Inter national J our nal of A pplied P o wer Engineering (IJ APE) V ol. 14, No. 3, September 2025, pp. 569 578 ISSN: 2252-8792, DOI: 10.11591/ijape.v14.i3.pp569-578 569 AI-dri v en solutions f or Li-ion battery perf ormance and pr ediction Sthitprajna Mishra 1 , Chinmoy K umar P anigrahi 1 , Subhra Debdas 1 , Atri Bandy opadh yay 2 , Srikanth V elpula 3 , Amit K umar Sahoo 4 , P abitra K umar T ripath y 5 1 School of Electrical Engineering, KIIT Deemed to be Uni v ersity , Bhubanesw ar , India 2 School of Computer Engineering, KIIT Deemed to be Uni v ersity , Bhubanesw ar , India 3 Department of Electrical and Electronics Engineering, SR Uni v ersity , W arang al, India 4 Department of Electrical and Electronics Engineering, Centurion Uni v ersity T echnology and Management, Bhubanesw ar , India 5 Department of Computer Science and Engineering, Kalam Institute of T echnology , Berhampur , India Article Inf o Article history: Recei v ed Jul 10, 2024 Re vised Jan 11, 2025 Accepted Jan 19, 2025 K eyw ords: Battery management system Lithium-ion battery Remaining useful life State of char ge State of health ABSTRA CT Batteries serv e as crucial po wer sources for essential port able de vices lik e elec- tric v ehicles, smartphones, and laptops. The widespread adoption of Li-ion bat- teries, while bene cial, has unfortunately led to a sur ge in adv erse incidents. The sudden f ailure of batteries in both industrial and lightweight applications poses signicant economic risks across v arious industries. Consequently , researchers are intensifying their focus on enhancing battery state estimation, and manage- ment systems and predicting remaining useful life (R UL). This paper is struc- tured into three main sections. Firstly , it delv es into the acquisition of battery data, encompassing both commercially a v ailable and freely accessible Li-ion battery datasets. Secondly , the e xploration e xtends to techniques for estimating battery states through adv anced battery management systems. The paper in v es- tig ates battery R UL estimation, c ate gorizing and e v aluating di v erse prognostic methods appl ied to Li-ion batteries based on crucial performance parameters. The re vie w includes scrutin y of commercially and publicly a v ailable datasets for v arious battery models and conditions, considering dif ferent battery states and the role of adv anced battery management system (BMS). In the nal section, the paper concludes with a comparati v e analysis of Li-ion battery R UL prediction, incorporating e xploration into v arious R UL prediction algorithms, and mathe- matical models, and introducing an AI-based cloud monitoring system. This is an open access article under the CC BY -SA license . Corresponding A uthor: Sthitprajna Mishra School of Electrical Engineering, KIIT Deemed to be Uni v ersity Bhubanesw ar , Odisha, India Email: sthitprajnamishra26@gmail.com 1. INTR ODUCTION It has turned out that the electric v ehicles (EVs) as well as the clean ener gy technologies continue t o e xpand at a f ast pace, which has become one of the dominant w ays of tackling global en vironmental and ener gy issues. The con v entional ener gy sources such as those that produce toxic g ases are gradually being substituted by en vironmentally friendly electric automobiles. Central to EV technology is the battery , which comprises four critical components: electric design, mechanical design, thermal design as well as battery management system (BMS) [1]. Due to the social de v elopment and manuf acturing of hi ghly ef cient and instantaneous po wer control technologies as a result of the adv ancing modern technologies, batteries particularly those used within J ournal homepage: http://ijape .iaescor e .com Evaluation Warning : The document was created with Spire.PDF for Python.
570 ISSN: 2252-8792 EVs ha v e become e xtremely vital in an attempt at of fering e v er reliable green ener gy storage mechanisms. Ener gy storage has hence emer ged as one of the most promising industries due to the increasing need for ener gy-intensi v e products such as consumer electronics and adv ancement in the utilization of rene w able ener gy [2], [3]. Ne w ener gy sources displacing the traditional and continuous supply systems including nuclear , coal, and oil and replacing them with rene w able ener gy systems including wind and solar ener gy is causing disruption of the systems especially in de v eloping countries. This shift is putting in place conditions that require ne w adv anced ener gy st orage systems to respond to comple x ener gy mark et structures [4], [5]. Dif fe rent types of storage such as electrochemical which has a high ef cienc y , mechanical, chemical, and thermal storage ha v e dif ferent capacities, and days of storage. From these, the rechar geable electrochemical systems bearing the most popularity due to con v eniences such as high ener gy density , light weight and e xibility particularly the lithium- ion (Li-ion) batteries. Li-ion batteries are higher performing than other battery technologies including lead- acid, redox o w , sodium sulphur batteries among others that mak e its application in a viation industry , satellite communication, marine applications and EVs. The y also dri v e v arious home appliances lik e refrigerators, laptops and w andering de vices lik e mobile phones among others [6]. This report implies that Li-ion batteries ha v e some of the unique benets such as long c ycle-life, high ener gy density and lo w maintenance hence the technology is uni v ersally applied. But the y also ha v e some disadv antages for e xample, the y are e xpensi v e, are easily damaged when the y are fully char ged or cause re risks. In the case of the electrical v ehicles, tw o v ariables namely the stat e of health, and the state of char ge signicantly af fect both the safety , and output of the v ehicle. Battery ener gy management (BEM) strate gies therefore seek to enhance the state of health (SoH) and state of char ge (SoC) of Li-ion batteries for increased life c ycle of the battery bank together with increasing ef cienc y of the induction motor [7], [8]. While there are certain con v entional w ays of controlling speed such as using dynamo-meter and other similar technologies, the recent inno v ati v e BEM methods in v olv e model-in-loop t echniques that mimic battery performance and there- fore reduces the battery’ s life c ycle. The y pre v ent the reduction of SoC rate and slo w do wn the SoH decline that distorts the general battery dependability [9]. As t he use of Li-ion batteries e xpanded widely there has been a signicant focus on the number of char ge dischar ge c ycles, remaining useful life (R UL), and de gradation anal- ysis. It is v ery important for accurate estimation of R UL pre v enting battery f ailures and maintaining superior system performance [10]. Although there are so man y benets associated with Li-ion batteries, the y e xperience high de gradation and f ailure rates, this mak es battery management system and accurate R UL models to be more crucial. Better estimation of R UL leads to the necessity of ha ving better datasets, and se v eral or g anizations ha v e been de v eloping datasets for dif ferent battery models [11], [12]. These datasets are v ery useful in enhancing battery health estimations, and also in minim izing the time tak en in de v eloping ne w datasets needed for battery research, as well as enhancing the dependability of systems used in battery management. Using these datasets, researchers will be able to increase the accurac y of R UL estimations as well as battery health assessment [13], [14]. Man y approaches ha v e been made to ef fecti v e and accurate assessment of SoH and R UL in the Li-ion bat- teries. Through a system simulation approach in v olving electrochemical techniques and data analysis methods of statistics, e v aluation of state estimation algorithms is pro vided [15]. Furthermore, support v ector machines (SVM) has been used to impro v e the accurac y of R UL of battery through impro ving the accurac y of the men- tioned model. The current de v elopments in the Li-ion battery technology call for more e xtensi v e and up-to-date re vie ws of the methods aimed at the estim ation of R UL [16], [17]. Gaps in current research: Although there are adv ancements being made on the study of battery health and performance de gradation, there are still loopholes in measuring dif ferent methodologies a v ailable. Se v eral studies are still missing in the current research that addresses the interaction between the battery management algorithms and the estimation of R UL [18], [19]. This g ap is important for maximizing the performance of batteries as well as eliminating f ai lures. Thus, there is a need to e xplore BMS and R UL estimation models to obtain the best results in battery control [20]. Research contrib utions and ne w directions: This research aims at lling these g aps by pro viding an assessment of both the public and commercial datasets in batteries storage. It asses ses superior state estima- tion methods utilizing BMS and dissects dif ferent types of R UL prediction techniques. The study focuses on three k e y areas: another sub-eld is the battery data acquis ition, deep estimation of battery health utilizing progressi v e BMS, and methods of R UL estimation [21], [22]. This w ork also identies the pros and cons of the approaches presented when analyzing datasets and comparing R UL prediction models. The contrib ution of this study is therefore in comparing a number R UL prediction al g or ithms and mathematical model [23]. The paper presents a cloud monitoring system that is inte n de d to increase the capabilities of stream processing and upgrade battery health predictions. This study establishes that it is possible to impro v e the accurac y of the R UL Int J Appl Po wer Eng, V ol. 14, No. 3, September 2025: 569–578 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Appl Po wer Eng ISSN: 2252-8792 571 of Li-ion cells by incorporating ndings from data analysis with other suitable methods [24], [25]. Future di- rections: In the future, the studies will continue on ho w to add machine learning and big data implemented into the battery c on t rol systems. The highlighted adv anced technologies in papers described ha v e the capabilities to enhance battery ef ci enc y , its durability and ener gy density . Furthermore, other methods such as correlation matrices, pair plot, time series plot and box plot will also be utilized to understand certain beha viour of Li-ion battery particularly during dischar ge c ycles. These visualization tools of fer v aluable insights into performance changes and de gradation patterns, further adv ancing battery health monitoring and R UL prediction beha viour during dischar ge c ycles. F or the prediction of R UL e xplanation sho wn in Figure 1. Figure 2 depicts a fundamental battery management system concept. This method aids in kno wledge of the dif ferences among numerous battery states and R UL, in particular within the conte xt of real-global b usiness and commercial conditions. The number one recognition of this o v ervie w is to pro vide a comparati v e look at battery R UL estimation procedures, incorporating battery management, and of fering complete statistics on commercially and freely a v ailable online battery datasets. Figure 1. Predicting R UL through Li-ion battery data acquisition process AI-driven solutions for Li-ion battery performance and pr ediction (Sthitpr ajna Mishr a) Evaluation Warning : The document was created with Spire.PDF for Python.
572 ISSN: 2252-8792 Figure 2. V isualizing battery management procedures 2. PR OPOSED METHOD The proposed frame w ork gi v es an impro v ed methodology and high le v el of sophistication for perfor - mance e v aluation and accurate estimation of the predicti v e capability of lithium-ion batteries supported by state of art data acquisition, BMS, and R UL assessment techniques. It unfolds across three k e y pillars: 2.1. Battery data acquisition This step is about the ef fecti v e use of the lithium-ion battery data pool which includes both proprietary and open-sourced data sets. Through highly tar geted choice of the best battery cells and the subsequent collec- tion of highly accurate parameter data related to the cell state of char ge, v oltage, current, and temperature, the study aims to arri v e at a set of v ery solid health indicators (HI) on which the accurate calculation of R ULs will be based. Some of the main trends observ ed in the process of data acquisition include the follo wing stringent requirements imposed on the functional performance of the data acquisition systems to achie v e e xtremely high dependability and v ersatility with respect to the operating conditions. 2.2. Estimation of battery states A wireless BMS on the programmable logic controllers (PLC), supervisory control and data acquisi- tion (SCAD A), and GSM modules is a modern solution that allo ws the monit oring of the battery’ s prerequisites such as v oltage, current, and temperature. This dynamic monitoring approach is highly reliable and e xible which mak es it possible to perform well in man y dif ferent en vironments. W ireless technology inte gration of fers enhanced data transfer and remote system management capabilities that enhance battery ef fecti v eness as well as decision-making. 2.3. R UL estimation methods R UL est imation methods: As in the pre vious phase, a detailed comparati v e e v aluation of dif ferent prognostic models is made with an emphasis put on R UL prediction. The technology comprises the AI-based methods of cloudy monitoring and the latest mathematical calculations f ac ilitating the further impro v ement of battery performance predictions. The y also augment the estimated R UL s precision while also pro viding insights into the battery beha vior under v arious scenarios, which is a paradigm shift in batt ery management. The proposed study is something unique i n the eld as i t embeds modern approaches to data acquisition, adv anced algorithms of prediction, and sophisticated BMS technologies into a single research for the purpose of maximizing ener gy storage solutions in electric v ehicle and portable applications. 3. METHODOLOGY W ireless implementation of battery management systems (BMS) is carried out with the help of PLC for data processing in real-t ime with the help of SCAD A systems and GSM modules for industrial temperature monitoring. This inte gration ensures seamless data handling and monitoring, enabl ing operators to k eep track of system parameters ef ciently . It also enhances s y s tem e xibility and reduces wiring comple xity , leading to easier maintenance and scalability . Int J Appl Po wer Eng, V ol. 14, No. 3, September 2025: 569–578 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Appl Po wer Eng ISSN: 2252-8792 573 3.1. Battery monitoring system (BMS) A 2.4 GHz wireless communication module also ensures lo w po wer and lo w-cost signal for transmi t- ting other important issues lik e v oltage, current, and temperature to f acilitate data acquisition from the battery system. The use of such modules enhances the reliability of wireless transmission while maintaining ener gy ef cienc y and cost-ef fecti v eness, making it suitable for lar ge-scale industrial applications. Additionally , it sup- ports seamless inte gration with IoT platforms, enabli ng remote monitoring and adv anced data analytics. This contrib utes to predicti v e maintenance and im pro v ed o v erall system performance. Furthermore, the scalability of these modules allo ws easy e xpansion of the monitoring netw ork as system requirements gro w . 3.2. Safety assessment In order to pre v ent early battery f ailures, a ne w safety assessment approach is established to detect and pre v ent possible battery f ailures. It monitors constantly all internal v ariables, including chemicals and ph ysical changes and e xternal f actors for instance, thermal and electrical loads. By analyzing condi tions in real-time, the system is able to pre v ent o v erchar ging, o v er -dischar ging, and o v erheating thus increasing battery life and pre v enting e xplosions. 3.3. Data acquisition and analysis In this w ork, a number of high-accurac y sensing elements are used in order to measure some char acter - istics of the battery such as its v oltage, its resistance, the current through it, the temperature and the capacitance. These measurements are tak en across eight modules; each of the modules comprises twelv e cells. The nominal v oltage is distrib uted with each of the indi vidual cells featuring 3.7 v olts of po wer potential, that tak es the total v oltage of a fully char ged module to 44.4 V . This w ay of connecting the battery allo ws the accurate and intricate observ ation of its w orking. Data acquisition process forms the type of c yclic beha vior of the battery in the process of dischar ging processes. Thus, through collecting such data, the researchers can plot time series plots illustrating the dynamic of these paramet ers. Such plots assist in tracking battery de gradation and the performance of the battery for each dischar ging c ycle being performed on the battery . In addition, it is possible to perform a correlation analysis aimed at making conclusions on interdependence of v arious parameters with each other , for instance, ho w temperature is dependent on v oltage or current. High accurac y sensors, collection of data at dif ferent stages during the dischar ging c ycles across long term gi v es a clear indication of the battery performance across the c ycles. It is also the information which can be v aluable to mak e the proper decisions on the use of the batteries in the future and increase the o v erall reliability of the system, as sho wn in Figure 3. Figure 3. V isualizing battery management procedures AI-driven solutions for Li-ion battery performance and pr ediction (Sthitpr ajna Mishr a) Evaluation Warning : The document was created with Spire.PDF for Python.
574 ISSN: 2252-8792 3.4. Statistical techniques and data-set selection Exploratory data analysis techniques including correlation analysis and box plots are emplo yed to de- termine the lik elihood of a relation between battery parameters as well as their trends within char ge-dischar ge c ycles. The data collected is subject to rigorous inclusion and e xclusion criteria, which means that only high quality data from commercial and public s ource is used. This mak e the outcome of the analysis reliable, repeat- able, and sound ha ving a solid groundw ork for future in v estig ations on lithium-ion battery performance and control. After the acquisition of e xperimental battery data, it becomes imperati v e to conduct assessments for impedance aging parameter estimation and capacity de gradation parameter estimation. These e v aluations are crucial in determining the remaining useful life (R UL) threshold or ascertaining whether it meets the specied criteria. This analytical process in v olv es measuring impedance aging and capacity de gradation to g auge the op- erational longe vity and performance deterioration of the battery , contrib uting to a comprehensi v e understanding of its life c ycle dynamics. 4. RESUL TS AND DISCUSSION Although there are distinct uses of R UL prediction methodologies for lithium-ion batteries (LIBs), most methodologies ha v e not yet been utilized to predict the R UL of LIBs, thus pointing t o areas that can be e xplored in the future. These une xplored areas of fer potential for adv ancing battery life forecasti n g tech- niques. Future research can le v erage emer ging tools lik e machine learning and h ybrid modeling to enhance prediction accurac y . By inte grating data dri v en approaches with ph ysics-based models, researchers can better capture comple x de gradation mechanisms under v aried operating conditions. Additionally , incorporating adap- ti v e algorithms that learn from real-time battery performance data can signicantly impro v e long-term R UL prediction and f acilitate smarter battery management systems. 4.1. Pr ediction of the beha viour of the battery The follo wing re gions present huge research scopes : In the outlined suggested model, there are three forms of assessment namely: i) The complementary nature of analytical methodologies with the similar method mak es it promising to enhance the prognostication accurac y of the R UL of lithium-ion cells. Thus, combining the identied basic approaches to data analysis with the other methods, Burns and Summer will belie v e that the dependability of the R UL results can be impro v ed considerably in practice. This combination of methods is necessary especially as the battery characteristics demonstrate more features in the de v elopment of its beha vior that af fects the predictions accurac y; ii) This has made it necessary to look for other better sources of data with which the accurac y of estimating the remaining useful life (R UL) can be enhanced. As mentioned earlier , to achie v e high le v els of R UL prediction it is imperati v e to w ork with high quality data therefore, future research should focus primary on seeking better data sources. This will be especially helpful because it will co v er data obtained under v arying operational conditions as well as bat tery states that will impro v e R UL estimation; iii) Multi-state joint estimation can be used to se v eral states such as state of char ge (SoC), state of ener gy (SoE), state of po wer (SoP), state of health (SoH), state of tempera ture (SoT), as well as state of stress (SoS) to e xpose the enhanced battery management methodologies. In other w ords, by jointly estimating these dif ferent states, the BMS pro vides a more accurate representation of battery conditions hence impro ving its capacity to estimate R UL whil e at the same time impro ving its o v erall performance; and i v) Hence other research potentia lities in the future of battery management may also include, AI and machine learning, and big data analysis. Thus, the combination of dif ferent approaches in the estimation enhances the le v el of precision and can form a basis for enhancement on the estimation of R UL with re g ard to BMS. 4.2. Dischar ging beha viour of Li-ion battery The conducted visualizations present a comprehensi v e analysis of the beha vior of lithium-ion (Li-ion) batteries during the dischar ge process. Through a series of analytical techniques, including the generation of a correlation matrix and the visualization of pair plots, the interrelationships between v arious parameters such as cell v oltage, current, and temperature were e xamined. T ime series plots were emplo yed to elucidate the temporal dynamics of these v ariables throughout the dischar ge c ycles, pro viding v aluable insights into their beha vior o v er tim e. Additionally , boxplots were utilized to illustrate the distri b ution of ce ll v oltage, current, and tempera ture across dif ferent c ycle numbers, enabling the identication of trends and patterns associated with c ycle progression. By systematically conducting these visualizations, a holistic understanding of the beha vior of Li-ion batteries during dischar ge w as attained, f acilitating informed decision-making and Int J Appl Po wer Eng, V ol. 14, No. 3, September 2025: 569–578 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Appl Po wer Eng ISSN: 2252-8792 575 optimization strate gies for battery performance enhancement. This approach contrib utes to the adv ancement of battery research and the de v elopment of ef cient ener gy storage solutions. W e mentioned the dif ferent techniques in Figure 4 and Figures 5(a), and 5(b) as correlation and box plot respecti v ely . Figure 4. Correlation matrix of Li-ion battery while dischar ging in dif ferent c ycle (a) (b) Figure 5. Dischar ging battery box plot according to the c ycles: (a) v oltage of the battery and (b) temperature of the battery 5. CONCLUSION In conclusion, this paper highlights the critical importance of adv ancing battery state estimation, and management systems, and forecasting the remaining useful life (R UL) of lithium-ion (Li-ion) batteries. The widespread adoption of Li-ion batteries in essential portable de vices such as smartphones, laptops, and EV AI-driven solutions for Li-ion battery performance and pr ediction (Sthitpr ajna Mishr a) Evaluation Warning : The document was created with Spire.PDF for Python.
576 ISSN: 2252-8792 has led to increased attention on mitig ating adv erse incidents and economic risks associated with battery f ail- ures. Through a structured re vie w , the paper emphasizes three main sections: the acquisition of battery data, techniques for estimating battery stat es using adv anced battery management systems, and battery R UL esti- mation methods. By scrutinizing commercially and publicly a v ailable datasets for v arious battery models and conditions, as well as e v aluating di v erse prognostic methods, the paper pro vides insights into the current land- scape of Li-ion battery research. The com parati v e analysis of R UL prediction algorithms and mathematical models, alongside the introduction of an AI-based cloud monitoring system, underscores the need for inno v a- ti v e approaches to enhance battery performance and safety in both industri al and lightweight applications. As researchers continue to e xplore and de v elop adv anced battery management techniques, this paper serv es as a v aluable resource for guiding future research directions and addressing the challenges associated with Li-ion battery technology . FUNDING INFORMA TION The authors declare that this research did not r ecei v e an y specic grant from public, commercial, or not-for -prot funding agencies. A UTHOR CONTRIB UTIONS ST A TEMENT This journal uses the C o nt rib utor Roles T axonomy (CRediT) to recognize indi vidual author contrib u- tions, reduce authorship disputes, and f acilitate collaboration. Name of A uthor C M So V a F o I R D O E V i Su P Fu Sthitprajna Mishra Chinmo y K umar P anigrahi Subhra Debdas Atri Bandyopadh yay Srikanth V elpula Amit K umar Sahoo P abitra K umar T ripath y C : C onceptualization I : I n v estig ation V i : V i sualization M : M ethodology R : R esources Su : Su pervision So : So ftw are D : D ata Curation P : P roject Administration V a : V a lidation O : Writing - O riginal Draft Fu : Fu nding Acquisition F o : F o rmal Analysis E : Writing - Re vie w & E diting CONFLICT OF INTEREST ST A TEMENT The authors declare that there are no kno wn nancial, personal, or professional conicts of i nterest that could ha v e inuenced the w ork reported in this paper . All contrib utions to the research and manuscript preparation were conducted impartial ly and independently . This declaration ensures transparenc y and upholds the inte grity of the research process. D A T A A V AILABILITY The data presented in this study are h ypothetical and were generated solely for the purpose ofconcep- tual analysis and methodological illustration. As such, no real-w orld datasets were used or made a v ailable. REFERENCES [1] J . Y un, Y . Choi, J. Lee, S. Choi, and C. Shin, “State-of-char ge estimation method for lithium-ion batteries using e xtended Kalman lter with adapti v e battery parameters, IEEE Access , v ol. 11, pp. 90901-90915, 2023, doi: 10.1109/A CCESS.2023.3305950. [2] P . J. Hall and E. J. Bain, “Ener gy-storage technologies and electricity generation, Ener gy Polic y , v ol. 36, no. 12, pp. 4352-4355, 2008, doi: 10.1016/j.enpol.2008.09.037. [3] A . K. Rohit, K. P . De vi, and S. Rangnekar , An o v ervie w of ener gy storage and its importance in Indian rene w able ener gy sector: P art I–technologies and comparison, Journal of Ener gy Storage , v ol. 13, pp. 10-23, 2017, doi: 10.1016/j.est.2017.06.005. Int J Appl Po wer Eng, V ol. 14, No. 3, September 2025: 569–578 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Appl Po wer Eng ISSN: 2252-8792 577 [4] A. Upadh yaya and C. Mahanta, An o v ervie w of battery based electric v ehicle technologies with emphasis on ener gy sources, their conguration topol ogies and management strate gies, IEEE T ransactions on Intelligent T ransportation Systems , v ol. 25, no. 2, pp. 1087-1111, Feb . 2024, doi: 10.1109/TITS.2023.3316191. [5] J. Gusta vsson, “Ener gy storage technology comparison: A kno wledge guide to simplify selection of ener gy storage technology , Ener gy T echnology , 2016. [6] R. R. K umar , C. Bharatiraja, K. Udhayakumar , S. De v akirubakaran, K. S. Sekar , and L. Mihet-Popa, Adv ances in batteries, battery modeling, battery management system, battery thermal management, SOC, SOH, and char ge/dischar ge characteristics in EV applications, IEEE Access , v ol. 11, pp. 105761-105809, 2023, doi: 10.1109/A CCESS.2023.3318121. [7] A. S. Abdelaal, S. Mukhopadh yay , and H. Rehman, “Battery ener gy management techniques for an electric v ehicle traction sys- tem, IEEE Access , v ol. 10, pp. 84015-84037, 2022, doi: 10.1109/A CCESS.2022.3195940. [8] K. Zhang, D. Bai, Y . Li, K. Song, B. Zheng, and F . Y ang, “Rob ust state-of-char ge estimator for lithium-ion batteries enabled by a ph ysics-dri v en dual-stage attention mechanism, Applied Ener gy , v ol. 359, Apr . 2024, doi: 10.1016/j.apener gy .2024.122666 [9] S. Singh, V . More, and R. Batheri, “Dri ving electric v ehicles into the future with battery management systems, IEEE Engineering Management Re vie w , v ol. 50, no. 3, pp. 157-161, Sep. 2022, doi: 10.1109/EMR.2022.3194655. [10] Y . Zhang, Q. T ang, Y . Zhang, J. W ang, U. Stimming, and A. A. Lee, “Identifying de gradation patterns of lithium ion batteries from impedance spectroscop y using machine learning, Nature Communications , 2020, doi: 10.1038/s41467-020-15235-7. [11] G. Dong, F . Y ang, Z. W ei, J. W ei, and K.-L. Tsui, “Data-dri v en battery health prognosis using adapti v e Bro wnian motion model, IEEE T ransactions on Industrial Informatics , v ol. 16, no. 7, pp. 4736-4746, Jul. 2020, doi: 10.1109/TII.2019.2948018. [12] R. Xiong, Y . Zhang, J. W ang, H. He, S. Peng, and M. Pecht, “Lithium-ion battery health prognosis based on a real battery manage- ment system used in electric v ehicles, IEEE T ransactions on V ehicular T echnology , v ol. 68, no. 5, pp. 4110-4121, May 2019, doi: 10.1109/TVT .2018.2864688. [13] K. Goebel, B. Saha, A. Sax ena, J. R. Celaya, and J. P . Christ ophersen, “Prognostics in battery health management, IEEE instru- mentation and Measurement Mag azine , v ol. 11, no. 4 pp. 33-40, Aug. 2008, doi: 10.1109/MIM.2008.4579269. [14] H. Meng and Y .-F . Li, A re vie w on prognostics and health management (PHM) methods of lithium-ion batteries, Rene w able and Sustainable Ener gy Re vie ws , v ol. 116 Dec. 2019, doi: 10.1016/j.rser .2019.109405. [15] C. L. Gan, “Prognostics and health management of electronics: fundamentals, machine learning, and the internet of things, Life Cycle Reliability and Safety Engineering , John W ile y & Sons. Ltd., pp. 225-226, 2022, doi: 10.1007/s41872-020-00119-y . [16] D. Andre, C. Appel, T . Soczka-Guth, and D. U. Sauer , Adv anced mathematical met hods of SOC and SOH estimation for lithium- ion batteries, Journal of Po wer Sources , v ol. 224, pp. 20-27, doi: 10.1016/j.jpo wsour .2012.10.001. [17] H. T ian, P . Qin, K. Li, and Z. Zhao, A re vie w of the state of health for lithium-ion batteries: Research status and suggestions, Jour - nal of Cleaner Production , v ol. 261, 2020, doi: 10.1016/j.jclepro.2020.120813. [18] X. Zhou, Z. P an, X. Han, L. Lu, and M. Ouyang, An easy-to-implement multi-point impedance technique for monitori ng aging of lithium-ion batteries, Journal of Po wer Sources , v ol. 417, pp. 188-192, 2019, doi: 10.1016/j.jpo wsour .2018.11.087. [19] G. K. Prasad and C. D. Rahn, “Model based identication of aging parameters in l ithium-ion batteries, Journal of Po wer Sources , v ol. 232, pp. 79-85, 2013, doi: 10.1016/j.jpo wsour .2013.01.041. [20] M. E. Orchard, P . He via-K och, B. Zhang, and L. T ang, “Risk measures for particle-ltering-based state-of-char ge progno- sis in lithium-ion batteries, IEEE T ransactions on Industrial Electronics , v ol. 60, no. 11, pp. 5260-5269, No v . 2013, doi: 10.1109/TIE.2012.2224079. [21] S. Mishra, C . K. P anigrahi, S. Debdas, D. V . P . V arma, and M. Y ada v , “IoT enabled battery status monitoring system for electric v ehicles, in 2023 IEEE 3rd International Conference on Sustainable Ener gy and Future Electric T ransportation (SEFET) , pp. 1-5, 2023, doi: 10.1109/SeFeT57834.2023.10245359. [22] B . Duan, Q. Zhang, F . Geng, and C. Zhang, “Rem aining useful life prediction of lithium-ion battery based on e xtended Kalman particle lter , International Journal of Ener gy Research , v ol. 44, no. 3, pp. 1724-1734 2019, doi: 10.1002/er .5002. [23] N. W illiard, W . He, M. Osterman, and M. Pecht, “Comparati v e analysis of features for determining state of health in lithium-ion batteries, International Journal of Prognostics and Health Management , v ol. 4, no. 1, 2013. [24] M. Berecibar , I. Gandiag a, I. V illarreal, N. Omar , J. V an Mierlo, and P . V an den Bossche, “Critical re vie w of state of health estimation methods of Li-ion batteries for real applications, Rene w able and Sustainable Ener gy Re vie ws , v ol. 56, pp. 572-587, 2016, doi: 10.1016/j.rser .2015.11.042. [25] S. W ang, R. Zhou, Y . Ren, M. Jiao, H. Liu, and C. Lian, Adv anced data-dri v en techniques in AI for predicting lithium-ion battery remaining useful life: a comprehensi v e re vie w , Green Chemical Engineering , 2024, doi: 10.1016/j.gce.2024.09.001. BIOGRAPHIES OF A UTHORS Sthitprajna Mishra holds a Bachelor of T echnology in Electrical and Electr onics Engi- neering from GIT A, BBSR, and a Master’ s de gree in Po wer Electronics and Dri v es from IGIT Sarang. He is currently pursuing his Ph.D. at KIIT in the area of IoT -based optimized smart-grid battery man- agement system (BMS), and also serv es as an IEEE chair member of the student branch at KIIT . Mr . Mishra’ s academic focus lies in the intersection of IoT technology and smart gr id optimization, aim- ing to contrib ute to adv ancements in ener gy mana gement and grid ef cienc y . He can be contacted at email: sthitprajnamishra26@gmail.com. AI-driven solutions for Li-ion battery performance and pr ediction (Sthitpr ajna Mishr a) Evaluation Warning : The document was created with Spire.PDF for Python.
578 ISSN: 2252-8792 Chinmoy K umar P anigrahi is a Professor and Director at KIIT DU’ s School of Electrical Engineering. His e xpertise includes soft computing, po wer systems, rene w able ener gy , and battery management systems. He has supervised 29 Ph.D. and 72 M.T ech. scholars, and guided four jointly . He has authored 182 research articles and presented 148 papers at conferences. He recei v ed se v eral accolades, including being rank ed among the T op 3 Ph.D. supervisors at KIIT (2022), and a w ards such as Outstanding Scientist (2020) and Best T eacher (2015). He is the Chair of the IEEE K olkata Section Industrial Electronics Society Chapter Bhubanesw ar and holds senior IEEE memberships. He has conducted collaborati v e research at the Uni v ersity of Shef eld and the Uni v ersity of Zurich (UZH), German y . He can be contacted at email: panigrahichinmo y@gmail.com. Subhra Debdas recei v ed his B.E. in Electrical Engineering and M.E. in Po wer Syst em En- gineering f rom Indian Institute of Engineering Science and T echnology , Shibpur , and his Ph.D. from Sainath Uni v ersity , Ranchi . Extensi v e design po wer engineer e xperience at DCPL and L and T Sar - gent and Lundy , managing impactful national and inte rnational projects. Ov er 21 years of teaching e xperience, including 8 years at Uni v ersity of T echnology and Applied Sciences in Nizw a, Sultanate of Oman. No w full-time f aculty at KIIT Deemed Uni v ersity’ s School of Electrical Engineering. His academic interests is in rene w able ener gy , smart grid technologies, Industry 4.0, IoT , cloud comput- ing, and focusing on pract ical applications. He can be contacted at email: subhra.debdas@gmail.com. Atri Bandy opadh yay is a dynamic computer scientist, systems engineer from Purulia, W est Beng al. Inno v ator in AI, deep learning, cryptograph y through impactful internships at High- Radius and DRDO. Kaling a Institute of Industrial T echnology graduate, e xcelling in projects lik e EmoSense, dyna mic train price prediction. St ellar academic record, numerous certications, publi- cations in IEEE and Springer . T railblazer in machine learning and IoT . UiP ath Stude nt De v eloper Champion, accolades at DRDO ITR. Leading research, inno v ation, and shaping future of technology . He can be contacted at email: atricc03@gmail.com. Srikanth V elpula recei v ed the B.T ech. and M.T ech. de grees from Ja w aharlal Nehru T echnological Uni v ersity , Hyderabad, India, in 2009 and 2012, respecti v ely . He completed his Ph.D. de gree in the year 2020 at V ellore Institute of T echnology , V ellore, T amilnadu, India. He w ork ed as an Assistant Professor in the Department of Electrical and Electronics Engineering at v arious engineering colle ges in India during 2011-2022. Currently , he is w orking as Assistant Professor at SR Uni v ersity , W arang al, T elang ana, India. His research interests include con v erter controls, DFIG based systems, electrical v ehicle dri v es and battery management system, and the inte gration of rene w able ener gies into the po wer systems. He can be contacted at email: srikv elpula@gmail.com. Amit K u mar Sahoo recei v ed his Ph.D. from Birla Institute of T echnology , Mesra, India in Control System and Master’ s de gree from National Institute of T echnology , Rourk ela, India. He is presently w orking as an Associate Professor in the Department of Electrical and Electronics Engi- neering, Centurion Uni v ersity of T echnology and Management, Odisha, India. He has more than 13 years of teaching e xperience. He is a life member of IEI, India. His spec ializations include system identication, linear and non-linear control system, control and automation, inte gra l and fractional order controller design, soft and e v olutionary computing, and machine learning. He can be contacted at email: amitkumar2687@gmail.com. P abitra K umar T ripath y is a Professor at Kalam Institute of T echnology , Berhampur , af liated with Biju P atnaik Uni v ersity of T echnology , Odisha, specializes in Machine Learning and E-Commerce. He holds a M aster’ s de gree in Mathematics from Berhampur Uni v ersity , an M.T ech. in Computer Science, and a Ph.D. from Kaling a Uni v ersity , Ra ipur . His e xpertise includes theory of computations, compiler design, cryptograph y , and computational number theory . He’ s authored tw o books published by CRC and W ile y . He can be contacted at email: pabitratripath y81@gmail.com. Int J Appl Po wer Eng, V ol. 14, No. 3, September 2025: 569–578 Evaluation Warning : The document was created with Spire.PDF for Python.