6 research outputs found
Performance Analysis of Power-Aware Node-Disjoint Multipath Source Routing in Mobile Ad Hoc Networks
No full textMaximizing Electric Vehicle Battery Efficiency: A Multi-Model Machine Learning Approach for RUL Prediction of NMC-LCO Batteries
No full textElectric vehicles (EV) are becoming more prevalent because they are good for the environment and don't cost much to run. One big problem with EVs, though, is that their batteries don't last long. There is a complete way to figure out how long Nickel Manganese Cobalt-Lithium Cobalt Oxide (NMC-LCO) batteries will still work after this study. The information used in this study comes via the Hawaii Natural Energy Institute consist of 15 different batteries that were put through over 1000 rounds of controlled settings. A method with several steps is used, starting with collecting data and preparing it, then choosing features and getting rid of outliers. The RUL forecast method is made with machine learning (ML) methods like Bagging Regressor, XG Boost, Cat Boost, Light GBM and Extra Trees Regressor. Feature value analysis helps find important factors that affect the health and lives of a battery. Statistical tests show that there are no missing as well as duplicate data points and getting rid of outliers makes the method more accurate. Not surprisingly, XG Boost turned out to be the best algorithm, making predictions that were very close to being correct. This study shows how important RUL forecast is for improving battery lifetime management, especially in electric car uses, to make sure that resources are used in the best way possible, costs are kept low, and the environment is protected
AI-Powered Optimization of Solar Absorbers: Enhancing Industrial Thermal Energy Harvesting Through Deep Learning
No full textThermal energy harvesting is a recent attention due to the possibility of harnessing the sun to generate sustainable energy. The solar collector is essential components of this process because it turns the sun's rays into heat. A solar deep learning model (SDLM) is used to improve the efficiency of solar absorber in current industrial settings for collecting thermal energy. Several devices in this model gather information over time about things like moisture, speed of the wind, temperature, pressure of air as well as sun energy. This information is utilized for ML program that can predict the energy of a certain panel. For the proposed SDLM, the thresholds were 75.05 percent for absorption prevalence, 69.89 percent for absorption discovery, 81.41 percent for absorption omission, 90.82 percent for crucial success index, and 73.20 percent for threshold. To estimate the amount of thermal energy that may be gathered more precisely, the system includes other parameters like motion as well as insulation. In order to turn sunlight into heat, solar filters are employed in manufacturing. This thermal energy is crucial for many electrical systems, including heating and cooling systems, and industrial activities. Before investing in solar absorbers, companies may use the SDLM to calculate their prospective thermal energy production
Optimizing Battery Charge Prediction Accuracy Utilizing Machine Learning Methods
No full textEnergy storage systems are more cost-effective when they correctly manage the capacity for lithium-ion batteries (LiBs), especially when they are used on a big scale. The design saves money, in the long run, to repair or fix LiBs less often. To determine the amount that LiBs were capable of holding, adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), gradient boosting, light gradient boosting machine (LightGBM), category boosting (CatBoost), as well as ensemble learning models are utilized. Employing the mean absolute error (MAE), and the mean squared error (MSE) along R2 numbers, the researcher compared the accuracy with which each model could predict future outcomes. For example, the LightGBM model had the least MAE (0.102) as well as MSE (0.018) values, as well as the greatest R-squared (0.886) value, which means that its predictions were most closely related to reality. It was about the same in terms of speed among the gradient boosting as well as XGBoost models, which came next to LightGBM. The ensemble model's efficiency suggests that integrating many models might result in an overall increase in performance. In addition, the research uses Shapley additive explanations (SHAP) values to analyze important aspects influencing model predictions within the context of explainable artificial intelligence (XAI). This study found that discharge capacity is strongly influenced by temperature, cycle index, voltage, and power. This study demonstrates that Machine Learning (ML) methods can improve energy storage systems and regulate LiB in XAI
DETECTION OF HEART DISEASE BY USING RELIABLE BOOLEAN MACHINE LEARNING ALGORITHM
No full textArtificial Intelligence (A.I) is one of most exciting fields of computer engineering today. It is the science and technique used to make machine intelligent and it is vast and truly universal field. However, tremendous growth has been observed in this filed in past two decade owing to valuable contributions from variety of domains. It has numerous potential applications such as computer vision, medicine, philosophy, psychology, linguistics, automatic programming, natural language processing, speech processing and robotics, etc. Machine Learning takes training from natural events and helps in predicting any type of event and is a branch of Artificial Intelligence (AI). Over the past two decades, Machine Learning became a major source for information technology in developing applications, such as manufacturing industry for automation in assembly line, biometric recognition, handwriting recognition, medical diagnosis, speech recognition, text retrieval, natural language processing and Machine Learning is widely using in Data Science (DS), it is predominant and hotcake field of 21st century. Today all of use machine learning several times a day, without knowing it. Examples of such "ubiquitous" or "invisible" usage include search engines, customer-adaptive web services, email managers (spam filters), computer network security, and so on. Since last few decades Cardiovascular(Heart) Diseases (CVDs) has emerged as the most life-threatening diseases and proved to be fatal not only in India but throughout the whole world. In time detection, diagnosis and treatment of the disease needs a reliable, accurate and feasible system. In this paper we proposed Reliable Boolean Machine Learning Algorithm (RBMLA) by using novel approach to predict heat disease. Finally performance of RBMLA is measured by using various performance metrics like accuracy, precision, recall, sensitivity, specificity, reliability, F-score and ROC curve. It is shown that it gives better performance for given any new test data and new real time data. It has given better accuracy of 86%
Prediction of Heart Disease by Developing the Hybrid Deep Learning Models to Attain Higher Accuracy
No full textOne of the most common long-term diseases in the world is cardiovascular disease, also called heart disease. It is hard to make accurate and quick predictions about heart problems. Most of the work that has been done so far to identify heart disease has used machine learning methods, but they have not been able to get more accurate results. Recent advances in deep learning methods have a big effect on data analysis. Combining convolutional neural networks via a memory (LSTM), which network is what this work is all about. It aims to be additional exact than added ML methods. The heart disease information was put into two groups: normal and abnormal. This was done using a mixed CNN and LSTM method. The k-fold cross-validating method remained used to prove that this combination system works 90% of the time. Different machine learning algorithms, like SVM, Naïve Bayes, and Decision Tree, are compared with the suggested method to see how well it works. The outcomes show that future method works improved than the ML models that are already in use
