Road transport significantly contributes to greenhouse gas emissions in all places where it is used and therefore also in Europe, prompting the EU to set ambitious objectives for CO2 reduction. In order to reach these objectives, the automotive industry is transitioning to electric vehicles, utilizing electric powertrains powered by battery packs. However, the longevity and reliability of these batteries are critical concerns. This review paper focuses on the advanced diagnostic techniques for effective battery State of Charge (SoC) and State of Health (SoH) monitoring. Accurate SoC/SoH estimation is crucial for optimizing battery performance, avoiding premature degradation, and ensuring driver safety. By investigating these areas, this paper aims to contribute to the development of more sustainable and durable electric vehicles, supporting the transition to cleaner transportation systems.
This paper introduces a data-driven methodology for accurately estimating the residual capacity (RC) of lithium iron phosphate (LFP) batteries through a tailored artificial neural network (ANN) architecture. The proposed model integrates a long short-term memory (LSTM) layer with a fully connected layer, leveraging their combined strengths to achieve precise RC predictions. A distinguishing feature of this study is its ability to deliver highly accurate estimates using a limited dataset that was derived from a single cylindrical LFP battery with a 40 Ah capacity and collected during a controlled experimental campaign. Despite the constraints imposed by the dataset size, the ANN demonstrates remarkable performance, underscoring the model’s capability to operate effectively with minimal data. The dataset is partitioned into the training and testing subsets to ensure a rigorous evaluation. Additionally, the robustness of the approach is validated by testing the trained ANN on data from a second battery cell subjected to a distinct aging process, which was entirely unseen during training. This critical aspect underscores the method’s applicability in estimating RC for batteries with varying aging profiles, a key requirement for real-world deployment. The proposed LSTM-based architecture was also benchmarked against a GRU-based model, yielding significantly lower prediction errors. Furthermore, beyond LFP chemistry, the method was tested on a broader NMC dataset comprising seven cells aged under different C-rates and temperatures, where it maintained high accuracy, confirming its scalability and robustness across chemistries and usage conditions. These results advance battery management systems by offering a robust, efficient modeling framework that optimizes battery utilization across diverse applications, even under data-constrained conditions.
Lithium-ion batteries (LiBs) undergo degradation influenced by storage and cycling conditions. Accurate state of health (SOH) assessment is crucial for predicting battery aging, which is generally marked by a decline in capacity (energy fade) or an increase in internal resistance (power fade). This study investigates the impulse response (IR) technique for assessing the SOH of lithium cobalt oxide batteries, addressing both capacity fade and rising internal resistance. The IR method relies on a predefined dataset that records the voltage response of the LiB to pulse current inputs across various states of charge (SOC), temperatures, and aging conditions to train a series of linear auto-regressive exogenous (ARX) models. This dataset is then used as a look-up table for subsequent SOH estimation under new operating conditions. The results demonstrate that the method can capture trends in capacity fade and resistance increase only when multiple battery temperatures are incorporated into the look-up table. In contrast, estimations based on ARX models trained at a single fixed temperature fail to provide reliable predictions of battery SOH.
2 April 2026 in IEEE Transactions on Instrumentation and Measurement
Resolvers are widely used in safety-critical and harsh environments due to their robustness against temperature, vibration, and electromagnetic interference. Conventional resolver-to-digital converters (RDCs) often rely on monolithic integrated circuits (ICs) integrating excitation, demodulation, and tracking loops, which add cost and board footprint while limiting flexibility. This paper presents a fully embedded RDC that eliminates dedicated RDC ICs by exploiting PWM and ADC resources of modern microcontrollers. The method combines: (i) a GaN half-bridge resonant excitation engineered to align resolver sine/cosine peaks with ADC sample-and-hold instants; (ii) differential 16-bit ADC acquisition with hardware PWM–ADC synchronization at sub-10-ns resolution; and (iii) a software phase-locked loop (PLL) with quality-driven autotuning that minimizes the ripple of a monitoring function to optimize sampling timing, and performs automatic gain/offset calibration. Experiments on a 1.1-kW PMSM test bench comparing the proposed method with a commercial RDC-IC channel show comparable or superior position/speed accuracy (ripple < 0.5%), finer velocity quantization, robust tracking during acceleration, and significant reductions in cost and PCB area. The results indicate that excitation engineering and metric-based autotuning are effective enablers for next-generation embedded resolver interfaces in electric drives.