The internal short-circuit (ISC) fault is one of the main faults, which results in thermal runaway of lithium-ion batteries. However, it is still challenging to detect the ISC fault when it comes to a battery energy storage system composed of battery packs. In response to this challenge, this paper proposes a novel ISC fault-diagnosis method based on fused features that could reflect the transient response and steady-state characteristics of the ISC fault. The features are obtained by measuring the battery voltage and estimating the state of charge (SOC) via an unscented Kalman filter. Then, the features are normalized and fused using Euclidean distance for fault determination, in order to collectively improve reliability and prevent misdiagnosis. The results of internal short-circuit experimental tests demonstrate that, compared to an existing method based only on SOC, adding voltage can significantly improve fault-diagnosis sensitivity and achieve at least 1.3 hrs earlier fault detection, with only a slightly increased computational cost. The proposed fault-diagnosis method for battery packs shows great potential for practical application.
Cassava is a widely cultivated, economically accessible tropical crop that has gained attention not only as a food staple but also as a promising feedstock for bioethanol production. This study explored the application of gamma-irradiation as a nuclear-based alternative or complement to conventional enzymatic hydrolysis in cassava bioethanol conversion. Two processing pathways were compared across irradiation doses of 0, 50, and 100 kGy: a direct fermentation-distillation pathway without hydrolysis (FD), and a sequential hydrolysis-fermentation-distillation pathway (HFD). Results showed that gamma-irradiation alone could not adequately replace hydrolysis, as the FD treatments yielded only 0.10-0.11 g sugar/g sample with negligible ethanol output. In contrast, combining irradiation with hydrolysis substantially improved process performance; the HFD treatments achieved ethanol yields of 0.20-0.25 g/g sugar and concentrations of 62% (v/v). Moreover, the combined treatments (50HFD and 100HFD) reduced fermentation time from 72 hr to approximately 12-24 hr. These findings indicate that gamma-irradiation functions most effectively as a synergistic pretreatment rather than a standalone process. To our knowledge, this is the first study to directly compare gamma-irradiation-only and gamma-irradiation-plus-hydrolysis pathways in cassava bioethanol production, identifying 50 kGy as an optimal irradiation dose. [GRAPHICS]
A comprehensive understanding of wind dynamics is essential for climate-resilient wind farm development under changing climatic conditions. This study evaluates wind energy potential at five coastal locations in Southeast Asia-Pangandaran (Indonesia), Ca Mau (Vietnam), Phuket (Thailand), Muar (Malaysia), and San Jose (Philippines) using 40 years of wind data at 50 m hub height. The analysis integrates seasonal variability, multi-decadal trends, and climate influence to evaluate wind availability and climate sensitivity. The results reveal that the peak wind speeds generally occur during December-January due to monsoon dominance, except in Pangandaran, where the maximum shifts to August. A modest decline in mean annual wind speeds is observed from 1981-2000 to 2001-2020; however, all sites remain suitable for modern wind energy applications. Wind speed ranges are 3.2-8.5 m/s in Ca Mau, 2.7-7.5 m/s in San Jose, 2.5-6.2 m/s in Phuket, 2.8-6.7 m/s in Muar, and 2.8-6.0 m/s in Pangandaran. Surface pressure and relative humidity are identified as key climatic drivers of wind variability. Among the six probability distribution models, the Nonlinear Least Squares method provides the best fit. Seasonal wind power density peaks during December-February at most sites, supporting sustainable wind energy deployment and long-term planning across Southeast Asia
Accurate building energy consumption prediction is crucial for optimizing energy efficiency and operational flexibility in modern buildings. However, existing prediction methods struggle to handle the highly complex, nonlinear, and multifactor-influenced nature of building energy data, often failing to fully exploit the intrinsic value and underlying characteristics of varying data components. To address these limitations, this study aims to develop a novel hybrid deep-learning framework that enhances multi-step prediction accuracy by decoupling and independently modeling the underlying patterns of energy time-series data. Methodologically, the proposed approach integrates STL with a divide-and-conquer predictive strategy. First, the original energy consumption series is decomposed into three interpretable components: trend, seasonality, and residual. Next, the random forest algorithm is applied to identify the most relevant input variables from historical energy, calendar, and meteorological data for each specific component. Subsequently, tailored deep-learning models are assigned based on component characteristics: LSTM networks model the long-term trend, the Transformer architecture captures the periodic seasonal patterns, and a Res-LSTM tackles the highly stochastic residual component. The proposed framework is validated using real-world data from the public BDG2 dataset. Experimental results demonstrate that the STL-based hybrid model significantly outperforms single prediction models (including BPNN, SVR, RNN, LSTM, transformer, and Res-LSTM). Specifically, compared to the best-performing single model, the proposed method yields improvements of 20.1%, 12.7%, and 23.9% in mean absolute error (MAE), root mean square error (RMSE), and coefficient of variation of RMSE (CV-RMSE), respectively. With achieved final values of 0.0397 kW for RMSE and 0.94 for the coefficient of determination (R 2), this study concludes that the proposed hybrid methodology effectively mitigates noise and non-stationarity in energy data. It provides a highly accurate, robust, and interpretable tool to support intelligent building energy management and system scheduling.
