Andrii Bosak

Andrii Bosak

Auto-generated profile, based on publicly available information

Andrii Bosak

General Energy Institute of NAS of Ukraine

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
9 March 2026 in System Research in Energy
Verified
Journal Article
Open access
PDF file
GKGanna Kostenko
Ganna Kostenko
ABAndrii Bosak
Andrii Bosak
YSYurii Sapsai
Yurii Sapsai
Artur ZaporozhetsArtur Zaporozhets
Artur Zaporozhets
The rapid expansion of battery-powered mobility and renewable energy integration has amplified the demand for energy storage as well as for reliable and standardized data supporting predictive modeling and lifecycle management of lithium-ion batteries. Experimental testing of batteries is costly, time-consuming, and limited by laboratory constraints, which makes open-access datasets an invaluable foundation for comparative studies, model validation, and reproducible analytics. However, the diversity of available datasets in terms of format, chemistry, and test conditions complicates their systematic use in degradation and KPI-based research. This study develops an information and analytical framework for the selection, evaluation, and classification of open-access Li-ion battery datasets applicable to both first- and second-life applications. Fifteen representative datasets were analyzed and grouped by chemistry, cycling depth, and metadata completeness, with additional assessment of data integrity and traceability according to FAIR principles. The analysis identifies dataset suitability for specific analytical domains: Sandia and HNEI provide long-term degradation and RUL modeling data; Stanford SLB and UC Davis Microgrid datasets enable operational and KPI analysis under realistic usage conditions; PulseBat and Panasonic PF datasets contribute to safety, reliability, and probabilistic risk evaluation. The proposed framework establishes clear connections between raw data, degradation indicators, and system-level metrics such as LCOS, utilization rate, and lifecycle efficiency. It also introduces a structured mapping of data relevance to various modeling objectives, supporting reproducible and cross-compatible research across laboratories and applications. Beyond comparative analysis, the study emphasizes the critical role of metadata completeness, DOI-based traceability, and repository-level version control in building trustworthy digital twins and regulatory tools such as the EU Battery Passport. The results provide a foundation for harmonized, data-driven methodologies that bridge experimental data, predictive models, and KPI-based lifecycle management, promoting transparency, interoperability, and sustainability in battery research and deployment. Additionally, a detailed comparative analysis of three representative high-quality datasets (Sandia NMC, HNEI LFP, and NASA/CALCE NMC/NCA) is presented to illustrate chemistry-dependent degradation behaviour and quantify inter-dataset divergence relevant for second-life modeling. Keywords: lithium-ion batteries, open datasets, lifecycle management, degradation modeling, KPI framework, second-life applications, data-driven energy systems.

31 October 2025 in POWER ENGINEERING: economics, technique, ecology
Verified
Journal Article
Open access
PDF file
ABAndrii Bosak
Andrii Bosak
GKGanna Kostenko
Ganna Kostenko
Artur ZaporozhetsArtur Zaporozhets
Artur Zaporozhets
Purpose of the research: This study examines the potential of Vehicle-to-Load (V2L) technology as a practical tool for enhancing grid load flexibility through decentralized energy support. Special emphasis is placed on the influence of battery characteristics on V2L effectiveness. Methodology: A comprehensive technical and comparative analysis is conducted, focusing on key battery-related parameters that affect V2L performance. These include usable capacity, inverter constraints, thermal regulation, and degradation behavior. The study evaluates three dominant battery chemistries—lithium iron phosphate (LFP), nickel manganese cobalt (NMC), and lithium titanate (LTO)—and three major cell formats (prismatic, pouch, cylindrical) to assess their suitability for real-world V2L use cases. Results: LFP batteries demonstrate the best overall trade-off between safety, durability, and cost-efficiency under partial load conditions typical for V2L. Inverter limitations and thermal management are shown to significantly affect performance and longevity. Among cell formats, pouch cells are more vulnerable to thermal stress, while prismatic cells perform more consistently under moderate cycling regimes. Conclusions: V2L technology offers a viable and increasingly accessible pathway to improve household-level energy autonomy and reduce strain on centralized grids during peak periods. The growing availability of commercial EVs equipped with LFP batteries strengthens the case for integrating V2L into smart grid strategies aimed at increasing load flexibility and resilience.

11 June 2021 in Transactions of Kremenchuk Mykhailo Ostrohradskyi National University
Verified
Journal Article
Closed access
ABAndrii Bosak
Andrii Bosak
NINational Technical University of Ukraine «Igor Sikorsky Kyiv Polytechnic Institute»
National Technical University of Ukraine «Igor Sikorsky Kyiv Polytechnic Institute»
ACAlexandr Chermalykh
Alexandr Chermalykh
BABosak Alla
Bosak Alla
VDVolodymyr Dubovyk
Volodymyr Dubovyk
Purpose. The purpose of this research is to increase the efficiency of the charging station, which is to meet the demand of energy for charging the batteries of electric vehicles without exceeding the limits of the grid restrictions. Methodology. The model of a controlled charging system was considered to study the efficiency of charging stations with the quantitative connection of electric vehicles to the charger under conditions of power limitation. Results. The load graph during the controlled charging method does not exceed the set limit voltage and power consumption settings, but with the uncontrolled method there is an overload of 16% during peak hours. The main disadvantage of the conventional method of charging with restrictions is that EVs that have been connected to the charging slots later than other EVs, were not charged in conditions of limited power supply. The charging stations require a charging system that does not have such a disadvantage – it is necessary that almost all connected electric vehicles fill their required level of charge and meet the limitations of the network at the same time. Investigating such a system is the task of future research. Originality. Every year the number of electric vehicles in Ukraine is growing rapidly. This also determines the development of a network of charging stations, both commercial and for domestic use. According to the Rules for the use of electricity, the contract for electricity consumption specifies the allowable capacity for the household or enterprise. Violation of this limit may cause the facility to become disconnected from the power supply with additional penalties. Practical value. The solutions to the problem of charging a large number of EVs are mainly based on different types of planning and creating a profile of their charging scenarios with technical limitations. These restrictions depend on the situation of the charging station (the number of connected EVs, their charging levels, the possibility of simultaneous charging for a few EVs, and the declared charging time). The charging period convenient for the EV owner and the network load schedule may conflict. The development of a realtime charging control system should provide each EV guaranteed charging even in conditions of limited power consumption. References 13, tables 4, figures 18.

28 March 2026 in Vidnovluvana energetika
Verified
Journal Article
Closed access
ЯСЯ. З. Сорочинський
Я. З. Сорочинський
ВДВ. Г. Дубовик
В. Г. Дубовик
ЛКЛ. Я. Кулаковський
Л. Я. Кулаковський
АБА. В. Босак
А. В. Босак
Accurate State of Health (SOH) estimation is critically important for the safe and long-term operation of lithium-ion batteries. However, the implementation of estimation algorithms in embedded Battery Management Systems (BMS) is significantly limited by available computational resources. This paper presents a scientifically grounded comparative analysis of two computationally efficient machine learning methods – Support Vector Regression (SVR) and Random Forest – to determine the optimal algorithm for practical implementation, with a dual focus on prediction accuracy and computational efficiency. Based on an open dataset, systematic hyperparameter tuning with GridSearchCV and cross-validation was conducted for both models. The performance of the final, optimized models was evaluated by accuracy metrics (RMSE, MAE) and indicators of suitability for embedded systems (forecasting time, model size). The results showed that even after thorough optimization, the SVR model demonstrated higher prediction accuracy (RMSE 2.67% vs. 3.08% for Random Forest). An even more significant advantage was found in the efficiency analysis: SVR was found to be almost 170 times faster (forecast time of 0.119 ms vs 20.570 ms) and 200 times more compact (model size 8.6 kB vs 1718.6 kB). It is concluded that, for the problem of estimating SOH based on cyclic data, the SVR model is the optimal candidate for practical implementation in embedded BMS. It offers the best balance of high accuracy and minimal hardware resource requirements, outperforming Random Forest across all key criteria for engineering practice.


28 October 2026 in Energy
Verified
Journal Article
Closed access
DMDmytro Matushkin
Dmytro Matushkin
Artur ZaporozhetsArtur Zaporozhets
Artur Zaporozhets
ABAndrii Bosak
Andrii Bosak
IBIhor Buratynskyi
Ihor Buratynskyi
VVVladyslav Verpeta
Vladyslav Verpeta