Shivanshu Shekhar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Shivanshu Shekhar has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Gajendra Singh Chawda
Department of Electrical & Computer Engineering, Tuskegee University, Tuskegee, AL 36088, USA
2 shared publications - Wencong Su
University of Michigan–Dearborn
2 shared publications Ben Oni
Department of Electrical and Computer Engineering (ECE) Tuskegee University Tuskegee Alabama USA
1 shared publication- Ben Oni
Department of Electrical & Computer Engineering, Tuskegee University, Tuskegee, AL 36088, USA
1 shared publication Carlos Hernández
1 shared publication- Guilherme Vieira Hollweg
Department of Electrical and Computer Engineering (ECE) University of Michigan‐Dearborn Dearborn Michigan USA
1 shared publication
Latest Publications
- 8 November 2026
- Published by Wiley in Energy Storage
- Vol. 8 (8)
A Monotone Rational–Quadratic Spline Driven High‐Fidelity Open Circuit Voltage Modeling for Lithium‐Ion Batteries
Reliable state estimation of Li‐ion batteries in energy applications depends largely on accurate modeling of the open‐circuit voltage (OCV), denoted by V oc , as a function of state of charge (SoC). The V oc –SoC characteristic is conventionally represented using high‐order polynomials or lookup tables stored in the battery management system (BMS). High‐order polynomial models may suffer from numerical issues and generally require numerical root‐finding for inverse mapping, whereas lookup‐table approaches may require a dense set of breakpoints, leading to increased memory usage. To address these limitations, this article proposes a monotone rational–quadratic spline (RQS) framework for V oc –SoC modeling. A greedy knot‐selection algorithm is incorporated into the RQS framework to adaptively select breakpoints along the V oc –SoC characteristic. The proposed approach is validated on test data from six Li‐ion cells with different chemistries, collected at multiple operating temperatures. The proposed method provides a smooth and monotonic representation with analytical invertibility and low computational complexity compared to conventional models. The efficient pseudo‐ V oc ‐based SoC reconstruction suggests the proposed framework's potential suitability for integration into real‐time BMS applications.
- 8 November 2026
- Published by Wiley in Energy Storage
- Vol. 8 (8)
A Monotone Rational–Quadratic Spline Driven High‐Fidelity Open Circuit Voltage Modeling for Lithium‐Ion Batteries
Reliable state estimation of Li‐ion batteries in energy applications depends largely on accurate modeling of the open‐circuit voltage (OCV), denoted by V oc , as a function of state of charge (SoC). The V oc –SoC characteristic is conventionally represented using high‐order polynomials or lookup tables stored in the battery management system (BMS). High‐order polynomial models may suffer from numerical issues and generally require numerical root‐finding for inverse mapping, whereas lookup‐table approaches may require a dense set of breakpoints, leading to increased memory usage. To address these limitations, this article proposes a monotone rational–quadratic spline (RQS) framework for V oc –SoC modeling. A greedy knot‐selection algorithm is incorporated into the RQS framework to adaptively select breakpoints along the V oc –SoC characteristic. The proposed approach is validated on test data from six Li‐ion cells with different chemistries, collected at multiple operating temperatures. The proposed method provides a smooth and monotonic representation with analytical invertibility and low computational complexity compared to conventional models. The efficient pseudo‐ V oc ‐based SoC reconstruction suggests the proposed framework's potential suitability for integration into real‐time BMS applications.
- Open Access
Critical Review of Artificial Intelligence and Machine Learning Methods for Data-Center Load Forecasting
Modern data centers are consuming more energy than ever before due to the rapid growth of cloud services, artificial intelligence (AI), and large-scale digital applications. As energy demand continues to rise, accurate load forecasting has become an important tool for improving energy management and operational planning. However, predicting data-center power demand remains challenging because computing workloads, cooling systems, and facility operations are closely connected and constantly changing. This review examines the use of artificial intelligence (AI) and machine learning (ML) techniques for data-center load forecasting. The surveyed literature covers traditional statistical methods, supervised learning algorithms, deep learning models, probabilistic forecasting techniques, and physics-informed hybrid approaches. Important topics such as forecasting horizons, feature selection, performance evaluation, and practical deployment challenges are also discussed. The reviewed studies indicate that advanced approaches, including long short-term memory (LSTM), gated recurrent unit (GRU), transformer-based models, and digital twin (DT) frameworks, can improve forecasting performance and support more energy-efficient operation. These methods can also assist carbon-aware and grid-interactive data-center management. Several challenges remain. These include limited data availability, poor generalization, interpretability issues, and real-time implementation constraints. In this paper, the reviewed studies are classified and analyzed to provide a clear reference for researchers working in AI-based data-center energy forecasting.
Open AccessCritical Review of Artificial Intelligence and Machine Learning Methods for Data-Center Load Forecasting
Modern data centers are consuming more energy than ever before due to the rapid growth of cloud services, artificial intelligence (AI), and large-scale digital applications. As energy demand continues to rise, accurate load forecasting has become an important tool for improving energy management and operational planning. However, predicting data-center power demand remains challenging because computing workloads, cooling systems, and facility operations are closely connected and constantly changing. This review examines the use of artificial intelligence (AI) and machine learning (ML) techniques for data-center load forecasting. The surveyed literature covers traditional statistical methods, supervised learning algorithms, deep learning models, probabilistic forecasting techniques, and physics-informed hybrid approaches. Important topics such as forecasting horizons, feature selection, performance evaluation, and practical deployment challenges are also discussed. The reviewed studies indicate that advanced approaches, including long short-term memory (LSTM), gated recurrent unit (GRU), transformer-based models, and digital twin (DT) frameworks, can improve forecasting performance and support more energy-efficient operation. These methods can also assist carbon-aware and grid-interactive data-center management. Several challenges remain. These include limited data availability, poor generalization, interpretability issues, and real-time implementation constraints. In this paper, the reviewed studies are classified and analyzed to provide a clear reference for researchers working in AI-based data-center energy forecasting.
- Open Access
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Smart Grid
- Vol. 9 (1)
Artificial Intelligence in Bidirectional Energy Management: A Critical Review of Control Strategies for V2G and G2V Systems
The growing adoption of electric vehicles (EVs) and renewable energy sources has accelerated the need for intelligent bidirectional energy management systems. Vehicle‐to‐Grid (V2G) and Grid‐to‐Vehicle (G2V) technologies offer promising solutions to enhance grid stability, optimise energy use and leverage EVs as mobile storage units. However, real‐time control of these systems remains challenging due to the dynamic nature of power flows, user behaviour and energy demand uncertainties. This review explores how Artificial Intelligence (AI) and Machine Learning (ML) methods–such as supervised learning, reinforcement learning, deep learning and hybrid models–address these challenges. This paper highlight key applications, recent case studies, and experimental advances in AI‐driven V2G and G2V control, emphasising improvements in system efficiency, battery health and grid reliability. Critical challenges including data quality, real‐time computation, and cybersecurity are discussed, along with future directions such as Explainable AI, blockchain integration, and lightweight decentralised models. This review aims to present a critical overview of AI‐enabled control strategies for V2G and G2V systems and to guide future innovations in sustainable and intelligent energy management.
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Smart Grid
- Vol. 9 (1)
Open AccessArtificial Intelligence in Bidirectional Energy Management: A Critical Review of Control Strategies for V2G and G2V Systems
The growing adoption of electric vehicles (EVs) and renewable energy sources has accelerated the need for intelligent bidirectional energy management systems. Vehicle‐to‐Grid (V2G) and Grid‐to‐Vehicle (G2V) technologies offer promising solutions to enhance grid stability, optimise energy use and leverage EVs as mobile storage units. However, real‐time control of these systems remains challenging due to the dynamic nature of power flows, user behaviour and energy demand uncertainties. This review explores how Artificial Intelligence (AI) and Machine Learning (ML) methods–such as supervised learning, reinforcement learning, deep learning and hybrid models–address these challenges. This paper highlight key applications, recent case studies, and experimental advances in AI‐driven V2G and G2V control, emphasising improvements in system efficiency, battery health and grid reliability. Critical challenges including data quality, real‐time computation, and cybersecurity are discussed, along with future directions such as Explainable AI, blockchain integration, and lightweight decentralised models. This review aims to present a critical overview of AI‐enabled control strategies for V2G and G2V systems and to guide future innovations in sustainable and intelligent energy management.
- 29 September 2025
- Published by Elsevier in Journal of Power Sources
- Vol. 660
Effect of variable temperatures on machine learning battery SoH estimation for auxiliary aircraft batteries
- 29 September 2025
- Published by Elsevier in Journal of Power Sources
- Vol. 660
Effect of variable temperatures on machine learning battery SoH estimation for auxiliary aircraft batteries