JURNAL INFOTEL

Development of Smart Hydroponics System using AI-based Sensing

Septafiansyah Dwi Putra — 2024-08-08

This paper proposes a smart hydroponic system that operates automatically using a fuzzy logic algorithm, integrating IoT functionalities to support smart agriculture. The system allows for remote monitoring and control via the internet, providing real-time data on water levels, pH levels, temperature, and nutrient solution temperature. Precise dosing and temperature control are critical for optimal plant growth, and the system schedules temperature measurements to ensure stability. Unstable temperature can affect pH levels, thereby impacting nutrient absorption. The proposed system employs sensors to continuously monitor the electrical conductivity (EC) and pH levels of the nutrient solution. Fuzzy control is utilized to regulate the nutrient solution pump, automatically adjusting EC and pH levels to promote optimal plant growth. This approach reduces the time burden on producers and provides more precise control over the nutrient solution, resulting in improved growth outcomes. The main contributions of this work are as follows: the development and implementation of an AI-based system integrating a controller, IoT environment, fuzzy logic algorithm, and NFT (nutrient film technology) hydroponics; the creation of a user-friendly interface for farmers through the Smart-Hydroponic application, enabling hybrid monitoring and control of hydroponic farms; the establishment of an IoT-based cloud environment for sensor data monitoring; the implementation of a smart hydroponic system for nutrient sensing, monitoring, and control; and a comparative analysis between smart and conventional hydroponics based on morphological results.

Recommender System for Group of Users using Matrix Factorization for Tourism Domain (Case Study: Bali)

Z. K. A. Baizal — 2024-08-09

Choosing a product that suits a customer's needs requires a recommendation system to provide suggestions on a collection of items of interest to the user. Recommendations can be applied in various fields such as entertainment, shopping sites, social networking, job portals, discovery of relevant web pages, and so on. There are many circumstances where recommendations are needed for a group such as in tourism and entertainment purposes. The development of a Group Recommendation System (GRS) was carried out in response to the need to provide several recommendations to a group of users. We conducted this research to build a GRS that can provide item recommendations using the Collaborative Filtering (CF) method with Matrix Factorization Technique, as well as three approaches, namely After Factorization (AF), Before Factorization (BF), and Weighted Before Factorization (WBF). Determine the best approach for the three categories of groups formed, namely small groups (three members), medium groups (five members), and large groups (ten members). The focus of this research is the tourism destination domain in Bali. In the evaluation results of the precision calculation, the medium group obtained the highest score for the AF, BF, and WBF approaches of 0.944. Meanwhile, in calculating recall, the small group achieved the highest scores for the AF, BF, and WBF approaches of 0.294, 0.259, and 0.259, respectively. From the results of this study, it appears that small groups are suitable for using the BF approach, while the AF method is more effective for large groups, and the best approach for medium groups is the WBF.

Rupiah Banknotes Detection Comparison of The Faster R-CNN Algorithm and YOLOv5

Muhammad Zuhdi Hanif — 2024-08-23

Money is an essential part of human life. Humans are never separated from activities related to money. As time goes by, money is not only a means of transactions between humans but also between humans and machines. Machines can recognize money in various ways, including object detection. Object detection is one of the most popular branches of computer vision. There are many methods for carrying out object detection, such as Faster R-CNN and YOLO. Faster R-CNN has been widely used in various fields to perform object detection tasks. Faster R-CNN has advantages over its predecessor because it uses a Region Proposal Network (RPN) as a substitute for selective search, which requires less compilation time. YOLO (You Only Look Once) is the most frequently used object detection method. This method divides the image into grids; each part of the grid predicts objects and their probabilities. The main advantages of YOLO are its high speed and ability to recognize objects in various conditions and positions with reasonably high accuracy. This research compares the Faster R-CNN algorithm model using the ResNet-50 architecture with YOLOv5 to recognize rupiah banknotes. The dataset used is 1120 images consisting of 8 classes. The YOLOv5 model trained on RGB data had the best results, with calculation accuracy reaching 1. Test results on three images also showed suitable results. The hope is that this research can be applied in other research to build a system for recognizing rupiah banknotes.

Prediction of student delay impact on achievement at smk telkom lampung using artificial neural network

Desi Susanti — 2024-08-09

Student delays are a significant concern that can detrimentally affect the learning process and academic achievement. To address this challenge, leveraging artificial intelligence (AI) technology for analyzing educational data becomes imperative for the early identification of students potentially experiencing delays. In this regard, artificial neural networks (ANNs) emerge as highly relevant and effective methods. ANNs, inspired by the structure and function of the human brain, comprise interconnected artificial neurons capable of learning from input data to generate complex outputs, such as predictions of student delays. This study aims to forecast student delays at Telkom Lampung Vocational High School (SMK) using the ANN method, comparing its performance with other techniques like Support Vector Machine (SVM). Primary data, totalling 4939 instances with 550 cases, 26 features, and 4 meta-attributes, were collected from SMK Telkom Lampung. Performance evaluation encompassed various metrics including Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and coefficient of determination (R2). Additionally, a comparative analysis of model performance through scatter plots and box plots was conducted. The research findings suggest that the Neural Network model slightly outperforms the Support Vector Machine model, exhibiting lower prediction error rates and a superior ability to elucidate data variability.

Coverage hetnet based picocell and femtocell for uplink condition around building environment with single knife edge method

Andrita Ceriana Eska — 2024-08-09

The development of HetNet (Heterogeneous Network) radio base stations has experienced many developments. This is indicated by the existence of microcells, macrocells, picocells, femtocells, and so on. In this research, the research is aimed at the propagation of user equipment communication systems in uplink conditions with HetNet picocells and femtocells. UE propagation is on a straight path between the building environment. The communication frequency used is 10 GHz. The communication between Tx and Rx is modeled as a diffraction mechanism, AWGN channel, and atmospheric attenuation. The Single Knife Edge (SKE) method is used to model the mechanism. The propagation channel is faced with AWGN (Additive White Gaussian Noise). The analysis of this research includes the SNR value, Adaptive Modulation and Coding (AMC) level, and percentage of communication coverage area. The AMC is based on the use of MCS (Modulation and Code Scheme). Some of the MCS modulations used QPSK, 16 QAM, and 64 QAM. As research results show that the percentage of communication coverage area obtained was gNB1 80.59 percent, gNB2 65.67%, and selection combining HetNet 95.52%.

Equal Incremental Cost Method with Adjustable Gamma Control to Solve Generator Scheduling

Basuki Rahmat — 2024-08-09

Generator scheduling remains an intriguing issue within the energy industry. It relates to the optimization of production costs, where system operators must select the optimal combination of available resources to minimize production costs. This paper proposes an enhancement to the Equal Incremental Cost (EIC) Method using Adjustable Gamma Control (AGC) in generator scheduling. Iterations begin with an initial lambda value, then gradually increase with the application of the factor until power demand is met. A variable of 10% is used as an adjustment step in the optimization method. The proposed method is capable of achieving convergence with 100% accuracy, where the power generated by all generators precisely matches the load demand (2,650 MW), at a cost of USD 32,289.03. EIC-AGC ranks second-best after VLIM, albeit with the consequence of consuming 195 seconds. This method is expected to have a significant impact on designing highly accurate economic dispatch techniques. Thus, generator scheduling will lead to a reduction in operational costs compared to current practices.