• Energy Research

Deep learning has gained much traction in application to state-of-charge (SOC) estimation for Li-ion batteries in electric vehicle applications. However, with the vast selection of architectures and hyperparameter combinations, it remains challenging to design an accurate and robust SOC estimation model with a sufficiently low computation cost. Therefore, this study provides a comparative evaluation among commonly used deep learning models from the recurrent, convolutional, and feedforward architecture benchmarked on an openly available Li-ion battery dataset. To evaluate model robustness and generalization capability, we train and test models on different drive cycles at various temperatures and compute the root mean squared error (RMSE) and mean absolute error metric. To evaluate model computation costs, we run models in real-time and record the model size, floating-point operations per second (FLOPS), and run-time duration per datapoint. This study proposes a two-hidden layer stacked gated recurrent unit model trained with a one-cycle policy learning rate scheduler. The proposed model achieves a minimum RMSE of 0.52% on the train dataset and 0.65% on the test dataset while maintaining a relatively low computation cost. Executing the proposed model in real-time takes up approximately 1 MB in disk space, 300K FLOPS, and 0.03 ms run-time per datapoint. This makes the proposed model feasible to be executed on lightweight battery management system processors.

The use of renewable energy techniques is becoming increasingly popular because of rising demand and the threat of negative carbon footprints. Wind power offers a great deal of untapped potential as an alternative source of energy. The rising demand for wind energy typically results in the generation of high-quality output electricity through grid integration. More sophisticated contemporary generators, power converters, energy management, and controllers have been recently developed to integrate wind turbines into the electricity system. However, a comprehensive review of the role of converters in the wind system’s power conversion, control, and application toward sustainable development is not thoroughly investigated. Thus, this paper proposes a comprehensive review of the impact of converters on wind energy conversion with its operation, control, and recent challenges. The converters’ impact on the integration and control of wind turbines was highlighted. Moreover, the conversion and implementation of the control of the wind energy power system have been analyzed in detail. Also, the recently advanced converters applications for wind energy conversion were presented. Finally, recommendations for future converters use in wind energy conversions were highlighted for efficient, stable, and sustainable wind power. This rigorous study will lead academic researchers and industry partners toward the development of optimal wind power technologies with improved efficiency, operation, and costs.

The overheating of heat pipes, poor transfer of heat across the absorber and finned heat pipes, and inability to provide hot water in the late evening hours are major problems associated with conventional heat pipe vacuum collector systems. The amalgamation of highly conductive storage material between the absorber tube (heat collecting surface) and the heat pipe is an effective way to overcome these problems. In this study, a stearic acid amalgamated vacuum tube solar collector system was designed and fabricated and its thermal output compared with a conventional vacuum tube system without storage material under the same environmental conditions. The experimental results showed that the amalgamation of stearic acid as storage material enhanced the thermal output of the solar system compared to the conventional one. The desired heat gain of the solar system with storage material increased by 31.30, 23.34, and 18.78% for Test 1_40 °C, Test 2_45 °C, and Test 3_50 °C, respectively. The technoeconomic analysis showed that almost 118.80 USD in revenue could be earned by the proposed solar system at the end of 15 years. The total running cost of ELG and the developed solar system was observed to be 202.62 and 86.70 USD, respectively. On average, the cost of hot water production using the solar system and ELG was found to be 0.0016 and 0.004 USD/L, respectively. The value of LEC was found to be 0.062 USD/electricity unit, which was much lower than the LEC value of ELG (0.116 USD/electricity unit). The value of NPW (73.73 USD) indicated high acceptability of the proposed system. The payback time is lower than the life of the system, indicating its suitability for use in the commercial sector. Therefore, the proposed solar system is highly recommended over conventional water heating systems in urban and rural areas.

The state of charge (SOC) is a crucial indicator of a Li-ion battery management system (BMS). A BMS with a good SOC assessment can dramatically improve the lifespan of the battery and ensure the safety of the end-user. With deep learning making tremendous strides in many other fields, this study aims to provide an empirical evaluation of commonly used deep learning methods on the task of SOC estimation. We propose the use of two-hidden-layer gated recurrent units (GRU) to estimate the SOC at various ambient temperatures. In this work, we conducted two experiment setups to showcase the capability of the proposed GRU model. In the first setup, the GRU was trained on the DST, BJDST and US06 drive cycle and evaluated the FUDS drive cycle upon convergence. The same procedure was repeated with the second setup except the GRU was trained on the DST, BJDST and FUDS drive cycle and evaluated on the US06 drive cycle. In both experiment setups, the proposed GRU was evaluated on a novel drive cycle that it has not encountered during the training phase. We show that a two-hidden-layer GRU with appropriate hyperparameter combination and training methodology can reliably estimate the SOC of novel drive cycles at various ambient temperatures in comparison with other deep learning methods such as simple recurrent network (SRNN), Long Short-Term Memory (LSTM), 1D Residual Network (Resnet), 1D Visual Geometry Group Network (VGG) and the Multilayer Perceptron (MLP). The proposed GRU achieves 2.3% RMSE on the FUDS drive cycle and 1.2% RMSE on the US06 drive cycle outperforming all other models.

Abstract Many countries around the world are planning to reach 100% renewable energy (RE) use by 2050. In this context and due to the recent sharp increase in RE utilization in the global energy mix along with its progressive impact on the world energy sector, the evaluation and investigation of its effect on achieving sustainable development goals (SDGs) are not covered sufficiently. Here, we present an assessment of the emerging role of RE utilization and artificial intelligence (AI) toward achieving SDGs. A total of 17 SDGs were divided into three groups, namely, environment, society, and economy, as per the three key pillars of sustainable development. The RE has a positive impact toward achieving 75 targets across all SDGs by using an expert elicitation method-based consensus. However, it may negatively affect the accomplishment of the 27 targets. In addition, the AI can help the RE to enable the attainment of 42 out of 169 targets. With the current exponential growth of RE share and AI development together with addressing certain present limitations, this impact may cover additional targets in the future. Nevertheless, the present research foci neglect significant facets. The exponential growth of RE share and rapid evolution of AI need to be accompanied through the requisite regulatory insight and technology regulation to cover additional targets in the future.