S. AKINBOHUN

State of Charge (SoC) estimation plays a crucial role in battery management systems (BMS), directly impacting the performance, safety, and longevity of lithium-ion batteries. This study presents a comparative review of three major categories of SoC estimation techniques: model-based, data-driven, and hybrid methods. The review is driven by the need to evaluate the accuracy, robustness, and practical applicability of these methods across various real-world conditions, including different temperature profiles, battery chemistries, and aging states. The research methodology involved a structured literature search, selection of 45 peerreviewed studies published between 2018 and 2025, and systematic data extraction. Model-based approaches, particularly those using Kalman filters and equivalent circuit models, demonstrated computational efficiency but showed sensitivity to parameter drift and aging. Data-driven techniques, including LSTM networks, Gaussian Process Regression (GPR), and Random Forests, offered high accuracy—often achieving <2% RMSE—but required large, diverse datasets. Hybrid methods, such as AEKF-LSTM and UKF-PSO-LSTM models, consistently achieved the highest accuracy (RMSE <1%) while balancing robustness and adaptability. The findings suggest that while model-based methods are suitable for resourceconstrained systems, hybrid approaches offer the most promising results in terms of overall performance and reliability. These insights can guide future BMS development
and inform system-level design choices in electric vehicle and energy storage applications.

Lithium-ion batteries presence today as cornerstone power solutions for all scales of energy management from consumer electronic devices to electric vehicles and renew able systems. The high energy density together with long life cycle of lithium-ion batteries exists with reliability and safety risks due to thermal runaway and internal short circuits and capacity degradation faults. A new generative deep learning framework emerged for lithium-ion battery fault diagnosis through advanced machine learning implementations which improved anomaly detection alongside predictive fault analysis capabilities. Reliability measurements in batteries could be assessed through a method which integrates Graph Neural Networks (GNNs) and attention mechanisms plus spiral correlation detection for analysing complex non-linear characteristics. The model required data processing and temporal sequence generation as well as feature graph construction steps to ensure accuracy and resistance to external conditions. The system utilized synthetic and real data sets to reach an accuracy level of 95.6% and achieved 92.8% recall performance together with 0.98 AUC- ROC score. These metrics show that the framework successfully detects minor anomalies in addition to making exact judgments about normal and faulty battery states. Analysis demonstrated that GNNs work well for parameter relationship modelling and attention methods direct the model toward significant features like temperature and voltage variations. Spiral correlation detection as a method delivered ground breaking understanding of nonlinear fault patterns especially in cases of capacity fade and electrode degradation. These study findings show how integrating deep learning methods with specialized domain understanding allows the resolution of essential battery diagnostic problems.

This article presents a thorough examination and assessment of a 5.5 kVA solar power inverter, focusing on its performance, efficiency, and reliability under various operational circumstances. The introduction emphasizes the significance of solar power inverters in renewable energy systems and underscores the necessity for comprehensive analysis and testing to gauge their effectiveness and dependability. The experimental approach details the setup of the testing environment, incorporating a simulated PV array, a programmable electronic load, and a comprehensive data collection system. Parameters such as input and output voltage, current, and power, inverter efficiency, total harmonic distortion, and operating temperature are meticulously measured during the assessment. Testing covers a spectrum of input voltage and output power levels, as well as diverse environmental conditions to mimic real-world usage scenarios. The findings reveal that the 5.5 kVA inverter consistently maintains high efficiency levels, typically exceeding 95%, across a broad range of operating conditions. It also demonstrates robust power handling capabilities, delivering the rated 5.5 kVA of power with minimal distortion in the output waveform. The inverter proves resilient to fluctuations in input voltage and output power, rendering it suitable for both grid-connected and off-grid solar energy applications. The discussion delves into a detailed analysis of the results, spotlighting the critical factors influencing the inverter's performance
and reliability, along with their implications for its deployment and operation.