS RAHARJA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33050/cerita.v10i2.3326

Abstract

Machine learning is a rapidly evolving technology that has led to numerous advancements, especially in the field of computer vision. Within this domain, many models have been introduced, including EfficientNetV2 and MobileNetV2. Given the wide range of model choices, it is essential to compare the accuracy of each model. This involves data collection, resizing and sorting data, visualizing data, and finally tuning and training processes. In this study, it was found that the accuracy of EfficientNetV2 is 31%, while MobileNetV2 achieves an accuracy of 99%.
Academic Advisory Guidance Application Website-Based Sukma, Fanni; Kasmar, Andre Febrian; Afyenni, Rita; -, Defni; Azizah, Nur
International Journal of Advanced Science Computing and Engineering Vol. 4 No. 3 (2022)
Publisher : SOTVI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (433.556 KB) | DOI: 10.62527/ijasce.4.3.111

Abstract

Academic Advisory Guidance at the Information Technology Department  Politeknik Negeri Padang is currently carried out manually, causing several obstacles, including time and place. To help with problems that occur, a website-based academic guidance system was built using the waterfall method. The stages of implementation methodology carried out are Analysis, System Design, Program Code Generation, and Program Testing. With this system, students and academic advisors can interact anytime and anywhere without having to meet face-to-face. This can also be a solution for students who have both academic and personal problems.
Design and Development of a Web-Based Mosque Management Information System: A Case Study of Darul Amal Mosque Azmi, Meri; Sonatha, Yance; Kirana, Salsabila Okta; Idmayanti, Rika; Afyenni, Rita
invotek Vol 24 No 3 (2024): INVOTEK: Jurnal Inovasi Vokasional dan Teknologi
Publisher : Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/invotek.v24i3.1205

Abstract

Many mosques in Indonesia face challenges in managing their operational activities, such as construction donations, donations to orphans, funeral donations, and religious assemblies as well as other financial administration. Conventionally, the recording of these activities is still done manually through financial recording in books and socializing the information using loudspeakers or utilizing mosque bulletin boards. To overcome these limitations, this research proposes the development of a Web-based Mosque Management Information System that aims to improve operational efficiency, transparency, and accessibility. The system was developed by taking a case study at Darul Amal Mosque which is a mosque in West Sumatra, Indonesia. The system was developed using PHP programming language and Laravel framework, with an online donation feature that allows donors to make donations without the need to be physically present. The research successfully developed a mosque management platform as expected. And it has been proven that a centralized and easily accessible online management process facilitates financial management, improves services to worshipers and donors, and provides better accountability in mosque operations.