JURNAL INFOTEL

Prediction model with artificial neural network for tidal flood events in the coastal area of bandar lampung City

Eka Suci Puspita Wulandari, Ramadhan Nurpambudi, RZ. Abdul Aziz — Mon, 08 May 2023 00:00:00 +0000

The fastest sea level rise began in 2013 and reached its highest level in 2021. This is part of the ongoing global warming impact, where polar ice continues to melt, glaciers also continue to melt, causing sea level rise. In the Bandar Lampung City area, there are several areas that are threatened with tidal flooding, namely Karang City Village and Kangkung Village, Bumi Waras Village, and Sukaraja Village. Bandar Lampung itself is the city center in the coastal area. Where the majority of the population is in the Coastal area so that the threat of tidal flooding is caused by rising sea levels. To study the occurrence of tidal floods in the past, this research uses an Artificial Neural Network which has the ability to study non-linear data which is then carried out by training and testing until the best configuration model is obtained. Based on the analysis and discussion that has been carried out, several important points can be drawn, including the results of training and dataset testing that has been carried out. , 80:20, and 90;10. This is evidenced by the results of the high accuracy of the model configuration and also the results of the prediction table which is able to describe the actual conditions, setting the model configuration experimentally is able to produce the best training accuracy value reaching 100% while for the best testing accuracy is 88%. The average correlation value of training with the 50:50 dataset is 0.975, the 60:40 dataset is 0.975, the 70:30 dataset is 0.951, the 80:20 dataset is 0.935, and the 90:10 dataset is 0.929. For the average value of the correlation test with the 50:50 dataset of 0.514, the 60:40 dataset is 0.362, the 70:30 dataset is 0.488, the 80:20 dataset is 0.284, and the 90:10 dataset is 0.402. Whereas the average error value for the 50:50 dataset is 0.006, the 60:40 dataset is 0.006, the 70:30 dataset is 0.010, the 80:20 dataset is 0.007, and the 90:10 dataset is 0.007, the flood prediction table is made based on 1 configuration the best with a training accuracy rate of 98% and a testing accuracy of 80% with an error value of 0.004, namely configuration model 14, this model is the best configuration model out of 3 dataset divisions out of a total of 5. The prediction table uses sea level tides of 1.5 meters. The prediction table is able to provide good tidal flood percentage values, especially when there are active astronomical phenomena. The results of this good flood prediction table illustrate that the backpropagation ANN is able to study datasets well and can be used by BMKG forecasters in making tidal flood early warnings.

Improved vanishing point reference detection to early detect and track distant oncoming vehicles for adaptive traffic light signaling

Yoanda Alim Syahbana, Dodi Zulherman, Yasunari Yokota — Tue, 09 May 2023 08:02:27 +0000

Real-time traffic monitoring is essential for the operation of an adaptive traffic lighting system and plays a significant role in decision-making, particularly signaling in roadworks. When only one lane is accessible due to temporary road blockage, early detection of oncoming vehicles is crucial to minimize bottlenecks near the traffic light that could result in congestion and accidents. This research aimed to enhance the detection and tracking of traffic at a distance from the traffic light. We utilized the vanishing point as a reference for detection and calculated the region of interest. We implemented the proposed method on twelve traffic surveillance videos and evaluated the system performance based on how quickly it could detect incoming traffic compared with the R-CNN method. The proposed method detected target vehicles in an average of 17.75 frames, while the R-CNN method required an average of 63.36 frames. Moreover, the proposed method’s precision depends on the number of pixel orientations used to estimate the vanishing point and the definition of the region of interest. Therefore, the proposed method for enhancing the safety and reliability of an adaptive traffic light system is reliable.

Internet of things for monitoring parking system using optical character recognition

Dona Yuliawati, Rio Kurniawan, Bayu Nugroho, Suhendro Yusuf Irianto, Sri Karnila — Sun, 14 May 2023 10:37:53 +0000

This research is in the form of an IoT-based parking system, which can help the transportation department. Currently, there are several obstacles experienced in collecting parking levies in the field, the absence of automatic and real-time information on four-wheeled and two-wheeled vehicles and the processing of vehicle parking tax levies is not transparent. One of the components of local revenue is the motor vehicle tax, in Bandar Lampung City, the implementation is still not optimal. This type of On Street Parking parking service uses the curb to park motor vehicles, generally guarded by a parking attendant with a parking location that has been determined by the parking manager. At each On Street Parking parking point, parking attendants are facilitated with a tool in the form of "Monitor Parking", with detection cameras that take pictures of motor vehicle license plates and store them in a database. OCR (Optical Character Recognition) technique of annotated plate data, and generates data again. The design results are in the form of a vehicle parking monitoring tool that can be run through portable gadgets. The "Monitor Parking" tool is easy to use and can help make it easier for parking attendants and the Transportation Agency to monitor parking in the field.

Multi industry stock forecasting using deep transfer learning

Ezra Julang Prasetyo, Kristoko Dwi Hartomo — Thu, 18 May 2023 12:38:13 +0000

After the Covid-19 pandemic, the number of investors in Indonesia has proliferated. In managing a good stock portfolio, investors need the right strategy too. One approach that can be applied is to predict stock movements by considering the company's industrial sector. This paper proposed a new framework for applying deep transfer learning for stock forecasting in multi-industry. The model used in the framework is a combined algorithm between Gated Recurrent Unit (GRU) and Long-Short Term Memory (LSTM). The author built the pre-trained model using Indeks Harga Saham Gabungan (IHSG) and transferred it to predict Indonesia's stock indexes based on industry classification (IDX-IC) as the measurer of stock movement in multiple industries. The outcomes reveal that this framework produces good model predictions and can be used to help analyze the evaluation of the pre-trained model to conduct transfer learning stock prediction in different industries efficiently. The model built using the IHSG indexes can predict stock prices best in the energy, technology, and industrial sectors.

Indonesian news classification application with named entity recognition approach

Nurchim Nurchim, Nurmalitasari Nurmalitasari, Zalizah Awang Long — Tue, 23 May 2023 06:32:53 +0000

Nowadays, many netizens search for news via search engines with countless amounts of information, so it is increasingly difficult to determine when the number of news articles that appear changes very quickly and dynamically. Thus, it is necessary to process the extraction of news information to display the core information of the news. Problems arise, especially in Indonesian, which has a structure of various noun phrase entities with shallow parsing or grammatical induction. Named Entity Recognition (NER) has the opportunity to overcome this because it can extract news entities in depth, starting from proper nouns in text documents containing information search, machine translation, answering questions, and automatic summarization. This study aims to apply NER in Indonesian language news classification. This study uses Design-Based Research whose process includes (1) pre-implementation, (2) design, (3) implementation and revision, and finally, (4) reflection and evaluation. This application was developed on the platform python, streamlit, BeautifulSoup, gnews, and spacy library. The results of application accuracy testing have an F1-score value of 89.69% for all entities consisting of place, figure, day, date, and organization.

Static and dynamic human activity recognition with VGG-16 pre-trained CNN model

Mawaddah Harahap, Valentino Damar, Sallyana Yek, Michael Michael, M. Ridha Putra — Wed, 24 May 2023 02:53:35 +0000

Human Activity Recognition has been widely studied using the Convolutional Neural Network (CNN) algorithm to classify a person's movements by utilizing data from devices that record movements such as cameras. The benefits generated by this technology are useful for modern devices such as Virtual Reality and Smart Home technology with CCTV cameras. The VGG-16 (Visual Geometric Group with 16 Layers) pre-trained model is one of the models used for transfer learning and has won the Image Net competition. In this study, the authors tested the performance of the VGG-16 model to identify two types of human activity, namely Static and Dynamic. This study uses 1,680 public datasets which are divided into 80% Data Train, 10% Data Validation, and 10% Data Test I. In addition, there are also 100 local datasets as Data Test II. There is no overfitting issue in the training and testing process. The accuracy of the Testing process with public and local images dataset produces a high accuracy of 98.8% and 97% respectively.

Fatigue detection using decision tree method based on PPG signal

Ilham Ari Elbaith Zaeni, Arya Kusuma Wardhana, Erianto Fanani — Mon, 08 May 2023 00:00:00 +0000

Fatigue is a complex psychophysiological condition marked by sleepiness or fatigue, poor performance, and a range of physiological changes. A decision tree may be used to categorize weariness based on the subject's heart rate data. To begin the experiment, a dataset of the heart rate signal was obtained. The signal has already undergone preprocessing. The feature obtained through preprocessing is then used to construct the decision model. Four traits were discovered. The HF power, LF power, normalized HF power, and normalized LF power are the characteristics. This research has a 75.94% accuracy rating. The precision, recall, and F-measure scores for this study were 0.736, 0.736, and 0.736, respectively.

Monitoring of three-phase distribution power transformer based on internet of things (IoT) and SCADA

Yusnan Badruzzaman, Revi Alvin Razaqi — Mon, 08 May 2023 00:00:00 +0000

The three-phase distribution transfomer, equipment for stepping down the voltage from medium (20/11,5 kV) to low voltage network (400/231 V) with a constant power, is a type of the PT. PLN (Persero) assets which has a direct relationship with customers. The condition and the performance of transformer are affecting on how the continuity of the electricity distributed. Hence, the monitoring process of three-phase distribution transfomer condition and performance should be done. Some elements which have to be monitored such as voltage (ZMPT101B sensor), current (ACS712 30 A sensor), power, and transfomer load. Those elements could be included as an electrical indicator.And then the transfomer’s temperature (DS18B20 sensor) and the oil transfomer level (HC SR04 sensor) could be included as a mechanical indicator. All of the sensors are processed and programmed with Arduino Mega 2560 which has been connected directly into an additional modul called Ethernet shield and router. The results then emitted by WiFi into SCADA to be shown. The results shown by SCADA is the information whether transformer need to be maintened or not

Prototype of cascade level and flow control system on steam drum based on IoT

Mon, 08 May 2023 09:03:00 +0000

In the industrial field, boiler functions to heat a fluid in the form of water, the boiler has a part which is a steam drum which has a function to produce steam for use for utility needs, and a steam turbine, in practice, the state of the water level must be maintained at the desired value or set. point so that carryover does not occur, and in overcoming these problems a control system is needed. This control works by comparing the value of the sensor and the set point, then gives an output signal to correct that to speed up the response, so it is necessary to use a cascade control configuration that adds an input flow control as a slave control. In this prototype, the cascade level control serves to control the level process. In addition, the human-machine interface has been designed to monitor processes in real-time. In addition, this prototype is equipped with an Internet of Things system that functions for the monitoring process as long as it is always connected to the internet. To run the control system, parameter control is needed, in this project the PID parameter setting uses the Ziegler-Nichols method with the parameter Kp level=20.25; Ki level = 1.51; Kp Flow = 5.14; Ki flow = 2.2.

A Fire suppression monitoring system for smart building

I Ketut Agung Enriko — Tue, 30 May 2023 23:53:38 +0000

A fire suppression system (FSS) monitoring system is a system to monitor the FSS devices’ status since FSS is a critical system to respond to fire disasters. The monitoring system collects data on important parameters which are water pressure, main power status, and backup power status. The FSS monitoring system is built with an IoT capability where data are collected from the FSS module and sent to the IoT platform through Wi-Fi based Internet connection. Then the data will be displayed in a dashboard application. A QoS assessment framework is referred to and performed to check the performance of the FSS monitoring system, namely the TIPHON framework, which consists of five parameters: bandwidth, throughput, packet loss, delay, and jitter. The overall score for the FSS system using the TIPHON standard is 3.2 or categorized as “good”.