The increasing adoption of batteries in a variety of applications has highlighted the necessity of accurate parameter identification and effective modeling, especially for lithium-ion batteries, which are preferred due to their high power and energy densities. This paper proposes a comprehensive framework using the Levenberg–Marquardt algorithm (LMA) for validating and identifying lithium-ion battery model parameters to improve the accuracy of state of charge (SOC) estimations, using only discharging measurements in the N-order Thevenin equivalent circuit model, thereby increasing computational efficiency. The framework encompasses two key stages: model parameter identification and model verification. This framework is validated using experimental measurements on the INR 18650-20R battery, produced by Samsung SDI Co., Ltd. (Suwon, Republic of Korea), conducted by the Center for Advanced Life Cycle Engineering (CALCE) battery group at the University of Maryland. The proposed framework demonstrates robustness and accuracy. The results indicate that optimization using only the discharging data suffices for accurate parameter estimation. In addition, it demonstrates excellent agreement with the experimental measurements. The research underscores the effectiveness of the proposed framework in enhancing SOC estimation accuracy, thus contributing significantly to the reliable performance and longevity of lithium-ion batteries in practical applications.
Accurate simulations of gas turbines’ dynamic performance are essential for improvements in their practical performance and advancements in sustainable energy production. This paper presents models with extremely accurate simulations for a real dual-fuel gas turbine using two state-of-the-art techniques of neural networks: the dynamic neural network and deep neural network. The dynamic neural network has been realized via a nonlinear autoregressive network with exogenous inputs (NARX) artificial neural network (ANN), and the deep neural network has been based on a convolutional neural network (CNN). The outputs selected for simulations are: the output power, the exhausted temperature and the turbine speed or system frequency, whereas the inputs are the natural gas (NG) control valve, the pilot gas control valve and the compressor variables. The data-sets have been prepared in three essential formats for the training and validation of the networks: normalized data, standardized data and SI units’ data. Rigorous effort has been carried out for wide-range trials regarding tweaking the network structures and hyper-parameters, which leads to highly satisfactory results for both models (overall, the minimum recorded MSE in the training of the MISO NARX was 6.2626 × 10−9 and the maximum MSE that was recorded for the MISO CNN was 2.9210 × 10−4, for more than 15 h of GT operation). The results have shown a comparable satisfactory performance for both dynamic NARX ANN and the CNN with a slight superiority of NARX. It can be newly argued that the dynamic ANN is better than the deep learning ANN for the time-based performance simulation of gas turbines (GTs).
Accurate and reliable calibration of power plant models in the power system is crucial for maintaining grid stability and preventing power outages. This becomes even more crucial as the power grids become more dynamic and stochastic with the ever-increasing penetration of renewable energy sources, mid-size generators, smart loads, and energy storage. Current practice such as staged-test is costly and timeconsuming. Alternatively, recent PMU-based approaches using online measurements without interfering with generators operation have been introduced to provide a scalable and low-cost to meet North American Electric Reliability Corporation (NERC) standards. This research paper proposes a PMU-based framework to validate and calibrate model parameters by a new iterative deep learning approach. The results show that the proposed approach can accurately calibrate power plantmodels from a single disturbance event without the urgent need of prior knowledge of the model’s initial parameters. The proposed method has been validated on real-world and simulated scenarios on a relatively large number of calibrated parameters with an average error of 1.43%.
Wireless Sensor Networks (WSNs) encounter a substantial challenge when it comes to energy conservation. As sensor nodes rely on battery power to operate in unattended environments, and it can be inconvenient to recharge the batteries. So, maximizing the network’s operational lifetime by minimizing energy consumption is crucial for improving the quality of service. Anomaly detection is widely used technique in WSNs to identify anomalies or unusual events. However, timely anomaly detection can be challenging to execute reliably in real-time. This study presents a methodology that focuses on energy efficiency when detecting anomalies in real-time settings for multivariate time series data by utilizing compression of measurement data and reducing operating time without compromising measurement resolution. This effectively reduces energy consumption. Instead of modeling the time series of each sensor individually, the proposed deep sequential learning approach models the time series of multiple sensors concurrently, accounting for potential interactions among them. Additionally, the proposed deep sequential learning approach eliminates the need for labeled data and can be directly applied to real-world scenarios where labeling a large data stream is impractical and time-consuming. Finally, experiments with a real-world WSN demonstrate that the proposed approach is both adequate and robust in practice.