Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Development of IoT based intelligent irrigation system using particle swarm optimization and XGBoost techniques
Get PDFA crop needs regular watering throughout its life to grow well. Irrigation improves food growth. Machines irrigate plants. The dry Sahel, which gets a lot of rain during the summer season but is dry in winter, needs irrigation. When it doesn't rain enough, crops need watering. By constantly monitoring soil moisture, humidity, temperature, and pH, precision agriculture reduces water use and increases crop output. Precision gardening uses less water. In many wealthy nations, efficient farming requires the internet of things (IoT). Particle swarm optimization (PSO) and XGBoost are used in this IoT-based intelligent watering system. Humidity and moisture sensors gather soil data at grass roots. Sensors constantly gather this data. These data are useless for smart watering. PSOselects smart watering data. This reduces central cloud info storage. Then, machine learning methods are trained using soil humidity, moisture, crop, and weather data. These programs can calculate a crop's water requirements. IoT devices control irrigation system water flow and results in saving fresh water. XGBoost algorithm is saving water from 23% to 27% for different crops
Firefly algorithm tuning of PID position control of DC motor using parameter estimator toolbox
Get PDFThis paper aims to design an accurate angular position control for DC motors using a proportional integral derivative (PID) controller. The estimated DC motor parameters have been calculated using the parameter estimator toolbox in MATLAB, Arduino Mega 2560 and speed sensor to build an accurate model in MATLAB. The optimized PID coefficients are found for the DC motor model using the firefly algorithm (FA), which aims to make the actual angle match the desired value without overshooting and oscillations. The PIC16F877A microcontroller was used to implement the code based on optimized PID coefficients found in MATLAB/Simulink to generate the suitable pulse width modulation (PWM) output. In this work, step input was tested to analyze the characteristics of the system response in terms of rise time, settling time and overshoot. It was found that the controller output response curve which is produced from FA-based-PID reached the desired position without overshoot and any oscillations. The findings established that closed-loop control of any system using a system identification toolbox and optimized PID technique can be applied in real applications using low-cost controllers and sensors such as PIC16F877A microcontroller and analog rotary position sensor, respectively
Implementing trajectory correction strategy through model prediction control for flight vehicle missions
Get PDFModeling a high-speed flying vehicle is imperative to ensure the success of vehicle development missions. Moreover, adherence to research protocols mandates a stepwise approach to testing the vehicle model, encompassing simulation trials using software-in-the-loop simulation (SILS), hardware-in-the-loop simulation (HILS), as well as diverse ground and environmental tests prior to flight testing. This study entailed a collaborative effort between MATLAB/Simulink and LabVIEW to seamlessly integrate the model developed in MATLAB/Simulink into LabVIEW for the implementation of model predictive control (MPC) strategy, aimed at trajectory correction (TC) missions for the vehicle. This MPC strategy was directly applied to the onboard flight control system (OBFCS) of the vehicle. Simulation results indicate the successful control of roll and pitch conditions by OBFCS in both SILS and HILS, ensuring the maintenance of flight conditions in accordance with predicted trajectories despite the presence of simulated disturbances. Notably, the simulation demonstrates the independence or absence of interference between each simultaneous MPC control for roll and pitch adjustments
Towards visual sentiment summary to understand customers’ satisfaction
Get PDFDue to the COVID-19 pandemic, the shopping behavior of customers has been significantly affected and is being shifted towards online shopping. Understanding the customers’ opinions, attitudes, and emotions in feedback and comments plays an essential role in making decisions for organizations and individuals (e.g., companies and customers). In this study, we propose sentiment summaries from the customer knowledgebase (SSoCK) framework that analyses customer feedback and improve a mechanism for sentiment summarization by using text analysis including sentiment analysis. In the experiments, various domains from customer reviews (e.g., computer and Canon) are used to conduct. The results show that the proposed SSoCK framework has the high performance of sentiment classification in terms of its accuracy when compared to the other approaches. Moreover, the proposed framework generates various kinds of sentiment summaries that can support managers/potential customers understand trending/interesting aspects of the product with customer satisfaction and can be easily updated with new reviews within the same domain without storing any original data
A hybrid facial features extraction-based classification framework for typhlotic people
Get PDFFacial features play a vital role in the real-time cloud-based applications. Since, most of the conventional models are difficult to detect heterogeneous facial features due to high computational memory and time for the internet of things (IoT) based video surveillance mechanisms. Video based facial features identification and extraction include a large number of candidates features which are difficult to detect the contextual similarity of the facial key points due to noise and computational memory. In order to resolve these issues, a hybrid multiple features extraction measures are implemented on the real-time video dataset to extract key points using the cloud-based classifier. In this work, a hybrid classifier is used to classify the key facial points in the cloud computing environment. Experimental results show that the proposed hybrid multiple feature extraction-based frameworks have better computational efficiency in terms of error rate, recall, precision, and accuracy than the conventional models
Differential evolution with adaptive mutation and crossover strategies for nonlinear regression problems
Get PDFThis paper presents the differential evolution algorithm with adaptive mutation and crossover strategies (DEAMC) for solving nonlinear regression problems. The DEAMC algorithm adaptively uses two mutation strategies and two ranges of crossover rate. We evaluate its performance on the National Institute of Standards and Technology (NIST) nonlinear-regression benchmark containing many models of varying levels of difficulty and compare it with classic differential evolution (DE), enhanced differential evolution algorithm with an adaptation of switching crossover strategy (DEASC), and controlled random search methods (CRS4HC, CRS4HCe). We also apply the proposed method to solve parameter identification applications and compare it with enhanced chaotic grasshopper optimization algorithms (ECGOA), self-adaptive differential evolution with dynamic mutation and pheromone strategy (SDE-FMP), and JAYA and its variant methods. The experimental results show that DEAMC is more reliable and gives more accurate results than the compared methods
Improving sentiment analysis using text network features within different machine learning algorithms
Get PDFSentiment analysis poses a significant challenge due to the inherent subjectivity of natural language and the prevalence of unstandardized dialects in social networks. Regrettably, existing literature lacks a dedicated focus on network representation learning for sentiment classification. This paper addresses this gap by investigating ten machine learning algorithms, including support vector machine (SVM), random forest (RF), logistic regression (LR), and Naive Bayes (NB). Our approach integrates text network analysis and sentiment analysis to propose a comprehensive solution. We begin by applying text preprocessing techniques and converting a text corpus into a text network using word co-occurrence. Subsequently, we employ network analysis techniques to extract features based on network topology and node attributes. These network-derived features serve as inputs for sentiment prediction on Yelp reviews. Through the incorporation of diverse text network features and various machine learning algorithms, we achieve significant enhancements in sentiment classification performance. Our evaluation demonstrates an improved area under curve (AUC) of 83% on the Yelp reviews corpus, underscoring the efficacy of integrating network features to enhance sentiment classifiers. This research underscores the critical role of network representation and its potential impact on sentiment analysis, highlighting the prospect of harnessing network features for sentiment classification tasks
A Thai-language chatbot analyzing mosquito-borne diseases using Jaccard similarity
Get PDFThe objective of this study is to create a Thai-language chatbot analyzing mosquito-borne diseases using Jaccard similarity with an aim to develop an artificial intelligence (AI)-based chatbot used to analyze Aedes-borne diseases through natural language processing. The analysis occurred when the symptoms provided by users on the chatbot were assessed to select relevant words as text attributes using the term frequency-inverse document frequency (TF-IDF) before the Jaccard similarity was used to measure the similarity of the information on the mosquito-borne disease database. The Line Messaging API was applied to facilitate communication between users and the chatbot through the Line application. The chatbot applied PHP 7.2.34 and MySQL 5.7.32 for database management, with Apache 2.2.29 serving as the bot server. The performance evaluation of the chatbot revealed that the chatbot accurately understood user intentions with an intent accuracy of 85.00%. Likewise, the usability of the chatbot was assessed using the system usability scale (SUS), and it received a score of 89.75, indicating a high level of user-friendliness. Furthermore, it has been found that appropriate tokenization enables accurate feature selection. This leads to improved accuracy in measuring Jaccard similarity. Consequently, the chatbot is capable of providing precise responses that align with the user's intent
Intelligent agriculture system using low energy and based on the use of the internet of things
Get PDFThe field of smart agriculture is ranked among the top areas that uses the internet of things (IoT), whose goal is to increase the quantity and quality of agricultural productivity. The aim of this work is to realize a new device that will be cost-effective, reliable, and autonomous using a solar panel to provide electricity in large-scale agricultural fields, ESP32 to interconnect IoT sensors and the long range (LoRa) data transmission protocol to guarantee connectivity in places where there is no internet, whose objective is to monitor and irrigate agricultural fields only when there is a need for water. The data received by the sensors is sent to mobile app users via the Blynk cloud. The performance of our new approach is measured in terms of energy savings. This new model of irrigation and smart monitoring will improve the efficiency of farming techniques
Distribution network reconfiguration utilizing the particle swarm optimization algorithm and exhaustive search methods
Get PDFThe load level for each period in the distribution network can be considered non-identical due to the increasing demand for loads and the bigger distribution network. The main problem in the transmission and distribution network system is power losses and voltage profiles, affecting the quality of service and operating costs. This study compares the reconfiguration of the network using exhaustive search techniques and particle swarm optimization (PSO) algorithms on the IEEE 33 bus distribution network system. The study’s results compare the study of power flow before and after network reconfiguration, which is a decrease in the value of power losses from 202.7 kW to 139.6 kW. Then voltage profile improved from 91.309% to 93.782%. The simulation results also found that this reconfiguration can improve the system voltage profile, which initially contained 21 buses outside the standard limits of IEEE Std 1159-1995 to 7 buses