The effectiveness of a battery management system (BMS) in lithium-ion batteries (LIBs) is significantly dependent on the accuracy of battery sensors. However, owing to the highly nonlinear nature of LIBs, detecting small uncertainties in sensor measurements, which can lead to high estimation errors, poses a remarkable challenge. Moreover, in conventional BMS, sensor uncertainty detection and state estimation are treated separately which greatly reduces its effectiveness. Hence, this article presents an integrated framework for fast bias detection and state estimation to ensure reliable and safe operations. First, extensive experimentation is performed to statistically analyze the effects of different sensor uncertainties on multiple chemistries of LIBs. Second, a fast bias detection algorithm is proposed to identify small sensor offsets effectively. Third, upon successful bias detection, a hybrid extended nonlinear observer is proposed for simultaneously estimating all states, in which bias is treated as an augmented state. Lastly, the observability of the proposed design is investigated across various practical scenarios. Validation results, using 58 Ah NMC and 25 Ah LiFePO4 LIB cells, underscore the potential of this integrated strategy to enhance the safety of LIBs in commercial applications.

The reliability, lower computational complexity, and ease of implementation of control observers make them one of the most promising methods for the state estimation of Li-ion batteries (LIBs) in commercial applications. To pave their way, this study performs a comprehensive and systematic evaluation of four main categories of control observer-based methods in different practical scenarios considering estimation accuracy, computational time convergence speed, stability, and robustness against measurement uncertainties. Observers are designed using a second-order equivalent circuit model whose observability against different scenarios is rigorously investigated to verify the feasibility of the proposed analysis. Established techniques then are validated against driving datasets and their comparative usefulness is evaluated using an experimental setup. The analysis also evaluates the adaptability of different techniques to electric vehicle (field data. The results indicate better accuracy, stability, robustness, and faster convergence for the PI and PID, while the estimations of the Luenberger observers find it hard to converge against highly dynamic loadfiles. Moreover, this study also discusses the sensitivity of observer-based techniques to battery ohmic polarization and voltage-related measurement uncertainties. The most remarkable contribution of the proposed study lies in providing guidance for researchers when choosing the control observers for online state estimation of LIBs.

An accurate and robust state estimation of Lithium-ion batteries (LIBs) is crucial for safe and reliable operations. However, due to the modeling uncertainties and complex nonlinear behavior of LIBs in practical scenarios, existing methods fell short of the desired accuracy and adaptability. This article anchors a proportional integral derivative (PID) observer with deep reinforcement learning (DRL) to develop a battery state estimation framework with enhanced performance, stability, and adaptability. First, the PID observer is constructed using a second-order battery equivalent circuit model, and an actor-critic DRL agent is then integrated with the designed observer to learn the optimal policy for observer control. A dynamic reward function with adaptive constraints is formulated to avoid unproductive explorations and aid efficient learning and convergence. In addition, an augmented model bias is also introduced into the battery model to compensate for the model inaccuracies in real time. After successful training, the trained DRL agent, based on the learned optimal policy, auto-adjusts the PID control according to operating conditions, making the proposed method highly adaptive. Finally, a test bench is designed to verify its improved accuracy, enhanced robustness against uncertainties, faster convergence, and comparative usefulness. Moreover, the high adaptability against real electric vehicle data highlights the potential of the proposed framework in improving the safety of LIBs in practical applications.

We propose an integrated design for the combined estimation of state-modified cell core temperature, temperature-corrected state of charge (SOC), and state of energy (SOE) of lithium-ion batteries (LIBs). First, a control-oriented integrated electrothermal coupled model is constructed. The observability of the proposed MIMO system is rigorously investigated, and an adaptive PI observer is formulated for the combined estimation task. Based on the proposed model, the electrical parameters are identified using improved experimental techniques, while the thermal model parameters are estimated online. Finally, the practicability of the proposed technique is verified using a test bench. The experimental results show the RMSE<0.5% for SOC and SOE estimation, and the RMSE<0.25°C for core temperature estimation. Moreover, the comparative results demonstrate high robustness against incorrect initial values, high-current dynamic loadfiles, and critical operating temperatures, which highlights the underscores of the proposed framework in improving the safety of LIBs in practical applications.

Large format Li-ion batteries (LIBs) are getting huge attention due to higher density, efficient packing, and lower system and design costs. However, due to internal inhomogeneities, they exhibit a strong intermittent behavior against critical operating conditions, making accurate state estimation quite challenging. Therefore, this article presents a novel framework with augmented model bias and intelligent derivative control to ensure the desired accuracy and robustness in state estimation of large format cells. First, an augmented bias is proposed in the LIB equivalent circuit model to compensate for the modeling uncertainties and improve accuracy. Second, observability of the proposed system is investigated rigorously. Third, a proportional integral derivative observer with an intelligent multi-layer error convergence mechanism is proposed to enhance convergence while maintaining stable estimations. The superiority of the proposed framework against critical operations is verified using an experimental setup. Moreover, the comparative results against benchmark algorithms demonstrate improved accuracy, high adaptability, enhanced robustness against measurement uncertainties and disturbances, low computational cost, and five times faster convergence, highlighting the proposed framework's potential to improve the safety of LIBs in practical applications.

An accurate estimation of battery state of charge (SOC) is crucial to ensure the safety of Li-ion batteries (LIBs), and the recent boom in machine learning (ML) has illustrated a strong potential for SOC estimation. However, the ignorance of LIB internal dynamics greatly limits the performance of ML-based methods. Therefore, this work incorporates multiple battery internal dynamics into the ML-based method to ensure enhanced performance. First, using a simple mechanism based on the LIB model, the open circuit voltage, ohmic response, and temperature-corrected pattern are extracted from the measured data and provided as inputs to the neural networks (NNs). Then, to ensure time adaptability and accuracy, the predictions of the proposed design are adjusted using short-term Ampere-hour results. The proposed technique is validated using different real-world driving profiles. The comparative results with benchmark NNs-based methods verify a >60% reduction in root mean square error, demonstrating the effectiveness of incorporating battery internal dynamics into NNs.