Yazar "Aksoz, Ahmet" seçeneğine göre listele
Listeleniyor 1 - 20 / 23
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe 6-switched 3-level inverter for PV power quality enhancement in smart grid application(Institute of Electrical and Electronics Engineers Inc., 2017) Saygin, Ali; Kerem, Alper; Aksoz, AhmetThis paper presents the integration of the photovoltaic panels which are modeled in the Matlab/Simulink to the smart grid using a 6-switched 3-level inverter. The system generates electricity by photovoltaic panels and feed load via 6-switched 3-level inverter. To reduce harmonics in the system and increase the quality of the generated signal, 6-switched 3-level inverter, which is contain an alternative topology to multi-level inverter topologies, is used. The required signals are generated for the operation of the inverting IGBTs using the space vector pulse width modulation technique. The system has fed the loads by generating electricity energy when the photovoltaic panels are favorable in terms of the energy production. By analyzing the performance of the modelled system the stability level is investigated. At the end of the process, a productive model has been developed in terms of power stability and quality. © 2017 IEEE.Öğe 6-Switched 3-Level Inverter for PV Power Quality Enhancement in Smart Grid Application(IEEE, 2017) Saygin, Ali; Kerem, Alper; Aksoz, AhmetThis paper presents the integration of the photovoltaic panels which are modeled in the Matlab/Simulink to the smart grid using a 6-switched 3-level inverter. The system generates electricity by photovoltaic panels and feed load via 6-switched 3-level inverter. To reduce harmonics in the system and increase the quality of the generated signal, 6-switched 3-level inverter, which is contain an alternative topology to multi-level inverter topologies, is used. The required signals are generated for the operation of the inverting IGBTs using the space vector pulse width modulation technique. The system has fed the loads by generating electricity energy when the photovoltaic panels are favorable in terms of the energy production. By analyzing the performance of the modelled system the stability level is investigated. At the end of the process, a productive model has been developed in terms of power stability and quality.Öğe A Comparative Study of AI Methods on Renewable Energy Prediction for Smart Grids: Case of Turkey(Mdpi, 2024) Unsal, Derya Betul; Aksoz, Ahmet; Oyucu, Saadin; Guerrero, Josep M.; Guler, MerveFossil fuels still have emerged as the predominant energy source for power generation on a global scale. In recent years, Turkey has experienced a notable decrease in the production of coal and natural gas energy, juxtaposed with a significant rise in the production of renewable energy sources. The study employed neural networks, ANNs (artificial neural networks), and LSTM (long short-term memory), as well as CNN (convolutional neural network) and hybrid CNN-LSTM designs, to assess Turkey's energy potential. Real-time outcomes were produced by integrating these models with meteorological data. The objective was to design strategies for enhancing performance by comparing various models of outcomes. The data collected for Turkey as a whole are based on average values. Machine learning approaches were employed to mitigate the error rate seen in the acquired outcomes. Comparisons were conducted across light gradient boosting machine (LightGBM), gradient boosting regressor (GBR), and random forest regressor (RF) techniques, which represent machine learning models, alongside deep learning models. Based on the findings of the comparative analyses, it was determined that the machine learning model, LightGBM, exhibited the most favorable performance in enhancing the accuracy of predictions. Conversely, the hybrid model, CNN-LSTM, had the greatest rate of inaccuracy. This study will serve as a guide for renewable energy researchers, especially in developing countries such as Turkey that have not switched to a smart grid system.Öğe A Novel Voltage-Current Characteristic Model for Understanding of Electric Arc Furnace Behavior Using Experimental Data and Grey Wolf Optimization Algorithm(MDPI, 2025) Seker, Mustafa; Unsal, Emre; Aksoz, Ahmet; Dursun, MahirThe control of nonlinear systems cannot be effectively achieved using linear mathematical methods. This paper introduces a novel mathematical model to characterize the voltage-current (V-I) characteristics of the electric arc furnace (EAF) melting process, incorporating experimental field data for validation. The proposed model integrates polynomial curve fitting, the modified Heidler function, and double S-curves, with the grey wolf optimization (GWO) algorithm applied for parameter optimization, enhancing accuracy in predicting arc dynamics. The performance of the model is compared against the exponential, hyperbolic, exponential-hyperbolic, and nonlinear resistance models, as well as real-time measurement data, to assess its effectiveness. The results show that the proposed model significantly reduces voltage and current harmonic distortion compared to existing models. Specifically, the total harmonic distortion (THD) for voltage is reduced to 2.34%, closely matching the real-time measured value of 2.30%. Similarly, in the current spectrum, the proposed model achieves a significant reduction in third harmonic distortion and a THD of 11.40%, compared to 13.76% in real-time measurements. Consequently, a more precise characterization of the EAF behavior enables more effective mitigation of harmonics and vibrations, enhancing the stability and power quality of the electrical networks to which they are connected.Öğe Advancing Electric Vehicle Infrastructure: A Review and Exploration of Battery-Assisted DC Fast Charging Stations(Mdpi, 2024) Aksoz, Ahmet; Asal, Burcak; Bicer, Emre; Oyucu, Saadin; Gencturk, Merve; Golestan, SaeedConcerns over fossil fuel depletion, fluctuating fuel prices, and CO2 emissions have accelerated the development of electric vehicle (EV) technologies. This article reviews advancements in EV fast charging technology and explores the development of battery-assisted DC fast charging stations to address the limitations of traditional chargers. Our proposed approach integrates battery storage, allowing chargers to operate independently of the electric grid by storing electrical energy during off-peak hours and releasing it during peak times. This reduces dependence on grid power and enhances grid stability. Moreover, the transformer-less, modular design of the proposed solution offers greater flexibility, scalability, and reduced installation costs. Additionally, the use of smart energy management systems, incorporating artificial intelligence and machine learning techniques to dynamically adjust charging rates, will be discussed to optimize efficiency and cost-effectiveness.Öğe Analysis of SARIMA Models for Forecasting Electricity Demand(IEEE, 2024) Aksoz, Ahmet; Oyucu, Saadin; Bicer, Emre; Bayindir, RamazanThis article presents an in-depth evaluation of electricity consumption predictions using the Seasonal Autoregressive Integrated Moving Average (SARIMA) model. Leveraging historical electricity consumption data, the SARIMA model demonstrates a commendable ability to forecast future consumption patterns. Our analysis reveals a strong alignment between the model's predictions and actual consumption data, affirming the efficacy of time series modeling in capturing complex energy consumption dynamics. Notably, while the model excels in predicting near-term consumption trends, uncertainties widen for long-term forecasts, prompting critical reflections on the model's evolving accuracy and reliability over time. For future research endeavors, we recommend comparing the performance of diverse time series models to discern optimal modeling approaches. Further optimization of model parameters stands as a paramount endeavor to refine prediction accuracy and mitigate uncertainties. Specifically, efforts to identify and address potential overfitting or underfitting tendencies within the model are advised. Additionally, leveraging supplementary data sources and integrating seasonal factors could bolster the reliability of future predictions, expanding the model's predictive scope and ensuring more robust and precise forecasts.Öğe Comparative Analysis of Commonly Used Machine Learning Approaches for Li-Ion Battery Performance Prediction and Management in Electric Vehicles(Mdpi, 2024) Oyucu, Saadin; Dogan, Ferdi; Aksoz, Ahmet; Bicer, EmreThe significant role of Li-ion batteries (LIBs) in electric vehicles (EVs) emphasizes their advantages in terms of energy density, being lightweight, and being environmentally sustainable. Despite their obstacles, such as costs, safety concerns, and recycling challenges, LIBs are crucial in terms of the popularity of EVs. The accurate prediction and management of LIBs in EVs are essential, and machine learning-based methods have been explored in order to estimate parameters such as the state of charge (SoC), the state of health (SoH), and the state of power (SoP). Various machine learning techniques, including support vector machines, decision trees, and deep learning, have been employed for predicting LIB states. This study proposes a methodology for comparative analysis, focusing on classical and deep learning approaches, and discusses enhancements to the LSTM (long short-term memory) and Bi-LSTM (bidirectional long short-term memory) methods. Evaluation metrics such as MSE, MAE, RMSE, and R-squared are applied to assess the proposed methods' performances. The study aims to contribute to technological advancements in the electric vehicle industry by predicting the performance of LIBs. The structure of the rest of the study is outlined, covering materials and methods, LIB data preparation, analysis, the proposal of machine learning models, evaluations, and concluding remarks, with recommendations for future studies.Öğe Creating a Quasi-Resonant Induction Cooktop Integrating Zero-Voltage Switching (ZVS) and Load Management(MDPI, 2024) Aksoz, AhmetThis study aims to elucidate the development and construction of a durable induction cooktop, with key considerations including efficiency, power customization, and safety features. The intricate processes involved in crafting a 3.5 kW induction burner are thoroughly examined, encompassing simulations for quasi-resonant inverters, the meticulous selection of induction coils and capacitors, the implementation of practical Analog-to-Digital Converter (ADC) filtration, pulse width modulation (PWM) driving techniques, and the integration of protection mechanisms. Leveraging an ARM-based microcontroller enabled the attainment of diverse objectives such as Zero-Voltage Switching (ZVS), safeguarding IGBTs, facilitating user interaction through a user-friendly interface, and enabling load detection capabilities. Furthermore, the capability to gauge and adjust output power based on user preferences was incorporated. Subsequently, rigorous testing was conducted to evaluate the functionality and applicability of the device in real-world scenarios.Öğe Deep Learning Forecasting Model for Market Demand of Electric Vehicles(MDPI, 2024) Simsek, Ahmed Ihsan; Koc, Erdinc; Tasdemir, Beste Desticioglu; Aksoz, Ahmet; Turkoglu, Muammer; Sengur, AbdulkadirThe increasing demand for electric vehicles (EVs) requires accurate forecasting to support strategic decisions by manufacturers, policymakers, investors, and infrastructure developers. As EV adoption accelerates due to environmental concerns and technological advances, understanding and predicting this demand becomes critical. In light of these considerations, this study presents an innovative methodology for forecasting EV demand. This model, called EVs-PredNet, is developed using deep learning methods such as LSTM (Long Short-Term Memory) and CNNs (Convolutional Neural Networks). The model comprises convolutional, activation function, max pooling, LSTM, and dense layers. Experimental research has investigated four different categories of electric vehicles: battery electric vehicles (BEV), hybrid electric vehicles (HEV), plug-in hybrid electric vehicles (PHEV), and all electric vehicles (ALL). Performance measures were calculated after conducting experimental studies to assess the model's ability to predict electric vehicle demand. When the performance measures (mean absolute error, root mean square error, mean squared error, R-Squared) of EVs-PredNet and machine learning regression methods are compared, the proposed model is more effective than the other forecasting methods. The experimental results demonstrate the effectiveness of the proposed approach in forecasting the electric vehicle demand. This model is considered to have significant application potential in assessing the adoption and demand of electric vehicles. This study aims to improve the reliability of forecasting future demand in the electric vehicle market and to develop relevant approaches.Öğe Discharge Capacity Estimation for Li-Ion Batteries: A Comparative Study(Mdpi, 2024) Oyucu, Saadin; Dumen, Sezer; Duru, Iremnur; Aksoz, Ahmet; Bicer, EmreLi-ion batteries are integral to various applications, ranging from electric vehicles to mobile devices, because of their high energy density and user friendliness. The assessment of the Li-ion state of heath stands as a crucial research domain, aiming to innovate safer and more effective battery management systems that can predict and promptly report any operational discrepancies. To achieve this, an array of machine learning (ML) and artificial intelligence (AI) methodologies have been employed to analyze data from Li-ion batteries, facilitating the estimation of critical parameters like state of charge (SoC) and state of health (SoH). The continuous enhancement of ML and AI algorithm efficiency remains a pivotal focus of scholarly inquiry. Our study distinguishes itself by separately evaluating traditional machine learning frameworks and advanced deep learning paradigms to determine their respective efficacy in predictive modeling. We dissected the performances of an assortment of models, spanning from conventional ML techniques to sophisticated, hybrid deep learning constructs. Our investigation provides a granular analysis of each model's utility, promoting an informed and strategic integration of ML and AI in Li-ion battery state of health prognostics. Specifically, a utilization of machine learning algorithms such as Random Forests (RFs) and eXtreme Gradient Boosting (XGBoost), alongside regression models like Elastic Net and foundational neural network approaches including Multilayer Perceptron (MLP) were studied. Furthermore, our research investigated the enhancement of time series analysis using intricate models like Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) and their outcomes with those of hybrid models, including a RNN-long short-term memory (LSTM), CNN-LSTM, CNN-Gated Recurrent Unit (GRU) and RNN-GRU. Comparative evaluations reveal that the RNN-LSTM configuration achieved a Mean Squared Error (MSE) of 0.043, R-Squared of 0.758, Root Mean Square Error (RMSE) of 0.208, and Mean Absolute Error (MAE) of 0.124, whereas the CNN-LSTM framework reported an MSE of 0.039, R-Squared of 0.782, RMSE of 0.197, and MAE of 0.122, underscoring the potential of deep learning-based hybrid models in advancing the accuracy of battery state of health assessments.Öğe Empowering Sustainability: A Consumer-Centric Analysis Based on Advanced Electricity Consumption Predictions(Mdpi, 2024) Durmus Senyapar, Hafize Nurgul; Aksoz, AhmetThis study addresses the critical challenge of accurately forecasting electricity consumption by utilizing Exponential Smoothing and Seasonal Autoregressive Integrated Moving Average (SARIMA) models. The research aims to enhance the precision of forecasting in the dynamic energy landscape and reveals promising outcomes by employing a robust methodology involving model application to a large amount of consumption data. Exponential Smoothing demonstrates accurate predictions, as evidenced by a low Sum of Squared Errors (SSE) of 0.469. SARIMA, with its seasonal ARIMA structure, outperforms Exponential Smoothing, achieving lower Mean Absolute Percentage Error (MAPE) values on both training (2.21%) and test (2.44%) datasets. This study recommends the adoption of SARIMA models, supported by lower MAPE values, to influence technology adoption and future-proof decision-making. This study highlights the societal implications of informed energy planning, including enhanced sustainability, cost savings, and improved resource allocation for communities and industries. The synthesis of model analysis, technological integration, and consumer-centric approaches marks a significant stride toward a resilient and efficient energy ecosystem. Decision-makers, stakeholders, and researchers may leverage findings for sustainable, adaptive, and consumer-centric energy planning, positioning the sector to address evolving challenges effectively and empowering consumers while maintaining energy efficiency.Öğe Ensemble Learning in Li-Ion Battery Management Systems: Focus on Voting Regression for Capacity Estimation(Institute of Electrical and Electronics Engineers Inc., 2024) Asal, Burcak; Oyucu, Saadin; Aksoz, AhmetAccurate estimation of discharge capacity in lithium-ion batteries is very important for optimizing their performance and longevity and directly affects the efficiency of battery management systems (BMS). Traditional models frequently encounter challenges in handling the non-linearities and complex interdependencies inherent in battery behavior. This study introduces a robust voting regression-based approach that combines multiple regression models to improve prediction accuracy and reliability. We employ four different regression configurations, Multi-Layer Perceptron (MLP), Random Forest (RF), Linear Regression and K-Nearest Neighbor (KNN) to form a combined estimator through voting mechanisms. The robustness of our voting-based approach is further validated through extensive experimentation on two real-world battery datasets including various operational conditions. Our results show that the voting regression approach we propose provides stable and accurate performance, making it a viable tool for different BMS applications. We also discuss the potential practical implications of our proposed voting approach and suggest directions for future research to further refine and adapt this approach to different types of battery technologies and configurations. © 2024 IEEE.Öğe Estimating Smart Grid Stability with Hybrid RNN plus LSTM Deep Learning Approach(IEEE, 2024) Oyucu, Saadin; Sagiroglu, Seref; Aksoz, Ahmet; Bicer, EmreSmart grids are faced with a range of challenges, such as the development of communication infrastructure, cybersecurity threats, data privacy, and the protection of user information, due to their complex structure. Another key challenge faced by smart grids is the stability issues arising from variable energy sources and consumption patterns. In these complex grid systems where energy demand and supply need to be balanced instantly, stability predictions play a significant role in foreseeing potential disruptions and optimizing energy flow. Therefore, within the scope of this study, a hybrid structure utilizing Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM) networks is employed for stability classification to predict grid stability. This hybrid model combines the ability of RNN to recognize relationships between consecutive data points with LSTM's capability to preserve long-term dependencies. The results obtained indicate that the model exhibited stable performance with accuracy rates of 98.06% and 98.02% at 50 and 100 epochs, respectively. The findings of this study contribute valuable insights to research on the management and stability of smart grids, enabling energy systems to be operated more reliably and efficiently.Öğe Hybrid AI-Powered Real-Time Distributed Denial of Service Detection and Traffic Monitoring for Software-Defined-Based Vehicular Ad Hoc Networks: A New Paradigm for Securing Intelligent Transportation Networks(MDPI, 2024) Polat, Onur; Oyucu, Saadin; Turkoglu, Muammer; Polat, Hueseyin; Aksoz, Ahmet; Yardimci, FahriVehicular Ad Hoc Networks (VANETs) are wireless networks that improve traffic efficiency, safety, and comfort for smart vehicle users. However, with the rise of smart and electric vehicles, traditional VANETs struggle with issues like scalability, management, energy efficiency, and dynamic pricing. Software Defined Networking (SDN) can help address these challenges by centralizing network control. The integration of SDN with VANETs, forming Software Defined-based VANETs (SD-VANETs), shows promise for intelligent transportation, particularly with autonomous vehicles. Nevertheless, SD-VANETs are susceptible to cyberattacks, especially Distributed Denial of Service (DDoS) attacks, making cybersecurity a crucial consideration for their future development. This study proposes a security system that incorporates a hybrid artificial intelligence model to detect DDoS attacks targeting the SDN controller in SD-VANET architecture. The proposed system is designed to operate as a module within the SDN controller, enabling the detection of DDoS attacks. The proposed attack detection methodology involves the collection of network traffic data, data processing, and the classification of these data. This methodology is based on a hybrid artificial intelligence model that combines a one-dimensional Convolutional Neural Network (1D-CNN) and Decision Tree models. According to experimental results, the proposed attack detection system identified that approximately 90% of the traffic in the SD-VANET network under DDoS attack consisted of malicious DDoS traffic flows. These results demonstrate that the proposed security system provides a promising solution for detecting DDoS attacks targeting the SD-VANET architecture.Öğe Impact of Environmental Conditions on Renewable Energy Prediction: An Investigation Through Tree-Based Community Learning(MDPI, 2025) Dogan, Ferdi; Oyucu, Saadin; Unsal, Derya Betul; Aksoz, Ahmet; Vafaeipour, MajidThe real-time prediction of energy production is essential for effective energy management and planning. Forecasts are essential in various areas, including the efficient utilization of energy resources, the provision of energy flexibility services, decision-making amidst uncertainty, the balancing of supply and demand, and the optimization of online energy systems. This study examines the use of tree-based ensemble learning models for renewable energy production prediction, focusing on environmental factors such as temperature, pressure, and humidity. The study's primary contribution lies in demonstrating the effectiveness of the bagged trees model in reducing overfitting and achieving higher accuracy compared to other models, while maintaining computational efficiency. The results indicate that less sophisticated models are inadequate for accurately representing complex datasets. The results evaluate the effectiveness of machine learning methods in delivering valuable insights for energy sectors managing environmental conditions and predicting renewable energy sourcesÖğe Integrating Machine Learning and MLOps for Wind Energy Forecasting: A Comparative Analysis and Optimization Study on Türkiye's Wind Data(Mdpi, 2024) Oyucu, Saadin; Aksoz, AhmetThis study conducted a detailed comparative analysis of various machine learning models to enhance wind energy forecasts, including linear regression, decision tree, random forest, gradient boosting machine, XGBoost, LightGBM, and CatBoost. Furthermore, it developed an end-to-end MLOps pipeline leveraging SCADA data from a wind turbine in T & uuml;rkiye. This research not only compared models using the RMSE metric for selection and optimization but also explored in detail the impact of integrating machine learning with MLOps on the precision of energy production forecasts. It investigated the suitability and efficiency of ML models in predicting wind energy with MLOps integration. The study explored ways to improve LightGBM algorithm performance through hyperparameter tuning and Docker utilization. It also highlighted challenges in speeding up MLOps development and deployment processes. Model performance was assessed using the RMSE metric, conducting a comparative evaluation across different models. The findings revealed that the RMSE values among the regression models ranged from 460 kW to 192 kW. Focusing on enhancing LightGBM, the research decreased the RMSE value to 190.34 kW. Despite facing technical and operational hurdles, the implementation of MLOps was proven to enhance the speed (latency of 9 ms), reliability (through Docker encapsulation), and scalability (using Docker swarm) of machine learning endeavors.Öğe Interpreting CNN-RNN Hybrid Model-Based Ensemble Learning with Explainable Artificial Intelligence to Predict the Performance of Li-Ion Batteries in Drone Flights(MDPI, 2024) Ersoz, Betul; Oyucu, Saadin; Aksoz, Ahmet; Sagiroglu, Seref; Bicer, EmreLi-ion batteries are important in modern technology, especially for drones, due to their high energy density, long cycle life, and lightweight properties. Predicting their performance is crucial for enhancing drone flight safety, optimizing operations, and reducing costs. This involves using advanced techniques like machine learning (e.g., Convolutional Neural Network-CNNs, Recurrent Neural Network-RNNs), statistical modeling (e.g., Kalman Filtering), and explainable AI (e.g., SHAP, LIME, PDP) to forecast battery behavior, extend battery life, and improve drone efficiency. The study aims to develop a CNN-RNN-based ensemble model, enhanced with explainable AI, to predict key battery metrics during drone flights. The model's predictions will aid in enhancing battery performance via continuous, data-driven monitoring, improve drone safety, optimize operations, and reduce greenhouse gas emissions through advanced recycling methods. In the present study, comparisons are made for the behaviors of two different drone Li-ion batteries, numbered 92 and 129. The ensemble model in Drone 92 showed the best performance with MAE (0.00032), RMSE (0.00067), and R2 (0.98665) scores. Similarly, the ensemble model in Drone 129 showed the best performance with MAE (0.00030), RMSE (0.00044), and R2 (0.98094) performance metrics. Similar performance results are obtained in the two predictions. However, drone 129 has a minimally lower error rate. When the Partial Dependence Plots results, which are one of the explainable AI (XAI) techniques, are interpreted with the decision tree algorithm, the effect of the Current (A) value on the model estimations in both drone flights is quite evident. When the current value is around -4, the model is more sensitive and shows more changes. This study will establish benchmarks for future research and foster advancements in drone and battery technologies through extensive testing.Öğe Multi-Stage Learning Framework Using Convolutional Neural Network and Decision Tree-Based Classification for Detection of DDoS Pandemic Attacks in SDN-Based SCADA Systems(Mdpi, 2024) Polat, Onur; Turkoglu, Muammer; Polat, Huseyin; Oyucu, Saadin; Uzen, Huseyin; Yardimci, Fahri; Aksoz, AhmetSupervisory Control and Data Acquisition (SCADA) systems, which play a critical role in monitoring, managing, and controlling industrial processes, face flexibility, scalability, and management difficulties arising from traditional network structures. Software-defined networking (SDN) offers a new opportunity to overcome the challenges traditional SCADA networks face, based on the concept of separating the control and data plane. Although integrating the SDN architecture into SCADA systems offers many advantages, it cannot address security concerns against cyber-attacks such as a distributed denial of service (DDoS). The fact that SDN has centralized management and programmability features causes attackers to carry out attacks that specifically target the SDN controller and data plane. If DDoS attacks against the SDN-based SCADA network are not detected and precautions are not taken, they can cause chaos and have terrible consequences. By detecting a possible DDoS attack at an early stage, security measures that can reduce the impact of the attack can be taken immediately, and the likelihood of being a direct victim of the attack decreases. This study proposes a multi-stage learning model using a 1-dimensional convolutional neural network (1D-CNN) and decision tree-based classification to detect DDoS attacks in SDN-based SCADA systems effectively. A new dataset containing various attack scenarios on a specific experimental network topology was created to be used in the training and testing phases of this model. According to the experimental results of this study, the proposed model achieved a 97.8% accuracy rate in DDoS-attack detection. The proposed multi-stage learning model shows that high-performance results can be achieved in detecting DDoS attacks against SDN-based SCADA systems.Öğe Optimizing Lithium-Ion Battery Performance: Integrating Machine Learning and Explainable AI for Enhanced Energy Management(Mdpi, 2024) Oyucu, Saadin; Ersoz, Betul; Sagiroglu, Seref; Aksoz, Ahmet; Bicer, EmreManaging the capacity of lithium-ion batteries (LiBs) accurately, particularly in large-scale applications, enhances the cost-effectiveness of energy storage systems. Less frequent replacement or maintenance of LiBs results in cost savings in the long term. Therefore, in this study, AdaBoost, gradient boosting, XGBoost, LightGBM, CatBoost, and ensemble learning models were employed to predict the discharge capacity of LiBs. The prediction performances of each model were compared based on mean absolute error (MAE), mean squared error (MSE), and R-squared values. The research findings reveal that the LightGBM model exhibited the lowest MAE (0.103) and MSE (0.019) values and the highest R-squared (0.887) value, thus demonstrating the strongest correlation in predictions. Gradient boosting and XGBoost models showed similar performance levels but ranked just below LightGBM. The competitive performance of the ensemble model indicates that combining multiple models could lead to an overall performance improvement. Furthermore, the study incorporates an analysis of key features affecting model predictions using SHAP (Shapley additive explanations) values within the framework of explainable artificial intelligence (XAI). This analysis evaluates the impact of features such as temperature, cycle index, voltage, and current on predictions, revealing a significant effect of temperature on discharge capacity. The results of this study emphasize the potential of machine learning models in LiB management within the XAI framework and demonstrate how these technologies could play a strategic role in optimizing energy storage systems.Öğe Parameters Estimation of Orthopedic Drill(IEEE, 2019) Torun, Yunis; Ozturk, Ahmet; Aksoz, Ahmet; Pazarci, OzhanIn medical drills, battery drills are highly preferred due to their comfort. Battery discharge during operation is a very important problem. For this reason, it is important to use the control method to minimize the energy consumption and to keep the drill speed in the desired range. In order to design an efficient control in any system, the system model and system parameters must be known. In this study, parameter estimation of a direct current geared motor used in medical drills was carried out by using system input/output signals. Nonlinear Least Squares method was used to estimate parameters then Genetic Algorithm optimization used to improve the model estimation performance. It was observed that the data obtained from the physical system and the model behaviors were similar in an acceptable range. Modeling improvement has been proven based on the criteria of Mean Square Error, Root Mean Square Error, Sum Square Error and Root Sum Square Error.
