Energy and AI最新文献

Predicting the thermal conductivity of polymer composites with one-dimensional oriented fillers using the combination of deep learning and ensemble learning 利用深度学习和集合学习相结合的方法预测带有一维定向填料的聚合物复合材料的导热性能

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-11-08 DOI: 10.1016/j.egyai.2024.100445

Yinzhou Liu , Weidong Zheng , Haoqiang Ai , Lin Cheng , Ruiqiang Guo , Xiaohan Song

Polymer composites with one-dimensional (1D) oriented fillers, recognized for their high thermal conductivity (TC), are extensively utilized in cooling electronic components. However, the prediction of the TC of composites with 1D oriented fillers poses a challenge due to the significant impact of filler orientation on composite TC. In this paper, we use a strategy that combines deep learning and ensemble learning to efficiently and quickly predict the TC of composites with 1D oriented fillers. First, as a control, we used convolutional neural network (CNN) model to predict the TC of 1D carbon fiber-epoxy composite, and the R-squared (R²) on the test set reached 0.924. However, for composites consist of different matrices and fillers, the CNN model needs to be retrained, which greatly wastes computing resources. Therefore, we define a descriptor ‘Orientation degree (O_d)’ to quantitatively describe the spatial distribution of the 1D fillers. CNN model was used to predict this structural parameter, the accuracy R² can reach 0.950 on the test set. Using O_d as a feature, random forest regression (RFR) was used to predict the TC, and the accuracy R² reached 0.954 on the test set, which was higher than that of CNN control group. We further successfully extended this strategy to composites consist of different 1D fillers and matrices, and only one CNN model and one RFR model needed to be trained to achieve fast and accurate TC prediction. This strategy provides valuable insights and guidance for machine learning-based material property prediction.

带有一维（1D）取向填料的聚合物复合材料具有公认的高热导率（TC），被广泛用于冷却电子元件。然而，由于填料取向对复合材料的热导率有很大影响，因此对带有一维取向填料的复合材料的热导率进行预测是一项挑战。在本文中，我们采用深度学习和集合学习相结合的策略，高效、快速地预测带有一维取向填料的复合材料的热导率。首先，作为对照，我们使用卷积神经网络（CNN）模型预测了一维碳纤维-环氧树脂复合材料的热导率，测试集上的 R 平方（R2）达到了 0.924。但是，对于由不同基材和填料组成的复合材料，CNN 模型需要重新训练，这极大地浪费了计算资源。因此，我们定义了一个描述符 "方向度（Od）"来定量描述一维填料的空间分布。我们使用 CNN 模型来预测这一结构参数，在测试集上的准确率 R2 可以达到 0.950。以 Od 为特征，使用随机森林回归（RFR）预测 TC，在测试集上的准确率 R2 达到 0.954，高于 CNN 对照组。我们进一步成功地将这一策略扩展到了由不同一维填料和矩阵组成的复合材料上，只需训练一个 CNN 模型和一个 RFR 模型即可实现快速准确的 TC 预测。该策略为基于机器学习的材料性能预测提供了宝贵的见解和指导。

{"title":"Predicting the thermal conductivity of polymer composites with one-dimensional oriented fillers using the combination of deep learning and ensemble learning","authors":"Yinzhou Liu , Weidong Zheng , Haoqiang Ai , Lin Cheng , Ruiqiang Guo , Xiaohan Song","doi":"10.1016/j.egyai.2024.100445","DOIUrl":"10.1016/j.egyai.2024.100445","url":null,"abstract":"<div><div>Polymer composites with one-dimensional (1D) oriented fillers, recognized for their high thermal conductivity (TC), are extensively utilized in cooling electronic components. However, the prediction of the TC of composites with 1D oriented fillers poses a challenge due to the significant impact of filler orientation on composite TC. In this paper, we use a strategy that combines deep learning and ensemble learning to efficiently and quickly predict the TC of composites with 1D oriented fillers. First, as a control, we used convolutional neural network (CNN) model to predict the TC of 1D carbon fiber-epoxy composite, and the R-squared (R<sup>2</sup>) on the test set reached 0.924. However, for composites consist of different matrices and fillers, the CNN model needs to be retrained, which greatly wastes computing resources. Therefore, we define a descriptor ‘Orientation degree (<em>O<sub>d</sub></em>)’ to quantitatively describe the spatial distribution of the 1D fillers. CNN model was used to predict this structural parameter, the accuracy R<sup>2</sup> can reach 0.950 on the test set. Using <em>O<sub>d</sub></em> as a feature, random forest regression (RFR) was used to predict the TC, and the accuracy R<sup>2</sup> reached 0.954 on the test set, which was higher than that of CNN control group. We further successfully extended this strategy to composites consist of different 1D fillers and matrices, and only one CNN model and one RFR model needed to be trained to achieve fast and accurate TC prediction. This strategy provides valuable insights and guidance for machine learning-based material property prediction.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"18 ","pages":"Article 100445"},"PeriodicalIF":9.6,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A hybrid wind power prediction model based on seasonal feature decomposition and enhanced feature extraction 基于季节特征分解和增强特征提取的混合风能预测模型

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-11-08 DOI: 10.1016/j.egyai.2024.100442

Weipeng Li , Yuting Chong , Xin Guo , Jun Liu

Efficient and accurate wind power prediction is crucial for enhancing the reliability and safety of power system. The data-driven forecasting methods are regarded as an effective solution. However, the inherent randomness and nonlinearity of wind power systems, along with the abundance of redundant information in measurement data, present challenges to forecasting methods. The integration of precise and efficient techniques for data feature decomposition and extraction is essential in conjunction with advanced data-driven forecasting models. Focus on the seasonal variation characteristics of wind energy, a hybrid wind power prediction model based on seasonal feature decomposition and enhanced feature extraction is proposed. The effectiveness and superiority of the proposed method in predictive accuracy are demonstrated through comprehensive multi-model experiment comparisons.

高效、准确的风电预测对提高电力系统的可靠性和安全性至关重要。数据驱动的预测方法被认为是一种有效的解决方案。然而，风力发电系统固有的随机性和非线性，以及测量数据中大量的冗余信息，给预测方法带来了挑战。将精确高效的数据特征分解和提取技术与先进的数据驱动预测模型相结合至关重要。针对风能的季节变化特征，提出了一种基于季节特征分解和增强特征提取的混合风能预测模型。通过多模型综合实验对比，证明了所提方法在预测精度方面的有效性和优越性。

引用次数: 0

Integrating local knowledge with ChatGPT-like large-scale language models for enhanced societal comprehension of carbon neutrality 将地方知识与类似 ChatGPT 的大规模语言模型相结合，提高社会对碳中和的理解能力

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-11-07 DOI: 10.1016/j.egyai.2024.100440

Te Han, Rong-Gang Cong, Biying Yu, Baojun Tang, Yi-Ming Wei

Addressing carbon neutrality presents a multifaceted challenge, necessitating collaboration across various disciplines, fields, and societal stakeholders. With the increasing urgency to mitigate climate change, there is a crucial need for innovative approaches in communication and education to enhance societal understanding and engagement. Large-scale language models like ChatGPT emerge as transformative tools in the AI era, offering potential to revolutionize how we approach economic, technological, social, and environmental issues of achieving carbon neutrality. However, the full potential of these models in carbon neutrality is yet to be realized, hindered by limitations in providing detailed, localized, and expert-level insights across an expansive spectrum of subjects. To bridge these gaps, this paper introduces an innovative framework that integrates local knowledge with LLMs, aiming to markedly enhance the depth, accuracy, and regional relevance of the information provided. The effectiveness of this framework is examined from government, corporations, and community perspectives. The integration of local knowledge with LLMs not only enriches the AI’s comprehension of local specificities but also guarantees an up-to-date information that is crucial for addressing the specific concerns and questions about carbon neutrality raised by a broad array of stakeholders. Overall, the proposed framework showcases significant potential in enhancing societal comprehension and participation towards carbon neutrality.

解决碳中和问题是一个多方面的挑战，需要不同学科、领域和社会利益相关方的合作。随着减缓气候变化的紧迫性与日俱增，亟需在交流和教育方面采取创新方法，以加强社会的理解和参与。像 ChatGPT 这样的大规模语言模型作为人工智能时代的变革性工具应运而生，有望彻底改变我们处理实现碳中和的经济、技术、社会和环境问题的方式。然而，这些模型在实现碳中和方面的全部潜力还有待发挥，其局限性在于无法在广泛的主题范围内提供详细、本地化和专家级的见解。为了弥补这些不足，本文介绍了一个创新框架，该框架将本地知识与本地土地利用模型相结合，旨在显著提高所提供信息的深度、准确性和区域相关性。本文从政府、企业和社区的角度探讨了这一框架的有效性。将本地知识与当地法律信息结合起来，不仅能丰富人工智能对当地特殊性的理解，还能确保提供最新信息，而这些信息对于解决广大利益相关者提出的有关碳中和的具体关切和问题至关重要。总之，建议的框架在加强社会对碳中和的理解和参与方面展示了巨大的潜力。

{"title":"Integrating local knowledge with ChatGPT-like large-scale language models for enhanced societal comprehension of carbon neutrality","authors":"Te Han, Rong-Gang Cong, Biying Yu, Baojun Tang, Yi-Ming Wei","doi":"10.1016/j.egyai.2024.100440","DOIUrl":"10.1016/j.egyai.2024.100440","url":null,"abstract":"<div><div>Addressing carbon neutrality presents a multifaceted challenge, necessitating collaboration across various disciplines, fields, and societal stakeholders. With the increasing urgency to mitigate climate change, there is a crucial need for innovative approaches in communication and education to enhance societal understanding and engagement. Large-scale language models like ChatGPT emerge as transformative tools in the AI era, offering potential to revolutionize how we approach economic, technological, social, and environmental issues of achieving carbon neutrality. However, the full potential of these models in carbon neutrality is yet to be realized, hindered by limitations in providing detailed, localized, and expert-level insights across an expansive spectrum of subjects. To bridge these gaps, this paper introduces an innovative framework that integrates local knowledge with LLMs, aiming to markedly enhance the depth, accuracy, and regional relevance of the information provided. The effectiveness of this framework is examined from government, corporations, and community perspectives. The integration of local knowledge with LLMs not only enriches the AI’s comprehension of local specificities but also guarantees an up-to-date information that is crucial for addressing the specific concerns and questions about carbon neutrality raised by a broad array of stakeholders. Overall, the proposed framework showcases significant potential in enhancing societal comprehension and participation towards carbon neutrality.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"18 ","pages":"Article 100440"},"PeriodicalIF":9.6,"publicationDate":"2024-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimization of a Bayesian game for Peer-to-Peer trading among prosumers under incomplete information via a CNN-LSTM-ATT 通过 CNN-LSTM-ATT，优化不完全信息条件下的贝叶斯博弈，促进消费者之间的点对点交易

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-11-06 DOI: 10.1016/j.egyai.2024.100437

Hongjie Jia , Wanxin Tang , Xiaolong Jin , Yunfei Mu , Dengxin Ai , Xiaodan Yu , Wei Wei

In modern low-carbon industrial parks, various distributed renewable energy resources are employed to fulfill production needs. Despite the growing capacity of renewable energy generation, a significant portion of the power produced by these renewable resources remains unconsumed, resulting in a waste of resources. Within an industrial park, microgrids that both generate and consume energy resources act as energy prosumers. Peer-to-peer (P2P) trading provides an efficient means of utilizing renewable energy among these energy prosumers, who possess both power generation and consumption capabilities. However, within the current market mechanism, each prosumer retains private information that is not disclosed on the network. To address the issue of incomplete information among multiple prosumers during the decision-making process, we develop a Bayesian game model based on the CNN-LSTM-ATT prediction method for P2P electricity transactions among multiple prosumers. The energy prosumers in each industrial park aim to minimize their energy consumption costs by adjusting strategies that include P2P energy trading and managing thermal loads. Prosumers make decisions on the basis of their own characteristics and estimates of other prosumer characteristics, which are obtained from the joint probability distribution predicted by the CNN-LSTM-ATT method. These decisions are aimed at minimizing each prosumer's electricity costs. The simulation results demonstrate the effectiveness of the Bayesian game model proposed in this study.

在现代低碳工业园区中，各种分布式可再生能源被用来满足生产需求。尽管可再生能源发电能力不断提高，但这些可再生能源产生的电能仍有很大一部分未被消耗，造成资源浪费。在工业园区内，既产生又消耗能源资源的微电网充当了能源消费者的角色。点对点（P2P）交易为这些同时具备发电和消费能力的能源消费商提供了有效利用可再生能源的手段。然而，在当前的市场机制中，每个能源消费者都保留着不在网络上公开的私人信息。为了解决决策过程中多个能源消费者之间信息不完整的问题，我们开发了一种基于 CNN-LSTM-ATT 预测方法的贝叶斯博弈模型，用于多个能源消费者之间的 P2P 电力交易。每个工业园区的能源消费商都希望通过调整策略（包括 P2P 能源交易和热负荷管理）最大限度地降低能耗成本。消费者根据自身特征和对其他消费者特征的估计做出决策，这些特征来自 CNN-LSTM-ATT 方法预测的联合概率分布。这些决策旨在最大限度地降低每个消费者的电费。模拟结果证明了本研究提出的贝叶斯博弈模型的有效性。

{"title":"Optimization of a Bayesian game for Peer-to-Peer trading among prosumers under incomplete information via a CNN-LSTM-ATT","authors":"Hongjie Jia , Wanxin Tang , Xiaolong Jin , Yunfei Mu , Dengxin Ai , Xiaodan Yu , Wei Wei","doi":"10.1016/j.egyai.2024.100437","DOIUrl":"10.1016/j.egyai.2024.100437","url":null,"abstract":"<div><div>In modern low-carbon industrial parks, various distributed renewable energy resources are employed to fulfill production needs. Despite the growing capacity of renewable energy generation, a significant portion of the power produced by these renewable resources remains unconsumed, resulting in a waste of resources. Within an industrial park, microgrids that both generate and consume energy resources act as energy prosumers. Peer-to-peer (P2P) trading provides an efficient means of utilizing renewable energy among these energy prosumers, who possess both power generation and consumption capabilities. However, within the current market mechanism, each prosumer retains private information that is not disclosed on the network. To address the issue of incomplete information among multiple prosumers during the decision-making process, we develop a Bayesian game model based on the CNN-LSTM-ATT prediction method for P2P electricity transactions among multiple prosumers. The energy prosumers in each industrial park aim to minimize their energy consumption costs by adjusting strategies that include P2P energy trading and managing thermal loads. Prosumers make decisions on the basis of their own characteristics and estimates of other prosumer characteristics, which are obtained from the joint probability distribution predicted by the CNN-LSTM-ATT method. These decisions are aimed at minimizing each prosumer's electricity costs. The simulation results demonstrate the effectiveness of the Bayesian game model proposed in this study.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"18 ","pages":"Article 100437"},"PeriodicalIF":9.6,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Parameter sensitivity analysis for diesel spray penetration prediction based on GA-BP neural network 基于 GA-BP 神经网络的柴油机喷雾渗透率预测参数敏感性分析

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-11-05 DOI: 10.1016/j.egyai.2024.100443

Yifei Zhang , Gengxin Zhang , Dawei Wu , Qian Wang , Ebrahim Nadimi , Penghua Shi , Hongming Xu

Machine learning has started to be used in engine research to optimize combustion and predict fuel spray characteristics. This paper presents the development of a machine learning model using a Genetic Algorithm-Backpropagation (GA-BP) neural network to predict spray penetration. The GA-BP neural network was selected for its ability to optimize neural network weights and thresholds, thereby improving model convergence and avoiding local minima, which are common challenges in complex, non-linear problems such as spray prediction. The model was trained using experimental data from diesel injector spray tests, and its accuracy was evaluated through parametric sensitivity analysis, examining the influence of various input factors. A comparison between the machine learning model and the traditional empirical formulas of spray penetration revealed that the machine learning model achieved greater accuracy. In terms of the sensitivity to inputs, it is interesting to find that the cognition of machines is different from that of humans. When an input parameter does not have any functional relationship with other input parameters, the absence of this input parameter will lead to a significant decrease in the accuracy of the output result. The results demonstrate that the machine learning approach offers higher accuracy and better generalizability compared to traditional empirical methods. This study recommends the ways to get better results of penetration prediction with BP neural networks, which is efficient in training and utilizing Artificial Neural Networks (ANNs).

机器学习已开始用于发动机研究，以优化燃烧和预测燃料喷射特性。本文介绍了利用遗传算法-反向传播（GA-BP）神经网络开发的机器学习模型，用于预测喷射穿透性。之所以选择 GA-BP 神经网络，是因为它能够优化神经网络权重和阈值，从而提高模型的收敛性并避免局部最小值，而局部最小值是喷雾预测等复杂非线性问题中常见的难题。利用柴油喷射器喷雾测试的实验数据对模型进行了训练，并通过参数敏感性分析评估了模型的准确性，检查了各种输入因素的影响。通过比较机器学习模型和传统的喷雾渗透经验公式，发现机器学习模型的准确性更高。在对输入的敏感性方面，有趣的是，机器的认知与人类不同。当一个输入参数与其他输入参数没有任何功能关系时，缺少这个输入参数将导致输出结果的准确性大大降低。结果表明，与传统的经验方法相比，机器学习方法具有更高的准确性和更好的普适性。本研究推荐了利用 BP 神经网络获得更好的渗透预测结果的方法，该方法可有效地训练和利用人工神经网络（ANN）。

{"title":"Parameter sensitivity analysis for diesel spray penetration prediction based on GA-BP neural network","authors":"Yifei Zhang , Gengxin Zhang , Dawei Wu , Qian Wang , Ebrahim Nadimi , Penghua Shi , Hongming Xu","doi":"10.1016/j.egyai.2024.100443","DOIUrl":"10.1016/j.egyai.2024.100443","url":null,"abstract":"<div><div>Machine learning has started to be used in engine research to optimize combustion and predict fuel spray characteristics. This paper presents the development of a machine learning model using a Genetic Algorithm-Backpropagation (GA-BP) neural network to predict spray penetration. The GA-BP neural network was selected for its ability to optimize neural network weights and thresholds, thereby improving model convergence and avoiding local minima, which are common challenges in complex, non-linear problems such as spray prediction. The model was trained using experimental data from diesel injector spray tests, and its accuracy was evaluated through parametric sensitivity analysis, examining the influence of various input factors. A comparison between the machine learning model and the traditional empirical formulas of spray penetration revealed that the machine learning model achieved greater accuracy. In terms of the sensitivity to inputs, it is interesting to find that the cognition of machines is different from that of humans. When an input parameter does not have any functional relationship with other input parameters, the absence of this input parameter will lead to a significant decrease in the accuracy of the output result. The results demonstrate that the machine learning approach offers higher accuracy and better generalizability compared to traditional empirical methods. This study recommends the ways to get better results of penetration prediction with BP neural networks, which is efficient in training and utilizing Artificial Neural Networks (ANNs).</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"18 ","pages":"Article 100443"},"PeriodicalIF":9.6,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimizing catalyst layer composition of PEM fuel cell via machine learning: Insights from in-house experimental data 通过机器学习优化 PEM 燃料电池催化剂层的组成：内部实验数据的启示

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-11-01 DOI: 10.1016/j.egyai.2024.100439

Yuze Hou, Patrick Schneider, Linda Ney, Nada Zamel

The catalyst layer (CL) is a pivotal component of Proton Exchange Membrane (PEM) fuel cells, exerting a significant impact on both performance and durability. Its ink composition can be succinctly characterized by platinum (Pt) loading, Pt/carbon ratio, and ionomer/carbon ratio. The amount of each substance within the CL must be meticulously balanced to achieve optimal operation. In this work, we apply an Artificial Neural Network (ANN) model to forecast the performance and durability of a PEM fuel cell based on its cathode CL composition. The model is trained and validated based on experimental data measured at our laboratories, which consist of data from 49 fuel cells, detailing their cathode CL composition, operating conditions, accelerated stress test conditions, polarization curves and ECSA measurements throughout their lifespan. The presented ANN model demonstrates exceptional reliability in predicting PEM fuel cell behavior for both beginning and end of life. This allows for a deeper understanding of the influence of each input on performance and durability. Furthermore, the model can be effectively applied to optimize the CL composition. This paper demonstrates the immense potential of AI, combined with a high-quality database, to advance fuel cell research.

催化剂层（CL）是质子交换膜（PEM）燃料电池的重要组成部分，对性能和耐用性都有重大影响。其油墨成分可通过铂（Pt）含量、铂/碳比率和离子聚合物/碳比率来简明描述。CL 中每种物质的含量都必须精确平衡，以达到最佳运行状态。在这项工作中，我们应用人工神经网络（ANN）模型，根据阴极 CL 的成分预测 PEM 燃料电池的性能和耐用性。该模型是根据我们实验室测量的实验数据进行训练和验证的，其中包括 49 个燃料电池的数据，详细说明了其阴极 CL 成分、运行条件、加速应力测试条件、极化曲线和整个寿命期间的 ECSA 测量结果。所介绍的 ANN 模型在预测 PEM 燃料电池寿命开始和结束时的行为方面都表现出了极高的可靠性。这样就能更深入地了解每项输入对性能和耐用性的影响。此外，该模型还能有效地用于优化 CL 成分。本文展示了人工智能与高质量数据库相结合在推动燃料电池研究方面的巨大潜力。

{"title":"Optimizing catalyst layer composition of PEM fuel cell via machine learning: Insights from in-house experimental data","authors":"Yuze Hou, Patrick Schneider, Linda Ney, Nada Zamel","doi":"10.1016/j.egyai.2024.100439","DOIUrl":"10.1016/j.egyai.2024.100439","url":null,"abstract":"<div><div>The catalyst layer (CL) is a pivotal component of Proton Exchange Membrane (PEM) fuel cells, exerting a significant impact on both performance and durability. Its ink composition can be succinctly characterized by platinum (Pt) loading, Pt/carbon ratio, and ionomer/carbon ratio. The amount of each substance within the CL must be meticulously balanced to achieve optimal operation. In this work, we apply an Artificial Neural Network (ANN) model to forecast the performance and durability of a PEM fuel cell based on its cathode CL composition. The model is trained and validated based on experimental data measured at our laboratories, which consist of data from 49 fuel cells, detailing their cathode CL composition, operating conditions, accelerated stress test conditions, polarization curves and ECSA measurements throughout their lifespan. The presented ANN model demonstrates exceptional reliability in predicting PEM fuel cell behavior for both beginning and end of life. This allows for a deeper understanding of the influence of each input on performance and durability. Furthermore, the model can be effectively applied to optimize the CL composition. This paper demonstrates the immense potential of AI, combined with a high-quality database, to advance fuel cell research.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"18 ","pages":"Article 100439"},"PeriodicalIF":9.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142592934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Machine learning based parameter estimation for an adapted finite element model of a blade bearing test bench 基于机器学习的叶片轴承试验台改良有限元模型参数估计

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-10-29 DOI: 10.1016/j.egyai.2024.100436

Luca Faller , Matthis Graßmann , Timo Lichtenstein

Improving the reliability of blade bearings is essential for the safe operation of wind turbines. This challenge can be met with the help of virtual testing and digital-twin driven condition monitoring. For such approaches, a precise digital representation of the blade bearing and its test bench is an essential prerequisite. However, various factors prevent the capture of all parameters of the blade bearing and the associated test bench. Parameters such as bearing preload, rolling element and raceway dimensions, and bolt preload during assembly vary with each bearing and test bench setup. As these parameters cannot be measured directly, an alternative solution is required. This article presents a methodology to efficiently estimate non-measurable parameters of the test bench using a combination of model-based and data-driven approaches, improving the detailed and accurate virtual testing of blade bearings. It must be ensured to enable the fastest possible, most computationally efficient estimation of parameters during virtual testing or condition monitoring. The developed methodology is evaluated using the example of bolt preload on the test bench. By employing a random forest model and the strain gauge measurements attached to the blade bearing, the bolt preload parameters are estimated. The results demonstrate that the accuracy of the digital model of the blade bearing test bench is improved by up to 11 % in three out of four test bench setups. The great improvement in the accuracy of the digital model highlights the effectiveness of the proposed methodology in enhancing virtual blade bearing testing and digital-twin driven condition monitoring.

提高叶片轴承的可靠性对于风力涡轮机的安全运行至关重要。这一挑战可以借助虚拟测试和数字双驱动状态监测来应对。对于此类方法，精确的叶片轴承及其测试台的数字表示是必不可少的先决条件。然而，各种因素阻碍了对叶片轴承及其相关测试台所有参数的捕捉。轴承预紧力、滚动体和滚道尺寸以及装配过程中的螺栓预紧力等参数因每个轴承和测试台的设置而异。由于这些参数无法直接测量，因此需要一种替代解决方案。本文介绍了一种方法，利用基于模型和数据驱动相结合的方法，有效估算测试台的不可测量参数，从而提高叶片轴承虚拟测试的详细性和准确性。在虚拟测试或状态监测过程中，必须确保以最快的速度、最有效的计算方法估算参数。以测试台上的螺栓预紧力为例，对所开发的方法进行了评估。通过采用随机森林模型和叶片轴承上的应变仪测量值，对螺栓预紧力参数进行了估算。结果表明，叶片轴承测试台数字模型的精确度在四个测试台设置中的三个提高了 11%。数字模型精确度的大幅提高凸显了所提方法在加强虚拟叶片轴承测试和数字双驱动状态监测方面的有效性。

{"title":"Machine learning based parameter estimation for an adapted finite element model of a blade bearing test bench","authors":"Luca Faller , Matthis Graßmann , Timo Lichtenstein","doi":"10.1016/j.egyai.2024.100436","DOIUrl":"10.1016/j.egyai.2024.100436","url":null,"abstract":"<div><div>Improving the reliability of blade bearings is essential for the safe operation of wind turbines. This challenge can be met with the help of virtual testing and digital-twin driven condition monitoring. For such approaches, a precise digital representation of the blade bearing and its test bench is an essential prerequisite. However, various factors prevent the capture of all parameters of the blade bearing and the associated test bench. Parameters such as bearing preload, rolling element and raceway dimensions, and bolt preload during assembly vary with each bearing and test bench setup. As these parameters cannot be measured directly, an alternative solution is required. This article presents a methodology to efficiently estimate non-measurable parameters of the test bench using a combination of model-based and data-driven approaches, improving the detailed and accurate virtual testing of blade bearings. It must be ensured to enable the fastest possible, most computationally efficient estimation of parameters during virtual testing or condition monitoring. The developed methodology is evaluated using the example of bolt preload on the test bench. By employing a random forest model and the strain gauge measurements attached to the blade bearing, the bolt preload parameters are estimated. The results demonstrate that the accuracy of the digital model of the blade bearing test bench is improved by up to 11 % in three out of four test bench setups. The great improvement in the accuracy of the digital model highlights the effectiveness of the proposed methodology in enhancing virtual blade bearing testing and digital-twin driven condition monitoring.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"18 ","pages":"Article 100436"},"PeriodicalIF":9.6,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel optimization framework for natural gas transportation pipeline networks based on deep reinforcement learning 基于深度强化学习的天然气运输管道网络新型优化框架

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-10-29 DOI: 10.1016/j.egyai.2024.100434

Zemin Eitan Liu , Wennan Long , Zhenlin Chen , James Littlefield , Liang Jing , Bo Ren , Hassan M. El-Houjeiri , Amjaad S. Qahtani , Muhammad Y. Jabbar , Mohammad S. Masnadi

Natural gas is an emerging and reliable energy source in transition to a low-carbon economy. The natural gas transportation pipeline network systems are crucial when transporting natural gas from the production endpoints to processing or consuming endpoints. Optimizing the operational efficiency of compressor stations within pipeline networks is an effective way to reduce energy consumption and carbon emissions during transportation. This paper proposes an optimization framework for natural gas transportation pipeline networks based on deep reinforcement learning (DRL). The mathematical simulation model is derived from mass balance, hydrodynamics principles of gas flow, and compressor characteristics. The optimization control problem in steady state is formulated into a one-step Markov decision process (MDP) and solved by DRL. The decision variables are selected as the discharge ratio of each compressor. By the comprehensive comparison with dynamic programming (DP) and genetic algorithm (GA) in three typical element topologies (a linear topology with gun-barrel structure, a linear topology with branch structure, and a tree topology), the proposed method can obtain 4.60% lower power consumption than GA, and the time consumption is reduced by 97.5% compared with DP. The proposed framework could be further utilized for future large-scale network optimization practices.

在向低碳经济转型的过程中，天然气是一种新兴的可靠能源。将天然气从生产终端输送到加工或消费终端，天然气运输管网系统至关重要。优化管网中压缩机站的运行效率是减少运输过程中能源消耗和碳排放的有效途径。本文提出了一种基于深度强化学习（DRL）的天然气运输管网优化框架。数学模拟模型源于质量平衡、气体流动的流体力学原理和压缩机特性。稳态优化控制问题被表述为一步马尔可夫决策过程（MDP），并通过 DRL 进行求解。决策变量选择为每台压缩机的排气比。通过与动态编程（DP）和遗传算法（GA）在三种典型元件拓扑结构（炮筒结构线性拓扑、分支结构线性拓扑和树形拓扑）中的综合比较，所提出的方法比 GA 的功耗低 4.60%，比 DP 的时间消耗减少 97.5%。提出的框架可进一步用于未来的大规模网络优化实践。

{"title":"A novel optimization framework for natural gas transportation pipeline networks based on deep reinforcement learning","authors":"Zemin Eitan Liu , Wennan Long , Zhenlin Chen , James Littlefield , Liang Jing , Bo Ren , Hassan M. El-Houjeiri , Amjaad S. Qahtani , Muhammad Y. Jabbar , Mohammad S. Masnadi","doi":"10.1016/j.egyai.2024.100434","DOIUrl":"10.1016/j.egyai.2024.100434","url":null,"abstract":"<div><div>Natural gas is an emerging and reliable energy source in transition to a low-carbon economy. The natural gas transportation pipeline network systems are crucial when transporting natural gas from the production endpoints to processing or consuming endpoints. Optimizing the operational efficiency of compressor stations within pipeline networks is an effective way to reduce energy consumption and carbon emissions during transportation. This paper proposes an optimization framework for natural gas transportation pipeline networks based on deep reinforcement learning (DRL). The mathematical simulation model is derived from mass balance, hydrodynamics principles of gas flow, and compressor characteristics. The optimization control problem in steady state is formulated into a one-step Markov decision process (MDP) and solved by DRL. The decision variables are selected as the discharge ratio of each compressor. By the comprehensive comparison with dynamic programming (DP) and genetic algorithm (GA) in three typical element topologies (a linear topology with gun-barrel structure, a linear topology with branch structure, and a tree topology), the proposed method can obtain 4.60% lower power consumption than GA, and the time consumption is reduced by 97.5% compared with DP. The proposed framework could be further utilized for future large-scale network optimization practices.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"18 ","pages":"Article 100434"},"PeriodicalIF":9.6,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142573484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A two-layer low-carbon economic planning method for park-level integrated energy systems with carbon-energy synergistic hub 具有碳-能协同枢纽的园区级综合能源系统的双层低碳经济规划方法

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-10-28 DOI: 10.1016/j.egyai.2024.100435

Yunfei Mu , Haochen Guo , Zhijun Wu , Hongjie Jia , Xiaolong Jin , Yan Qi

Building a low-carbon park is crucial for achieving the carbon neutrality goals. However, most research on low-carbon economic planning methods for park-level integrated energy systems (PIES) has focused on multi-energy flow interactions, neglecting the “carbon perspective” and the impact of the dynamic coupling characteristics between multi-energy flows and carbon emission flow (CEF) on carbon reduction and planning schemes. Therefore, this paper proposes a two-layer low-carbon economic planning method for park-level integrated energy systems with carbon-energy synergistic hub (CESH). Firstly, this paper establishes a CESH model for PIES to describe the synergistic relationship between CEF and multi-energy flows from input, conversion, storage, to output. Secondly, a PIES two-layer low-carbon economic planning model with CESH is proposed. The upper model determines the optimal device types and capacities during the planning cycle. The lower model employs the CESH model to promote carbon energy friendly interactions, optimize the daily operation scheme of PIES. The iterative process between the two layers, initiated by a genetic algorithm (GA), ensures the speed and accuracy. Finally, case studies show that, compared to planning methods without the CESH model, the proposed method is effective in reducing carbon emissions and total costs during the planning cycle. From a dual “carbon-energy” perspective, it enhances investment effectiveness and carbon reduction sensitivity by deeply exploring the energy conservation and carbon reduction potential of PIES.

建设低碳园区对于实现碳中和目标至关重要。然而，关于园区级综合能源系统（PIES）低碳经济规划方法的研究大多集中在多能源流的相互作用上，忽视了 "碳视角 "以及多能源流与碳排放流（CEF）之间的动态耦合特性对碳减排和规划方案的影响。因此，本文提出了碳能协同枢纽（CESH）的园区级综合能源系统双层低碳经济规划方法。首先，本文建立了 PIES 的 CESH 模型，以描述 CEF 与多能源流之间从输入、转换、存储到输出的协同关系。其次，本文提出了具有 CESH 的 PIES 双层低碳经济规划模型。上层模型确定规划周期内的最佳设备类型和容量。下层模型利用 CESH 模型促进碳能源友好互动，优化 PIES 的日常运行方案。两层模型之间的迭代过程由遗传算法（GA）启动，确保了速度和准确性。最后，案例研究表明，与不使用 CESH 模型的规划方法相比，所提出的方法能有效减少规划周期内的碳排放量和总成本。从 "碳-能 "双重角度来看，该方法通过深入挖掘 PIES 的节能和减碳潜力，提高了投资效益和减碳敏感性。

{"title":"A two-layer low-carbon economic planning method for park-level integrated energy systems with carbon-energy synergistic hub","authors":"Yunfei Mu , Haochen Guo , Zhijun Wu , Hongjie Jia , Xiaolong Jin , Yan Qi","doi":"10.1016/j.egyai.2024.100435","DOIUrl":"10.1016/j.egyai.2024.100435","url":null,"abstract":"<div><div>Building a low-carbon park is crucial for achieving the carbon neutrality goals. However, most research on low-carbon economic planning methods for park-level integrated energy systems (PIES) has focused on multi-energy flow interactions, neglecting the “carbon perspective” and the impact of the dynamic coupling characteristics between multi-energy flows and carbon emission flow (CEF) on carbon reduction and planning schemes. Therefore, this paper proposes a two-layer low-carbon economic planning method for park-level integrated energy systems with carbon-energy synergistic hub (CESH). Firstly, this paper establishes a CESH model for PIES to describe the synergistic relationship between CEF and multi-energy flows from input, conversion, storage, to output. Secondly, a PIES two-layer low-carbon economic planning model with CESH is proposed. The upper model determines the optimal device types and capacities during the planning cycle. The lower model employs the CESH model to promote carbon energy friendly interactions, optimize the daily operation scheme of PIES. The iterative process between the two layers, initiated by a genetic algorithm (GA), ensures the speed and accuracy. Finally, case studies show that, compared to planning methods without the CESH model, the proposed method is effective in reducing carbon emissions and total costs during the planning cycle. From a dual “carbon-energy” perspective, it enhances investment effectiveness and carbon reduction sensitivity by deeply exploring the energy conservation and carbon reduction potential of PIES.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"18 ","pages":"Article 100435"},"PeriodicalIF":9.6,"publicationDate":"2024-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring public attention in the circular economy through topic modelling with twin hyperparameter optimisation 通过双参数优化的主题建模探索循环经济中的公众关注度

IF 9.6 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Energy and AI

Pub Date : 2024-10-22 DOI: 10.1016/j.egyai.2024.100433

Junhao Song , Yingfang Yuan , Kaiwen Chang , Bing Xu , Jin Xuan , Wei Pang

To advance the circular economy (CE), it is crucial to gain insights into the evolution of public attention, cognitive pathways related to circular products, and key public concerns. To achieve these objectives, we collected data from diverse platforms, including Twitter, Reddit, and The Guardian, and utilised three topic models to analyse the data. Given the performance of topic modelling may vary depending on hyperparameter settings, we proposed a novel framework that integrates twin (single- and multi-objective) hyperparameter optimisation for CE analysis. Systematic experiments were conducted to determine appropriate hyperparameters under different constraints, providing valuable insights into the correlations between CE and public attention. Our findings reveal that economic implications of sustainability and circular practices, particularly around recyclable materials and environmentally sustainable technologies, remain a significant public concern. Topics related to sustainable development and environmental protection technologies are particularly prominent on The Guardian, while Twitter discussions are comparatively sparse. These insights highlight the importance of targeted education programmes, business incentives adopt CE practices, and stringent waste management policies alongside improved recycling processes.

为了推动循环经济（CE）的发展，深入了解公众注意力的演变、与循环产品相关的认知路径以及公众的主要关注点至关重要。为了实现这些目标，我们从 Twitter、Reddit 和《卫报》等不同平台收集数据，并利用三种话题模型对数据进行分析。鉴于话题模型的性能可能因超参数设置的不同而有所差异，我们提出了一个新颖的框架，将双（单目标和多目标）超参数优化整合到消费电子分析中。我们进行了系统性实验，以确定不同约束条件下的适当超参数，从而为了解行政首长协调会与公众关注度之间的相关性提供了有价值的见解。我们的研究结果表明，可持续发展和循环实践的经济影响，特别是围绕可回收材料和环境可持续技术的经济影响，仍然是公众关注的一个重要问题。与可持续发展和环保技术相关的话题在《卫报》上尤为突出，而推特上的讨论则相对稀少。这些见解凸显了有针对性的教育计划、商业激励措施、严格的废物管理政策以及改进的回收流程的重要性。

{"title":"Exploring public attention in the circular economy through topic modelling with twin hyperparameter optimisation","authors":"Junhao Song , Yingfang Yuan , Kaiwen Chang , Bing Xu , Jin Xuan , Wei Pang","doi":"10.1016/j.egyai.2024.100433","DOIUrl":"10.1016/j.egyai.2024.100433","url":null,"abstract":"<div><div>To advance the circular economy (CE), it is crucial to gain insights into the evolution of public attention, cognitive pathways related to circular products, and key public concerns. To achieve these objectives, we collected data from diverse platforms, including Twitter, Reddit, and The Guardian, and utilised three topic models to analyse the data. Given the performance of topic modelling may vary depending on hyperparameter settings, we proposed a novel framework that integrates twin (single- and multi-objective) hyperparameter optimisation for CE analysis. Systematic experiments were conducted to determine appropriate hyperparameters under different constraints, providing valuable insights into the correlations between CE and public attention. Our findings reveal that economic implications of sustainability and circular practices, particularly around recyclable materials and environmentally sustainable technologies, remain a significant public concern. Topics related to sustainable development and environmental protection technologies are particularly prominent on The Guardian, while Twitter discussions are comparatively sparse. These insights highlight the importance of targeted education programmes, business incentives adopt CE practices, and stringent waste management policies alongside improved recycling processes.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"18 ","pages":"Article 100433"},"PeriodicalIF":9.6,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142573483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0