Future Energy

Variability in initial battery cell characteristics and its implications for manufacturing quality control

2025-03-26T13:05:14+00:00

Ensuring manufacturing consistency in lithium-ion batteries is critical for reliable performance, safety, and longevity. This study examines the variability in initial pouch cell characteristics, including voltage, current, charge capacity, and discharge capacity, across 192 samples from 24 batches. Statistical analysis reveals that voltage remains relatively stable (mean = 3.951V, CV ≈ 6.45%), while charge and discharge capacities exhibit moderate variability (mean = 2.286Ah, CV ≈ 47.99% and mean = 2.350Ah, CV ≈ 52.53%, respectively). Current demonstrates the highest variability, with a mean of 0.280A and a CV of 195.25%, suggesting significant fluctuations possibly due to non-constant current and also likely influenced by process inconsistencies, operational conditions, or measurement sensitivity. Box plot and control chart analyses link many outliers to specific production factors, such as raw material lot changes and equipment maintenance cycles, pinpointing electrode preparation and formation as critical stages for mitigating variability. By integrating statistical insights with practical manufacturing considerations, this work provides a framework for proactive quality control, ultimately supporting scalable and high-quality lithium-ion battery production. While this study focuses on pouch cells, the underlying principles of variability analysis and targeted process improvements remain broadly relevant to other battery formats.

Rotor system fault detection utilizing semi-supervised and unsupervised machine learning

2025-06-05T04:19:41+00:00

Detecting multiple simultaneous faults in rotor systems is challenging, especially when labelled data is limited. This paper presents a novel framework combining unsupervised and semi-supervised machine learning to enhance fault diagnosis in rotor systems with various fault types. Using finite element method simulations, 100 vibration signal observations were generated for rotor systems under three fault conditions: imbalance, imbalance with shaft bending, and imbalance with cracking. Features were extracted via a multi-layer autoencoder in an unsupervised manner, followed by sequential feature selection to identify the most informative attributes. Two classification approaches were then applied: k-means clustering for unsupervised fault detection and a semi-supervised model with a Softmax layer for classification. The semi-supervised method achieved over 95% accuracy using only three selected features, effectively distinguishing different fault types. In contrast, the unsupervised approach proved better suited for anomaly detection rather than precise fault identification. These results demonstrate the potential of integrating unsupervised feature extraction with semi-supervised classification for reliable fault diagnosis in rotor systems with scarce labelled data.

Renewable energy in transportation: economic and environmental trade-offs

2025-06-13T06:48:30+00:00

This paper explores the benefits and challenges of transitioning from fossil-fueled vehicles to electric vehicles in Iran, with a focus on economic and environmental analysis. To this end, three different scenarios were considered to evaluate the impacts of this transition: the baseline system (fossil-fueled vehicles only), electric vehicles powered by fossil-based electricity, and electric vehicles powered by renewable energy. Each scenario was analyzed using various criteria, including fuel and maintenance costs, greenhouse gas emissions, required infrastructure investments, and return on investment. The results reveal that in the baseline scenario, annual CO₂ emissions of 73.25 million tons and total annual costs of $1.92 billion are among the main challenges. In the second scenario, with a 50% penetration of electric vehicles powered by fossil-based electricity, CO₂ emissions are reduced by 36.76 million tons, and the return on investment is achieved within five years. In the third scenario, assuming renewable energy sources supply electricity and a 70% penetration of electric vehicles, CO₂ emissions are reduced by 114.49 million tons, and a return on investment of 32.6% is achieved. These findings underscore the importance of integrating electric vehicles with renewable energy to achieve economic and environmental sustainability. The study highlights the critical need for developing renewable energy infrastructure and implementing appropriate policies to accelerate the transition to electric vehicles.

Transfer learning for power system fault location using artificial neural networks

2025-06-26T10:36:00+00:00

This paper investigates the application of transfer learning techniques to artificial neural networks (ANNs) for fault detection in power distribution systems, formulated as a classification problem. Comprehensive datasets are developed using multiple IEEE test feeders of varying complexity, including the 13-bus, 34-bus, 37-bus, and 123-bus test feeders. Various fault types are simulated across all three-phase buses in each system. Baseline performance is established by independently training ANNs on each feeder. Subsequently, knowledge learned from the complex 123-bus feeder is transferred to accelerate and improve fault location in simpler networks. The results demonstrate that transfer learning significantly improves both training efficiency and classification performance. Training convergence is accelerated by a factor of 1.68 to 2.56 across target feeders, corresponding to epoch reductions between 40.6% and 61.0%. Additionally, computational time is reduced by 24.0% to 49.5%, further enhancing the practical viability of the proposed approach. These findings suggest that transfer learning offers a powerful strategy to address data scarcity and computational challenges in fault location, enabling utilities to deploy accurate, efficient fault detection systems across diverse distribution networks with minimal retraining effort.

Third-generation biodiesel development in Bangladesh: a review of recent trends, prospects, and economic analysis

2025-06-27T19:37:29+00:00

Researchers worldwide are seeking alternative sources of energy that can meet the future energy demand while significantly mitigating greenhouse gas (GHG) emissions. In addition, increasing the global energy demand at a faster rate, dependency on fossil fuels, and the price of fossil fuels are increasing at an alarming rate day by day. Amongst the options, biodiesel as an environmentally sustainable renewable fuel is considered to make a substantial contribution to the future transport energy demands locally and internationally. Among different biodiesel sources, advanced biodiesel feedstock, so-called 3rd generation biodiesel, which is mainly derived from microalgae, can be taken into account as a promising potential feedstock for biodiesel production due to their fast growth rates, high lipid content, high production rate, and ability to capture carbon dioxide. This paper discusses the selection of 3rd generation biodiesel, recent trends in microalgae-based biodiesel production, challenges in large-scale commercialization, and prospects for its development, particularly in the scenario of Bangladesh. The study estimates biodiesel can produce 10,000 liters per acre per year, with great potential for CO₂ sequestration and job development. The study also looks at how Bangladesh's plentiful freshwater and saline water supplies might be used for microalgae farming in desolate areas. Microalgae biodiesel can be generated at 0.50–0.75 USD per liter, according to a cost study; government subsidies and economies of scale will help to further lower this figure. Microalgae biodiesel can help Bangladesh reach energy security, lower greenhouse gas emissions, and encourage sustainable economic development by being included in their renewable energy plan. This paper will provide a clear understanding of the potential usages of microalgae biodiesel as an alternative source to fossil fuel.