Energy and AI最新文献

Proton exchange membrane (PEM) fuel cells have significant potential for clean power generation, yet challenges remain in enhancing their performance, durability, and cost-effectiveness, particularly concerning metallic bipolar plates, which are pivotal for lightweight compact fuel cell stacks. Protective coatings are commonly employed to combat metallic bipolar plate corrosion and enhance water management within stacks. Conventional methods for predicting coating performance in terms of corrosion resistance involve complex physical-electrochemical modelling and extensive experimentation, with significant time and cost. In this study machine learning techniques are employed to model metallic bipolar plate coating performance, diamond-like-carbon coatings of varying thicknesses deposited on SS316L are considered, and coating performance is evaluated using potentiodynamic polarization and electrochemical impedance spectroscopy. The obtained experimental data is split into two datasets for machine learning modelling: one predicting corrosion current density and another predicting impedance parameters. Machine learning models, including extreme gradient boosting (XGB) and artificial neural networks (ANN), are developed, and optimized to predict coating performance attributes. Data preprocessing and hyperparameter tuning are carried out to enhance model accuracy. Results show that ANN outperforms XGB in predicting corrosion current density, achieving an R² > 0.98, and accurately predicting impedance parameters with an R² > 0.99, indicating that the models developed are very promising for accurate prediction of the corrosion performance of coated metallic bipolar plates for PEM fuel cells.

The combination of Machine Learning (ML), smart energy meters, and availability of household appliance energy profile data has opened new opportunities for Non-Intrusive Load Monitoring (NILM). However, the number of options makes it challenging in selecting optimal combinations for different energy applications, which requires studies to examine their trade-offs.

This paper contributes one such study that investigated four established ML approaches – K Nearest Neighbour (KNN), Support Vector Machine (SVM), eXtreme Gradient Boosting (XGBoost) and Convolutional Neural Network (CNN) – and their performance in classifying appliance events from Alternating Current (AC) and Root Mean Square (RMS) energy data where the sampling frequency and training dataset set size was varied (10 Hz–1 kHz and 50–2000 examples per class, respectively). The computational expense during training, testing and storage was also assessed and evaluated with reference to real-world applications.

The CNN classifier trained on AC data at 500 Hz and 11,000 examples gave the best F1-score $\left(0.989\right)$ followed by the KNN classifier $\left(0.940\right)$. The storage size required by the CNN models was 3̃MB, which is very close to fitting on cost-effective embedded system microcontrollers. This would prevent high-rate data needing to be sent to the cloud as analysis could be performed on edge computing Internet-of-Things (IoT) devices.

National-level climate action plans are often formulated broadly. Spatially disaggregating these plans to individual municipalities can offer substantial benefits, such as enabling regional climate action strategies and for assessing the feasibility of national objectives. Numerous spatial disaggregation approaches can be found in the literature. This study reviews and categorizes these. The review is followed by a discussion of the relevant methods for the disaggregation of climate action plans. It is seen that methods employing proxy data, machine learning models, and geostatistical ones are the most relevant methods for the spatial disaggregation of national energy and climate plans. The analysis offers guidance for selecting appropriate methods based on factors such as data availability at the municipal level and the presence of spatial autocorrelation in the data.

As the urgency of addressing climate change escalates, understanding the spatial aspects of national energy and climate strategies becomes increasingly important. This review will serve as a valuable guide for researchers and practitioners applying spatial disaggregation in this crucial field.

Insulators are essential equipment to ensure the safety and reliability of power transmission systems. Defective insulators may cause partial discharge and even lead to serious safety accidents. Hence it is necessary to accurately identify the defective insulator from a string of insulators. However, small defect poses significant challenges for recognizing the defective insulator from a large number of insulators. To address these issues, we collect and annotate the randomly generated defect dataset (RGDD). Further, the discriminative feature learning-based detector (DFLD) is constructed based on the pattern of backbone-neck-head. Specifically, considering the simultaneous existence of many insulators, attention-based bidirectional feature pyramid (ABFP) is designed to capture the discriminative information. Considering the small size of defective part, the efficient receptive field adaptation (ERFA) module is constructed to enhance the perception of contextual information related to defective insulators. Meanwhile, the two-stage detection head is designed to correct the location of defective insulators. It also adapts to the shape variation of insulators by the deformable convolution. On this basis, the keypoints method is introduced to more accurately represent the location of defective insulators. Due to the imbalance between positive and negative samples, the Adaptive Threshold Sample Assignment (ATSA) Strategy is proposed for selecting the best positive samples. DFLD has achieved good detection performance compared with classical object detection networks on the RGDD dataset and CPLID dataset. The ablation experiments are conducted on the RGDD dataset. It is verified that the discriminative features from DFLD can effectively recognize the small defect from insulators.

Deep geothermal energy presents large untapped renewable energy potential could significantly contribute to global energy needs. However, developing geothermal projects involves uncertainties regarding adequate geothermal brine extraction and huge costs related to preparation phases and consequently drilling and stimulation activities. Therefore, evaluating utilization alternatives of such projects is a complex decision-making problem effectively addressed using multi-criteria decision-making (MCDM) methods. This study introduces the MCDM method utilizing analytic hierarchy process (AHP) and weighted decision matrix (WDM) to assess different utilization alternatives (electricity generation, direct heat use and cogeneration). The AHP method determines the weight of each criterion and sub-criterion, while the WDM calculates the final project grade. Five criteria groups - technological, geological, economic, societal and environmental – comprising twenty-eight influencing factors were selected and used for the assessment of investment in Enhanced Geothermal Systems (EGS) projects. The AHP-WDM method was used by 38 experts from six categories: industry, educational institution, research and technology organization (RTO), small- and medium-sized enterprises (SME), local community and other. These diverse expert inputs aimed to capture varying perspectives and knowledge influence investment decisions in geothermal energy. The results were analysed accordingly. The results underscore the importance of incorporating different viewpoints to develop robust, credible, and effective investment strategies for EGS projects. Therefore, this method will contribute to more efficient EGS project development, enabling thus a greater penetration of the EGS into the market. Additionally, the proposed AHP-WDM method was implemented for a case study examining two locations. Locations were assessed and compared on scenario-based evaluation. The results confirmed the method's adequacy for assessing various end uses and comparing project feasibility across different locations.

Coal-gangue object detection has attracted substantial attention because it is the core of realizing vision-based intelligent and green coal separation. However, most existing studies have been focused on laboratory datasets and prioritized model lightweight. This makes the coal-gangue object detection challenging to adapt to the complex and harsh scenes of real production environments. Therefore, our project collected and labeled image datasets of coal and gangue under real production conditions from a coal preparation plant. We then designed a one-stage object model, named STATNet, following the “backbone-neck-head” architecture with the aim of enhancing the detection accuracy under industrial coal preparation scenarios. The proposed model utilizes Swin Transformer as backbone module to extract multi-scale features, improved path augmentation feature pyramid network (iPAFPN) as neck module to enrich feature fusion, and task-aligned head (TAH) as head module to mitigate conflicts and misalignments between classification and localization tasks. Experimental results on a real-world industrial dataset demonstrate that the proposed STATNet model achieves an impressive AP₅₀ of 89.27 %, significantly surpassing several state-of-the-art baseline models by 2.02 % to 5.58 %. Additionally, it exhibits stronger robustness in resisting image corruption and perturbation. These findings demonstrate its promising prospects in practical coal and gangue separation applications.

The physicochemical properties of liquid alternative fuels are important but difficult to measure/predict, especially when complex surrogate fuels are concerned. In the present work, machine learning is used to develop quantitative structure–property relationship models. The fuel chemical structure is represented by molecular descriptors, allowing the linking of important features of the fuel composition and key properties of fuel utilization. Feature selection is employed to select the most relevant features that describe the chemical structure of the fuel and several machine learning algorithms are tested to construct interpretable models. The effectiveness of the methodology is demonstrated through the development of accurate and interpretable predictive models for cetane numbers, with a focus on understanding the link between molecular structure and fuel properties. In this context, matrix-based descriptors and descriptors related to the number of atoms in the molecule are directly linked with the cetane number of hydrocarbons. Furthermore, the results showed that molecular connectivity indices play a role in the cetane number for aromatic molecules. Also, the methodology is extended to predict the cetane number of ester and ether molecules, leveraging the design of alternative fuels towards fully sustainable fuel utilization.

The marine climate conditions are intricate and variable. In scenarios characterized by high proportions of wind and solar energy access, the uncertainty regarding the energy sources for island microgrid is significantly exacerbated, presenting challenges to both the economic viability and reliability of the capacity configuration for island microgrids. To address this issue, this paper proposes a distributionally robust optimization (DRO) method for island microgrids, considering extreme scenarios of wind and solar conditions. Firstly, to address the challenge of determining the probability distribution functions of wind and solar in complex island climates, a conditional generative adversarial network (CGAN) is employed to generate a scenario set for wind and solar conditions. Then, by combining k-means clustering with an extreme scenario selection method, typical scenarios and extreme scenarios are selected from the generated scenario set, forming the scenario set for the DRO model of island microgrids. On this basis, a DRO model based on multiple discrete scenarios is constructed with the objective of minimizing the sum of investment costs, operation and maintenance costs, fuel purchase costs, penalty costs of wind and solar curtailment, and penalty costs of load loss. The model is subjected to equipment operation and power balance constraints, and solved using the columns and constraints generation (CCG) algorithm. Finally, through typical examples, the effectiveness of this paper’s method in balancing the economic viability and robustness of the configuration scheme for the island microgrid, as well as reducing wind and solar curtailment and load loss, is verified.

Proton-exchange membrane fuel cells (PEMFCs) are critical components of zero-emission electro-hydrogen generators. Accurate performance prediction is vital to the optimal operation management and preventive maintenance of these generators. Polarization curve remains one of the most important features representing the performance of PEMFCs in terms of efficiency and durability. However, predicting the polarization curve is not trivial as PEMFCs involve complex electrochemical reactions that feature multiple nonlinear relationships between the operating variables as inputs and the voltage as outputs. Herein, we present an artificial-intelligence-based approach for predicting the PEMFCs’ performance. In that way, we propose first an explainable solution for selecting the relevant features based on kernel principal component analysis and mutual information. Then, we develop a machine learning approach based on XGBRegressor and Bayesian optimization to explore the complex features and predict the PEMFCs’ performance. The performance and the robustness of the proposed machine learning based prediction approach is tested and validated through a real industrial dataset including 10 PEMFCs. Furthermore, several comparison studies with XGBRegressor and the two popular machine learning-based methods in predicting PEMFC performance, such as artificial neural network (ANN) and support vector machine regressor (SVR) are also conducted. The obtained results show that the proposed approach is more robust and outperforms the two conventional methods and the XGBRegressor for all the considered PEMFCs. Indeed, according to the coefficient of determination criterion, the proposed model gains an improvement of 6.35%, 6.8%, and 4.8% compared with ANN, SVR, and XGBRegressor respectively.

Data-driven models for battery state estimation require extensive experimental training data, which may not be available or suitable for specific tasks like open-circuit voltage (OCV) reconstruction and subsequent state of health (SOH) estimation. This study addresses this issue by developing a transfer-learning-based OCV reconstruction model using a temporal convolutional long short-term memory (TCN-LSTM) network trained on synthetic data from an automotive nickel cobalt aluminium oxide (NCA) cell generated through a mechanistic model approach. The data consists of voltage curves at constant temperature, C-rates between $\text{C}/30$ to $1\text{C}$, and a SOH-range from 70 % to 100 %. The model is refined via Bayesian optimization and then applied to four use cases with reduced experimental nickel manganese cobalt oxide (NMC) cell training data for higher use cases. The TL models’ performances are compared with models trained solely on experimental data, focusing on different C-rates and voltage windows. The results demonstrate that the OCV reconstruction mean absolute error (MAE) within the average battery electric vehicle (BEV) home charging window (30 % to 85 % state of charge (SOC)) is less than 22 mV for the first three use cases across all C-rates. The SOH estimated from the reconstructed OCV exhibits an mean absolute percentage error (MAPE) below 2.2 % for these cases. The study further investigates the impact of the source domain on TL by incorporating two additional synthetic datasets, a lithium iron phosphate (LFP) cell and an entirely artificial, non-existing, cell, showing that solely the shifting and scaling of gradient changes in the charging curve suffice to transfer knowledge, even between different cell chemistries. A key limitation with respect to extrapolation capability is identified and evidenced in our fourth use case, where the absence of such comprehensive data hindered the TL process.