Feature Evaluation in Machine Learning-Based Soc Estimation for PV Battery Systems

This paper proposes a novel grid-connected Solar Photovoltaic battery energy storage system (PV-BESS) architecture with the support of both DC loads and grid connection. The system combines photovoltaic panels, DCDC converters with Maximum Power Point Tracking (MPPT), Battery Energy Storage System (BESS) based on an intelligent Battery Management System (BMS), and bidirectional converters with a net metering policy. The real-time demand management further facilitates the Smart Energy Management System (SEMS,) which optimizes an optimum dispatching mechanism for energy flow inside constituent components of the system. Simulation results confirm efficient PV generation and battery recharging, accurate State of Charge (SOC) control, and steady synchronization with the inverter and the grid. The estimated power efficiency of the DCconnected system is $10-15 \%$ higher than that of the conventional AC-coupled systems. The future is envisioned with AI-based forecasting, advanced storage technology, and integration with other sectors due to the system’s scalability and resilience. The PV-BESS dual system is both economical and eco-friendly to future generations, as per the research acknowledgments.

Battery Monitoring is very important for most of the electric vehicle(EV) and battery energy storage system (BESS) since the safety, operation and even the life of the passenger/operator depends on the battery system. Checking and controlling the status of battery within their specified safe operating conditions is exactly the major function of battery management system (BMS). The state of charge (SOC) is a critical parameter of a Li-ion battery, an accurate on-line estimation of the SOC is important for forecasting the EV driving range and BESS power dispatching. A widely used method to estimate SOC is based on coulomb counting, due to the uncertainty, including unit-to-unit variation, measurement noise, operational uncertainties, and model inaccuracy, it's difficult to estimate the SOC by using coulomb counting, an online inference of open-circuit voltage (OCV) is used to calibrate the SOC, however, a relatively flat OCV curve of LiFePO4 will lead to the error of SOC is large. In this paper, an algorithm using the derivation curve (dV/dQ vs Q) to calibrate the SOC on-line is bring out. The SOC estimation has been implemented using coulomb counting and derivation curve methods thereby eliminating the limitation of the stand-alone coulomb counting method. Finally, by utilizing the actual operation data, experiments and numerical table were conducted, show that this SOC calibrate algorithm based on characteristic curve has better robust performance of the practical application of LiFePO4 battery energy storage system.

An optimized long short-term memory-weighted fading extended Kalman filtering model with wide temperature adaptation for the state of charge estimation of lithium-ion batteries

There have been strong drivers toward lessening diesel dependency in remote communities. These off-grid communities can be difficult to access, and often face challenges with diesel supply. More importantly, governments are striving to improve the environmental outlook in these areas. Tying this need with associated high operational costs and fuel handling challenges, alternative sources of energy and energy storage have emerged as a viable option to reduce diesel consumption. This paper presents the experience of a Canadian electric utility in planning the application of utility scale photovoltaic generation and Battery Energy Storage Systems to supplement existing diesel generators in a remote isolated community in Alberta. Economic drivers and opportunities to partner with first nations have sparked stakeholder attention and positioned the project as a national interest. Technical and economic studies were conducted to address all aspects of the project. These include forecasting load, sizing and siting each alternative generation option, customizing dispatch strategies, and determining required spinning reserve from diesel generators through time-domain studies. This paper covers the size and location optimization of the PV and BESS assets and dispatch strategies.

With the widespread application of electric vehicles and electrical energy storage systems, the accurate monitoring of lithium battery states has become crucial for ensuring safety and improving efficiency in terms of the applications. For this reason, this study proposes an algorithm focusing on Bayesian optimization-based adaptive extended Kalman filter (BO-AEKF) to enhance the numerical accuracy and stability of state-of-charge (SOC) estimation for lithium batteries under various operating conditions. By comparing with traditional methods, the proposed algorithm, introducing a temperature-adaptive mechanism and a dynamic parameter updating strategy, can effectively address the estimation limitations under severe temperature variations and initial SOC uncertainties. Experimental results demonstrate that the proposed algorithm exhibits superior estimation performance at different temperatures, including −10 °C, 0 °C, 25 °C, and 50 °C; particularly under dynamic operating conditions, the maximum error (MAX) and root mean square error (RMSE) are reduced by 51.9% and 74.5%, respectively, compared to the extended Kalman filter (EKF) and adaptive extended Kalman filter (AEKF) algorithms. Furthermore, the BO-AEKF achieves rapid convergence even with unknown initial SOC values, demonstrating its robustness and adaptability. These findings provide more reliable technical support for the development of battery management systems, making them suitable for state estimation in electric vehicles and renewable energy storage systems.

Characterization and optimization of total solids content of drilling fluids is critical for the efficiency and success of drilling operations. Traditional solids content analysis methods, such as retort analysis, require substantial human intervention and time, which can lead to inaccuracies, time-management issues, and increased operational risks. To address these issues and complement automated rheological property measurements during drilling, this study aims to develop and validate a machine learning-based framework for estimating solids content in drilling fluids from readily available rheological parameters. A comprehensive data set was compiled from more than 1600 laboratory reports of drilling fluid analyses across 130 oil wells globally. Text mining packages were used to convert laboratory reports into a numeric data set which consists of drilling fluid properties. Pre-processing steps were taken to clean, filter and reduce the dimensionality of the data. Unrealistic measurements, and missing data were filtered out and several feature-selection approaches such as least absolute shrinkage and selection operator (LASSO) regression, permutation feature importance, and correlation coefficient matrix were employed to remove redundant variables from the data set. Cleaned and dimensionality-optimized data set was then used for machine-learning model development. Due to their ability to capture non-linear relationships among multiple variables, artificial neural networks employing the resilient backpropagation algorithm were selected to fit regression models for estimating total solids content. Various configurations were tested, including different numbers of layers and neurons, as well as alternative data partitioning schemes. Considering reasonable values of rheological properties based on domain knowledge, filtering was applied to remove outliers which indicate experimental error. As a result of combining the observations from different feature importance analyses and considering physical relationships between rheological variables, redundant features as well as reports with many missing values were removed from the data set. Five features, including mud weight, chloride content, plastic viscosity, oil-water ratio and API fluid loss, were selected to carry over into the modeling stage. As a result of experimenting with 120 different neural network configurations, a two-hidden layer neural network with 40-30 neurons in each hidden layer was selected. The model achieved an R2 value of 0.96 for the training set and 0.89 for the testing set. The root-mean-square error (RMSE) was 0.87% for training and 1.27% for testing, both of which were considered acceptable levels of error. An analysis of the network weights indicated that mud weight (MW) and oil–water ratio (OWR) were the most influential features, contributing more significantly to the model’s predictions compared to other input variables.

Fractional-order models (FOMs) have been recognized as superior tools for capturing the complex electrochemical dynamics of lithium-ion batteries, outperforming integer-order models in accuracy, robustness, and adaptability. Parameter identification (PI) is essential for FOMs, as its accuracy directly affects the model’s ability to predict battery behavior and estimate critical states such as state of charge (SOC) and state of health (SOH). In this study, a hybrid deep learning approach has been introduced for FOM PI, representing the first application of deep learning in this domain. A simulation platform was developed to generate datasets using Sobol and Monte Carlo sampling methods. Five deep learning models were constructed: long short-term memory (LSTM), gated recurrent unit (GRU), one-dimensional convolutional neural network (1DCNN), and hybrid models combining 1DCNN with LSTM and GRU. Hyperparameters were optimized using Optuna, and enhancements such as Huber loss for robustness to outliers, stochastic weight averaging, and exponential moving average for training stability were incorporated. The primary contribution lies in the hybrid architectures, which integrate convolutional feature extraction with recurrent temporal modeling, outperforming standalone models. On a test set of 1000 samples, the improved 1DCNN + GRU model achieved an overall root mean square error (RMSE) of 0.2223 and a mean absolute percentage error (MAPE) of 0.27%, representing nearly a 50% reduction in RMSE compared to its baseline. This performance surpasses that of the improved LSTM model, which yielded a MAPE of 9.50%, as evidenced by tighter scatter plot alignments along the diagonal and reduced error dispersion in box plots. Terminal voltage prediction was validated with an average RMSE of 0.002059 and mean absolute error (MAE) of 0.001387, demonstrating high-fidelity dynamic reconstruction. By advancing data-driven PI, this framework is well-positioned to enable real-time integration into battery management systems.

• Seasonally adaptive PV–BESS dispatch framework for high EV penetration grids. • CNN–BiLSTM model achieves >92 % accuracy in short-term PV forecasting. • GA optimizes multi-objective PV-EV-BESS scheduling. • Reduces feeder losses by 63.55 % and Voltage Unbalance Factor by 55 %. • Enhances PV utilization, power quality, and grid resilience under stress. The rapid proliferation of photovoltaic (PV) systems and electric vehicles (EVs) poses significant challenges for distribution networks, including voltage unbalance, high feeder losses, and deteriorating power quality, further complicated by seasonal variability in solar output and demand. This paper proposes a seasonally adaptive PV–battery energy storage system (BESS) dispatch strategy that integrates short-term PV forecasting with multi-objective optimization. The primary novelty of this work lies in the tight integration of a seasonally adaptive, forecasting-driven dispatch strategy that synergistically combines a hybrid CNN-BiLSTM model for high-accuracy PV prediction with a Genetic Algorithm for multi-objective optimization in unbalanced distribution grids. A hybrid CNN–BiLSTM model, trained on distinct seasonal datasets (summer and winter) from NASA irradiance data, enables accurate real-time PV prediction, achieving an RMSE of 13.99 kWh, MAE of 12.94 kWh, and MAPE of 7.45 %, corresponding to a forecasting accuracy of 92.55 %. Meanwhile, a genetic algorithm (GA) optimizes coordinated PV–BESS scheduling under diverse seasonal scenarios. The approach is validated on a modified IEEE 13-bus system, demonstrating robustness across a wide range of EV penetration levels (40–70 %) and both summer and winter conditions, and showing reductions of up to 55 % in Voltage Unbalance Factor (VUF) and 63.55 % in feeder losses, alongside enhanced PV utilization. Compared to uncoordinated dispatch, the proposed framework delivers superior power quality, higher grid resilience, and practical scalability. The framework is designed for real-world application, with the forecasting model providing high-frequency predictions and the GA-based optimization operating on a computationally feasible hourly rolling horizon, making it suitable for deployment in larger distribution networks. These findings highlight the potential of forecast-driven, seasonally aware coordinated dispatch as a practical pathway toward reliable and sustainable smart grid operation.

This paper presents a modern yearly based park-and-ride management scheme. The electric vehicles’ owners are encouraged to keep their cars away from the crowded areas in cities and use the public facilities such as bus, train, and metro. This action will help the owners to reach their work on time inside these crowded cities. Electric vehicle charging stations are designed to charge 1000 electric vehicles using the proposed park-and-ride control approach. A case study of Fuxin, China is considered. The electric vehicle charging stations demand is met using renewable energy sources, namely photovoltaic and battery energy storage systems. Meeting the load demand and minimizing the total life cycle cost are considered two objective functions to formulate a multiobjective approach. The optimal sizes of the photovoltaic and battery energy storage systems are obtained using a multiobjective genetic algorithm and ε-MOGA. The robustness and effectiveness of the proposed control methodology are verified by detailed analysis and comparison using MATLAB®.

Advancing battery state of charge estimation in electric vehicles through deep learning: A comprehensive study using real-world driving data

Road crashes cause significant traffic delay, which can bring unnecessary financial losses. The objective of this study is to predict the level of delay caused by crashes (LDC) and discuss significant risk factors. To ensure the efficiency and accuracy of prediction, an improved stacking model was developed using Texas crash data of 2020. The first layer integrates seven base classifiers and the second layer tests three classifiers with different advantages. To improve and simplify the stacking model, three state-of-the-art methods—Bayesian hyperparameter optimization (BO), multiobjective feature selection (FS), and ensemble selection (ES)—were used. First, the hyperparameters and the least and most effective features were selected for each base classifier by BO and FS, respectively. Then ES, considering diversity and performance, selects the least base classifiers to reduce the input of the second layer. Finally, permutation feature importance was used to interpret the best stacking model. The results indicate that the stacking model achieves superior performance on four indicators: recall, G mean, F1 score, and area under the receiver operating characteristic (ROC) curve (AUC-ROC). FS significantly improves the efficiency of the stacking model and ES obtains a simplified stacking model without significantly reducing performance. In addition, the combination of the two methods (FS and ES) tends to achieve the best performance, and six risk factors have the greatest contributions in prediction using permutation feature importance. The prediction of LDC and the analysis of the main contributing factors help road managers respond to the rescue strategies to mitigate traffic congestion caused by crashes in a timely manner, thus minimizing economic losses.

ABSTRACTLithium‐ion batteries (LIBs) have gained significant attention in recent years due to their widespread applications in electric vehicles, portable electronics, energy storage, and renewable energy systems. However, their increasing use raises concerns about safety, reliability, and performance under various operating conditions. Among these, the impact of vibrations encountered during transportation, operation in vehicles, and industrial environments on battery performance, state of charge (SoC) and other critical parameters is of particular importance. This study is an overview that focuses on understanding the effects of vibrations on Li‐ion batteries (especially cylindrical, pouch, and prismatic cells) through a combination of experimental testing and simulation modeling results. The experimental results highlight the influence of vibration‐induced stress on electrical performance and battery degradation behavior. Simulations complement these findings by providing insights into the mechanical and electrochemical responses, effect on battery thermal management systems under different vibration frequencies and amplitudes. By addressing these effects comprehensively, this overview aims to contribute to the design of more robust Li‐ion battery systems capable of withstanding dynamic environments. The experimental studies show that discharge capacities were consistently decreasing depending on the cycles, frequencies, and the amplitudes of vibrations; this could be attributed to the separator and graphite anode material degradation. Also, the simulation results showed that the battery temperature management systems will be more effective under vibration conditions.

Evaluating the state of health (SOH) of lithium-ion batteries (LIBs) is essential for their safe deployment and the advancement of electric vehicles (EVs). Existing machine learning methods face challenges in the automation and effectiveness of feature extraction, necessitating improved computational efficiency. To address this issue, we propose a collaborative approach integrating an enhanced whale optimization algorithm (EWOA) for feature selection and a lightweight support vector regression (SVR) model for SOH estimation. Key features are extracted from charging voltage, current, temperature, and incremental capacity (IC) curves. The EWOA selects features by initially assigning weights based on importance scores from a random forest model. Gaussian noise increases population diversity, while a dynamic threshold method optimizes the selection process, preventing local optima. The selected features construct the SVR model for SOH estimation. This method is validated using four aging datasets from the NASA database, conducting 50 prediction experiments per battery. The results indicate optimal average absolute error (MAE) and root mean square error (RMSE) within 0.41% and 0.71%, respectively, with average errors below 1% and 1.3%. This method enhances automation and accuracy in feature selection while ensuring efficient SOH estimation, providing valuable insights for practical LIB applications.

Biochar-based persulfate activation: Rate constant prediction, key variables identification, and system optimization

Recent developments in lithium-ion (Li-ion) storage technology have enabled a revolution in the automotive industry. Fully electric vehicles (EVs) operate under the most diverse combination of driving and environmental conditions affecting the autonomy range. In other words, an equal state-of-charge (SOC) on two same model EV does not mean the same traveling distance since the conditions such as the state-of-health (SOH) of the battery, type of driver and even the type of route will influence the EV performance. Typically, SOC estimation algorithms are proposed and validated under controlled laboratory conditions. However, when real conditions are present, it is necessary to incorporate new tools capable of handling the diverse variability present in all the conditions. For instance, the topography of the route influences the current that the battery pack delivers, and the performance on the same route can be affected by the SOH. One of the main concerns for EV owners is that once a battery pack is installed, it becomes almost impossible to perform laboratory tests under controlled conditions. This paper proposes a novel approach to estimate the SOC by extending an existing SOC model (obtained in laboratory conditions) with the novelty of the assistance of Particle-Swarm-Optimization (PSO) to estimate the model parameters using real EV driving data. The data was obtained by a real-driving experiment, which consists on driving the EV in a complete discharge cycle on a highway. During this experiment, the initial SOC was at 100\%, and the idea was to discharge the battery pack driving through a highway where the driving conditions are almost uniform making it possible to characterize the SOC curve. Then, PSO is used to estimate the model parameters, and afterwards the model is validated in different types of routes. The obtained results show that the proposed approach can estimate the SOC satisfactorily. In this regard, this type of real-driving experiment can be performed by any driver, and by combining the particular results with the proposed approach, the users can personalize the SOC estimation model to their vehicles, and even more, create their own knowledge base of their EV performance through time. Therefore, the real-driving experiment can be replicated when needed to update the model parameters, thus allowing a better understanding of the actual SOH of the battery pack. Furthermore, by combining the obtained model with the elevation profile of a given route, the user can assess where to stop in case that a recharge is necessary.