Yichen Jiang, Bowen Zhou, Guangdi Li, Yanhong Luo, Bo Hu, Yubo Liu
In order to address the challenges posed by the integration of regional electric vehicle (EV) clusters into the grid, it is crucial to fully utilize the scheduling capabilities of EVs. In this study, to investigate the energy storage characteristics of EVs, we first established a single EV virtual energy storage (EVVES) model based on the energy storage characteristics of EVs. We then further integrated four types of EVs within the region to form EV clusters (EVCs) and constructed an EVC virtual energy storage (VES) model to obtain the dynamic charging and discharging boundaries of the EVCs. Next, based on the dispatch framework for the participation of renewable energy sources (RESs) and loads in the distribution network, we established a dual-objective optimization dispatch model, with the objectives of minimizing system operating costs and load fluctuations. We solved this model with NSGA-II and TOPSIS, which guided and optimized the charging and discharging of EVCs. Finally, the simulation results show that the system operating cost was reduced by 7.81%, and the peak-to-valley difference of the load was reduced by 3.83% after optimization. The system effectively achieves load peak shaving and valley filling, improving economic efficiency.
{"title":"Virtual Energy Storage-Based Charging and Discharging Strategy for Electric Vehicle Clusters","authors":"Yichen Jiang, Bowen Zhou, Guangdi Li, Yanhong Luo, Bo Hu, Yubo Liu","doi":"10.3390/wevj15080359","DOIUrl":"https://doi.org/10.3390/wevj15080359","url":null,"abstract":"In order to address the challenges posed by the integration of regional electric vehicle (EV) clusters into the grid, it is crucial to fully utilize the scheduling capabilities of EVs. In this study, to investigate the energy storage characteristics of EVs, we first established a single EV virtual energy storage (EVVES) model based on the energy storage characteristics of EVs. We then further integrated four types of EVs within the region to form EV clusters (EVCs) and constructed an EVC virtual energy storage (VES) model to obtain the dynamic charging and discharging boundaries of the EVCs. Next, based on the dispatch framework for the participation of renewable energy sources (RESs) and loads in the distribution network, we established a dual-objective optimization dispatch model, with the objectives of minimizing system operating costs and load fluctuations. We solved this model with NSGA-II and TOPSIS, which guided and optimized the charging and discharging of EVCs. Finally, the simulation results show that the system operating cost was reduced by 7.81%, and the peak-to-valley difference of the load was reduced by 3.83% after optimization. The system effectively achieves load peak shaving and valley filling, improving economic efficiency.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"25 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141925282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yi Wei, Yanfeng Xing, Xiaobing Zhang, Ying Wang, Juyong Cao, Fuyong Yang
The sealing technology of proton exchange membrane fuel cells (PEMFCs) is a critical factor in ensuring their performance, impacting driving safety and range efficiency. To guarantee the safe operation of PEMFCs in complex environments, it is essential to conduct related sealing research. The structure of the fuel cell sealing system is complex, with components in close contact, and identifying factors that affect its sealing performance is crucial for the development and application of the cells. This paper briefly describes the sealing mechanism of PEMFCs and introduces four typical sealing structures. It considers both the assembly and operation processes, summarizing assembly errors, sealing gaskets, and sealing leaks as well as vibration, cyclic temperature and humidity, and cyclic assembly. The research status of the sealing system in simulations and experiments is reviewed in detail. The key factors affecting the sealing performance of fuel cells are emphasized, highlighting the significance of dynamic detection of the gasket status, stack performance improvement under cumulative errors, and multi-objective optimization models combining contact pressure with the characteristics of stack components.
{"title":"A Review of Sealing Systems for Proton Exchange Membrane Fuel Cells","authors":"Yi Wei, Yanfeng Xing, Xiaobing Zhang, Ying Wang, Juyong Cao, Fuyong Yang","doi":"10.3390/wevj15080358","DOIUrl":"https://doi.org/10.3390/wevj15080358","url":null,"abstract":"The sealing technology of proton exchange membrane fuel cells (PEMFCs) is a critical factor in ensuring their performance, impacting driving safety and range efficiency. To guarantee the safe operation of PEMFCs in complex environments, it is essential to conduct related sealing research. The structure of the fuel cell sealing system is complex, with components in close contact, and identifying factors that affect its sealing performance is crucial for the development and application of the cells. This paper briefly describes the sealing mechanism of PEMFCs and introduces four typical sealing structures. It considers both the assembly and operation processes, summarizing assembly errors, sealing gaskets, and sealing leaks as well as vibration, cyclic temperature and humidity, and cyclic assembly. The research status of the sealing system in simulations and experiments is reviewed in detail. The key factors affecting the sealing performance of fuel cells are emphasized, highlighting the significance of dynamic detection of the gasket status, stack performance improvement under cumulative errors, and multi-objective optimization models combining contact pressure with the characteristics of stack components.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"36 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141924794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The modern Internet of Autonomous Vehicles (IoVs) has enabled the development of autonomous vehicles that can interact with each other and their surroundings, facilitating real-time data exchange and communication between vehicles, infrastructure, and the external environment. The lack of security procedures in vehicular networks and Controller Area Network (CAN) protocol leaves vehicles exposed to intrusions. One common attack type is the message injection attack, which inserts fake messages into original Electronic Control Units (ECUs) to trick them or create failures. Therefore, this paper tackles the pressing issue of cyber-attack detection in modern IoV systems, where the increasing connectivity of vehicles to the external world and each other creates a vast attack surface. The vulnerability of in-vehicle networks, particularly the CAN protocol, makes them susceptible to attacks such as message injection, which can have severe consequences. To address this, we propose an intelligent Intrusion detection system (IDS) to detect a wide range of threats utilizing machine learning techniques. However, a significant challenge lies in the inherent imbalance of car-hacking datasets, which can lead to misclassification of attack types. To overcome this, we employ various imbalanced pre-processing techniques, including NearMiss, Random over-sampling (ROS), and TomLinks, to pre-process and handle imbalanced data. Then, various Machine Learning (ML) techniques, including Logistic Regression (LR), Linear Discriminant Analysis (LDA), Naive Bayes (NB), and K-Nearest Neighbors (k-NN), are employed in detecting and predicting attack types on balanced data. We evaluate the performance and efficacy of these techniques using a comprehensive set of evaluation metrics, including accuracy, precision, F1_Score, and recall. This demonstrates how well the suggested IDS detects cyberattacks in external and intra-vehicle vehicular networks using unbalanced data on vehicle hacking. Using k-NN with various resampling techniques, the results show that the proposed system achieves 100% detection rates in testing on the Car-Hacking dataset in comparison with existing work, demonstrating the effectiveness of our approach in protecting modern vehicle systems from advanced threats.
{"title":"An Intelligent Attack Detection Framework for the Internet of Autonomous Vehicles with Imbalanced Car Hacking Data","authors":"S. Alshathri, A. Sayed, E. E. Hemdan","doi":"10.3390/wevj15080356","DOIUrl":"https://doi.org/10.3390/wevj15080356","url":null,"abstract":"The modern Internet of Autonomous Vehicles (IoVs) has enabled the development of autonomous vehicles that can interact with each other and their surroundings, facilitating real-time data exchange and communication between vehicles, infrastructure, and the external environment. The lack of security procedures in vehicular networks and Controller Area Network (CAN) protocol leaves vehicles exposed to intrusions. One common attack type is the message injection attack, which inserts fake messages into original Electronic Control Units (ECUs) to trick them or create failures. Therefore, this paper tackles the pressing issue of cyber-attack detection in modern IoV systems, where the increasing connectivity of vehicles to the external world and each other creates a vast attack surface. The vulnerability of in-vehicle networks, particularly the CAN protocol, makes them susceptible to attacks such as message injection, which can have severe consequences. To address this, we propose an intelligent Intrusion detection system (IDS) to detect a wide range of threats utilizing machine learning techniques. However, a significant challenge lies in the inherent imbalance of car-hacking datasets, which can lead to misclassification of attack types. To overcome this, we employ various imbalanced pre-processing techniques, including NearMiss, Random over-sampling (ROS), and TomLinks, to pre-process and handle imbalanced data. Then, various Machine Learning (ML) techniques, including Logistic Regression (LR), Linear Discriminant Analysis (LDA), Naive Bayes (NB), and K-Nearest Neighbors (k-NN), are employed in detecting and predicting attack types on balanced data. We evaluate the performance and efficacy of these techniques using a comprehensive set of evaluation metrics, including accuracy, precision, F1_Score, and recall. This demonstrates how well the suggested IDS detects cyberattacks in external and intra-vehicle vehicular networks using unbalanced data on vehicle hacking. Using k-NN with various resampling techniques, the results show that the proposed system achieves 100% detection rates in testing on the Car-Hacking dataset in comparison with existing work, demonstrating the effectiveness of our approach in protecting modern vehicle systems from advanced threats.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"18 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141926283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kiran Raut, A. Shendge, Jagdish Chaudhari, Ravita Lamba, Tapas Mallick, Anurag Roy
Considering environmental concerns, electric vehicles (EVs) are gaining popularity over conventional internal combustion (IC) engine-based vehicles. Hybrid energy-storage systems (HESSs), comprising a combination of batteries and supercapacitors (SCs), are increasingly utilized in EVs. Such HESS-equipped EVs typically outperform standard electric vehicles. However, the effective management of power sources to meet varying power demands remains a major challenge in the hybrid electric vehicles. This study presents the development of a MATLAB Simulink model for a hybrid energy-storage system aimed at alleviating the load on batteries during periods of high power demand. Two parallel combinations are investigated: one integrating the battery with a supercapacitor and the other with a photovoltaic (PV) system. These configurations address challenges encountered in EVs, such as power fluctuations and battery longevity issues. Although lead- batteries are commonly used in conjunction with solar PV systems for energy storage, they incur higher operating costs due to the necessity of converters. The findings suggest that the proposed supercapacitor–battery configuration reduces battery peak power consumption by up to 39%. Consequently, the supercapacitor–battery HESS emerges as a superior option, possibly prolonging battery cycle life by mitigating stress induced by fluctuating power exchanges during the charging and discharging phases.
{"title":"Performance Analysis of Multiple Energy-Storage Devices Used in Electric Vehicles","authors":"Kiran Raut, A. Shendge, Jagdish Chaudhari, Ravita Lamba, Tapas Mallick, Anurag Roy","doi":"10.3390/wevj15080357","DOIUrl":"https://doi.org/10.3390/wevj15080357","url":null,"abstract":"Considering environmental concerns, electric vehicles (EVs) are gaining popularity over conventional internal combustion (IC) engine-based vehicles. Hybrid energy-storage systems (HESSs), comprising a combination of batteries and supercapacitors (SCs), are increasingly utilized in EVs. Such HESS-equipped EVs typically outperform standard electric vehicles. However, the effective management of power sources to meet varying power demands remains a major challenge in the hybrid electric vehicles. This study presents the development of a MATLAB Simulink model for a hybrid energy-storage system aimed at alleviating the load on batteries during periods of high power demand. Two parallel combinations are investigated: one integrating the battery with a supercapacitor and the other with a photovoltaic (PV) system. These configurations address challenges encountered in EVs, such as power fluctuations and battery longevity issues. Although lead- batteries are commonly used in conjunction with solar PV systems for energy storage, they incur higher operating costs due to the necessity of converters. The findings suggest that the proposed supercapacitor–battery configuration reduces battery peak power consumption by up to 39%. Consequently, the supercapacitor–battery HESS emerges as a superior option, possibly prolonging battery cycle life by mitigating stress induced by fluctuating power exchanges during the charging and discharging phases.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"103 9","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141926463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This study introduces an innovative Teleoperated Driving (ToD) system integrated with virtual twin technology using the MORAI simulator. The system minimizes the need for extensive video data transmission by utilizing text-based vehicle information, significantly reducing the communication load. Key technical advancements include the use of high-precision GNSS devices for accurate vehicle location tracking, robust data communication via the MQTT protocol, and the implementation of the Ego Ghost mode in the MORAI simulator for precise vehicle simulation. The integration of these technologies enables efficient data transmission and enhanced system reliability, effectively mitigating issues such as communication blackouts and delays. Our findings demonstrate that this approach ensures stable and efficient operation, optimizing communication resource management and enhancing operational stability, which is crucial for scenarios requiring high video quality and real-time response. This research represents a significant advancement in ToD technology, establishing a precedent for integrating virtual twin systems to create more resource-efficient and reliable autonomous driving backup solutions. The virtual twin-based ToD system provides a robust platform for remote vehicle operation, ensuring safety and reliability in various driving conditions.
本研究利用 MORAI 模拟器介绍了一种与虚拟孪生技术相结合的创新型远程驾驶（ToD）系统。该系统通过利用基于文本的车辆信息，最大限度地减少了对大量视频数据传输的需求，从而显著降低了通信负荷。主要的技术进步包括使用高精度 GNSS 设备进行精确的车辆位置跟踪，通过 MQTT 协议进行稳健的数据通信，以及在 MORAI 模拟器中实施 Ego Ghost 模式进行精确的车辆模拟。这些技术的集成实现了高效的数据传输和更高的系统可靠性，有效缓解了通信中断和延迟等问题。我们的研究结果表明，这种方法可确保稳定高效的运行，优化通信资源管理，提高运行稳定性，这对于要求高视频质量和实时响应的场景至关重要。这项研究代表了 ToD 技术的重大进步，开创了整合虚拟孪生系统以创建资源效率更高、更可靠的自动驾驶备份解决方案的先例。基于虚拟孪生的 ToD 系统为远程车辆操作提供了一个强大的平台，确保了在各种驾驶条件下的安全性和可靠性。
{"title":"Teleoperated Driving with Virtual Twin Technology: A Simulator-Based Approach","authors":"Keonil Kim, Seok-Cheol Kee","doi":"10.3390/wevj15070311","DOIUrl":"https://doi.org/10.3390/wevj15070311","url":null,"abstract":"This study introduces an innovative Teleoperated Driving (ToD) system integrated with virtual twin technology using the MORAI simulator. The system minimizes the need for extensive video data transmission by utilizing text-based vehicle information, significantly reducing the communication load. Key technical advancements include the use of high-precision GNSS devices for accurate vehicle location tracking, robust data communication via the MQTT protocol, and the implementation of the Ego Ghost mode in the MORAI simulator for precise vehicle simulation. The integration of these technologies enables efficient data transmission and enhanced system reliability, effectively mitigating issues such as communication blackouts and delays. Our findings demonstrate that this approach ensures stable and efficient operation, optimizing communication resource management and enhancing operational stability, which is crucial for scenarios requiring high video quality and real-time response. This research represents a significant advancement in ToD technology, establishing a precedent for integrating virtual twin systems to create more resource-efficient and reliable autonomous driving backup solutions. The virtual twin-based ToD system provides a robust platform for remote vehicle operation, ensuring safety and reliability in various driving conditions.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"2 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141640340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To increase the package space for the battery pack in the rear of battery electric vehicles (BEVs), and thus extend their driving range, a novel rear axle concept called the multi-link torsion axle (MLTA) has been developed. In this work, the kinematic design was extended with an elastokinematic concept, and the MLTA was designed in CAD and realized as a prototype. It was then integrated into a B-class series-production vehicle by adding masses in different locations of the vehicle to replicate the mass distribution of a BEV. Both objective and subjective vehicle dynamic evaluations were conducted, which included kinematic and compliance tests, constant-radius cornering, straight-line braking, and a frequency response test, as well as subjective evaluations by both expert and normal drivers. These test results were analyzed and compared to a production vehicle. It can be concluded that the vehicle dynamic performance of the MLTA-equipped vehicle is, overall, 0.67 grades lower than that of the comparable production vehicle on a 10-grade scale. According to OEM experts, this deficit can be eliminated by tuning the different components of the MLTA and meeting the tolerance requirements of series production vehicles.
{"title":"Evaluation of Vehicle Lateral and Longitudinal Dynamic Behavior of the New Package-Saving Multi-Link Torsion Axle (MLTA) for BEVs","authors":"Jens Olschewski, Xiangfan Fang","doi":"10.3390/wevj15070310","DOIUrl":"https://doi.org/10.3390/wevj15070310","url":null,"abstract":"To increase the package space for the battery pack in the rear of battery electric vehicles (BEVs), and thus extend their driving range, a novel rear axle concept called the multi-link torsion axle (MLTA) has been developed. In this work, the kinematic design was extended with an elastokinematic concept, and the MLTA was designed in CAD and realized as a prototype. It was then integrated into a B-class series-production vehicle by adding masses in different locations of the vehicle to replicate the mass distribution of a BEV. Both objective and subjective vehicle dynamic evaluations were conducted, which included kinematic and compliance tests, constant-radius cornering, straight-line braking, and a frequency response test, as well as subjective evaluations by both expert and normal drivers. These test results were analyzed and compared to a production vehicle. It can be concluded that the vehicle dynamic performance of the MLTA-equipped vehicle is, overall, 0.67 grades lower than that of the comparable production vehicle on a 10-grade scale. According to OEM experts, this deficit can be eliminated by tuning the different components of the MLTA and meeting the tolerance requirements of series production vehicles.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"41 23","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141649021","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we propose a vehicle pedestrian detection and tracking method based on the improved YOLOv7 and DeepSort algorithms. We aim to improve the quality of vehicle pedestrian detection and tracking, addressing the challenges that current commercially available autonomous driving technologies face in complex and changing road traffic situations. First, the NMS (non-maximum suppression) algorithm in YOLOv7 is replaced with a modified Soft-NMS algorithm to ensure that targets can be accurately detected at high densities, and second, the CCBA (coordinate channel attention module) attention mechanism is incorporated to improve the feature extraction and perception capabilities of the network. Finally, a multi-scale feature network is introduced to extract features of small targets more accurately. Finally, the MobileNetV3 lightweight module is introduced into the feature extraction network of DeepSort, which not only reduces the number of model parameters and network complexity, but also improves the tracking performance of the target. The experimental results show that the improved YOLOv7 algorithm improves the average detection accuracy by 3.77% compared to that of the original algorithm; on the MOT20 dataset, the refined DeepSort model achieves a 1.6% increase in MOTA and a 1.9% improvement in MOTP; in addition, the model volume is one-eighth of the original algorithm. In summary, our model is able to achieve the desired real-time and accuracy, which is more suitable for autonomous driving.
{"title":"CCBA-NMS-YD: A Vehicle Pedestrian Detection and Tracking Method Based on Improved YOLOv7 and DeepSort","authors":"Zhenhao Yuan, Zhiwen Wang, Ruonan Zhang","doi":"10.3390/wevj15070309","DOIUrl":"https://doi.org/10.3390/wevj15070309","url":null,"abstract":"In this paper, we propose a vehicle pedestrian detection and tracking method based on the improved YOLOv7 and DeepSort algorithms. We aim to improve the quality of vehicle pedestrian detection and tracking, addressing the challenges that current commercially available autonomous driving technologies face in complex and changing road traffic situations. First, the NMS (non-maximum suppression) algorithm in YOLOv7 is replaced with a modified Soft-NMS algorithm to ensure that targets can be accurately detected at high densities, and second, the CCBA (coordinate channel attention module) attention mechanism is incorporated to improve the feature extraction and perception capabilities of the network. Finally, a multi-scale feature network is introduced to extract features of small targets more accurately. Finally, the MobileNetV3 lightweight module is introduced into the feature extraction network of DeepSort, which not only reduces the number of model parameters and network complexity, but also improves the tracking performance of the target. The experimental results show that the improved YOLOv7 algorithm improves the average detection accuracy by 3.77% compared to that of the original algorithm; on the MOT20 dataset, the refined DeepSort model achieves a 1.6% increase in MOTA and a 1.9% improvement in MOTP; in addition, the model volume is one-eighth of the original algorithm. In summary, our model is able to achieve the desired real-time and accuracy, which is more suitable for autonomous driving.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"30 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141650357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Machine learning techniques have advanced rapidly, leading to better prediction accuracy within a short computational time. Such advancement encourages various novel applications, including in the field of operations research. This study introduces a novel way to utilize regression machine learning models to predict the objectives of vehicle routing problems that are solved using a genetic algorithm. Previous studies have generally discussed how (1) operations research methods are used independently to generate optimized solutions and (2) machine learning techniques are used independently to predict values from a given dataset. Some studies have discussed the collaborations between operations research and machine learning fields as follows: (1) using machine learning techniques to generate input data for operations research problems, (2) using operations research techniques to optimize the hyper-parameters of machine learning models, and (3) using machine learning to improve the quality of operations research algorithms. This study differs from the types of collaborative studies listed above. This study focuses on the prediction of the objective of the vehicle routing problem directly given the input and output data, without optimizing the problem using operations research algorithms. This study introduces a straightforward framework that captures the input data characteristics for the vehicle routing problem. The proposed framework is applied by generating the input and output data using the genetic algorithm and then using regression machine learning models to predict the obtained objective values. The numerical experiments show that the best models are random forest regression, a generalized linear model with a Poisson distribution, and ridge regression with cross-validation.
{"title":"Regression Machine Learning Models for the Short-Time Prediction of Genetic Algorithm Results in a Vehicle Routing Problem","authors":"Ivan Kristianto Singgih, M. Singgih","doi":"10.3390/wevj15070308","DOIUrl":"https://doi.org/10.3390/wevj15070308","url":null,"abstract":"Machine learning techniques have advanced rapidly, leading to better prediction accuracy within a short computational time. Such advancement encourages various novel applications, including in the field of operations research. This study introduces a novel way to utilize regression machine learning models to predict the objectives of vehicle routing problems that are solved using a genetic algorithm. Previous studies have generally discussed how (1) operations research methods are used independently to generate optimized solutions and (2) machine learning techniques are used independently to predict values from a given dataset. Some studies have discussed the collaborations between operations research and machine learning fields as follows: (1) using machine learning techniques to generate input data for operations research problems, (2) using operations research techniques to optimize the hyper-parameters of machine learning models, and (3) using machine learning to improve the quality of operations research algorithms. This study differs from the types of collaborative studies listed above. This study focuses on the prediction of the objective of the vehicle routing problem directly given the input and output data, without optimizing the problem using operations research algorithms. This study introduces a straightforward framework that captures the input data characteristics for the vehicle routing problem. The proposed framework is applied by generating the input and output data using the genetic algorithm and then using regression machine learning models to predict the obtained objective values. The numerical experiments show that the best models are random forest regression, a generalized linear model with a Poisson distribution, and ridge regression with cross-validation.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"41 18","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141649999","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The accuracy and real-time performance of 3D object detection are key factors limiting its widespread application. While cameras capture detailed color and texture features, they lack depth information compared to LiDAR. Multimodal detection combining both can improve results but incurs significant computational overhead, affecting real-time performance. To address these challenges, this paper presents a real-time multimodal fusion model called Fast Transfusion that combines the benefits of LiDAR and camera sensors and reduces the computational burden of their fusion. Specifically, our Fast Transfusion method uses QConv (Quick Convolution) to replace the convolutional backbones compared to other models. QConv concentrates the convolution operations at the feature map center, where the most information resides, to expedite inference. It also utilizes deformable convolution to better match the actual shapes of detected objects, enhancing accuracy. And the model incorporates EH Decoder (Efficient and Hybrid Decoder) which decouples multiscale fusion into intra-scale interaction and cross-scale fusion, efficiently decoding and integrating features extracted from multimodal data. Furthermore, our proposed semi-dynamic query selection refines the initialization of object queries. On the KITTI 3D object detection dataset, our proposed approach reduced the inference time by 36 ms and improved 3D AP by 1.81% compared to state-of-the-art methods.
三维物体检测的准确性和实时性是限制其广泛应用的关键因素。虽然相机能捕捉到详细的颜色和纹理特征，但与激光雷达相比，它们缺乏深度信息。结合两者的多模态检测可以提高检测结果，但会产生大量计算开销，影响实时性。为了应对这些挑战，本文提出了一种名为 "快速融合"（Fast Transfusion）的实时多模态融合模型，它结合了激光雷达和摄像头传感器的优势，并减轻了它们融合时的计算负担。具体来说，与其他模型相比，我们的快速融合方法使用 QConv（快速卷积）来替代卷积骨干。QConv 将卷积操作集中在信息量最大的特征图中心，以加快推理速度。它还利用可变形卷积来更好地匹配检测到的物体的实际形状，从而提高准确性。该模型采用了 EH 解码器（高效混合解码器），将多尺度融合分解为尺度内交互和跨尺度融合，从而高效地解码和整合从多模态数据中提取的特征。此外，我们还提出了半动态查询选择，改进了对象查询的初始化。在 KITTI 3D 物体检测数据集上，我们提出的方法与最先进的方法相比，推理时间减少了 36 毫秒，3D AP 提高了 1.81%。
{"title":"Real-Time Multimodal 3D Object Detection with Transformers","authors":"Hengsong Liu, Tongle Duan","doi":"10.3390/wevj15070307","DOIUrl":"https://doi.org/10.3390/wevj15070307","url":null,"abstract":"The accuracy and real-time performance of 3D object detection are key factors limiting its widespread application. While cameras capture detailed color and texture features, they lack depth information compared to LiDAR. Multimodal detection combining both can improve results but incurs significant computational overhead, affecting real-time performance. To address these challenges, this paper presents a real-time multimodal fusion model called Fast Transfusion that combines the benefits of LiDAR and camera sensors and reduces the computational burden of their fusion. Specifically, our Fast Transfusion method uses QConv (Quick Convolution) to replace the convolutional backbones compared to other models. QConv concentrates the convolution operations at the feature map center, where the most information resides, to expedite inference. It also utilizes deformable convolution to better match the actual shapes of detected objects, enhancing accuracy. And the model incorporates EH Decoder (Efficient and Hybrid Decoder) which decouples multiscale fusion into intra-scale interaction and cross-scale fusion, efficiently decoding and integrating features extracted from multimodal data. Furthermore, our proposed semi-dynamic query selection refines the initialization of object queries. On the KITTI 3D object detection dataset, our proposed approach reduced the inference time by 36 ms and improved 3D AP by 1.81% compared to state-of-the-art methods.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"21 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141652715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Qinguo Zhang, Xiaoyang Wang, Zheming Tong, Zhewu Cheng, Xiaojian Liu
The prominent problem of articulated steering structure of engineering vehicle is that there is pressure oscillation in the hydraulic system during steering, which seriously affects the performance of steering system. To solve this problem, the maximum stroke difference of left and right cylinders and the minimum maximum cylinder pressure are the optimization objectives, and the position of cylinder hinge point is the design variable. The multi-objective optimization design of articulated steering system is carried out by using the particle swarm optimization algorithm. After optimization, the maximum pressure of the steering system is reduced by 13.5%, and the oscillation amplitude is reduced by 16%, so the optimization effect is obvious. The dynamic characteristics of the hydraulic steering system under different loads, such as pressure and flow rate, are obtained through field steering tests of wheel loaders. The results show that the load has an important effect on the pressure response of the system, and the causes and influencing factors of pressure and flow fluctuation are determined. The relationship between mileage and hydrogen consumption is obtained, which provides data support for vehicle control strategy. The high-pressure overflow power consumption accounts for 60% of the total work, and the work lost on the steering gear reaches 36 kJ. The test results verify the rationality and correctness of the optimization method of steering mechanism and provide data support for the improvement in steering hydraulic system.
{"title":"Experimental Study on Structure Optimization and Dynamic Characteristics of Articulated Steering for Hydrogen Fuel Cell Engineering Vehicles","authors":"Qinguo Zhang, Xiaoyang Wang, Zheming Tong, Zhewu Cheng, Xiaojian Liu","doi":"10.3390/wevj15070306","DOIUrl":"https://doi.org/10.3390/wevj15070306","url":null,"abstract":"The prominent problem of articulated steering structure of engineering vehicle is that there is pressure oscillation in the hydraulic system during steering, which seriously affects the performance of steering system. To solve this problem, the maximum stroke difference of left and right cylinders and the minimum maximum cylinder pressure are the optimization objectives, and the position of cylinder hinge point is the design variable. The multi-objective optimization design of articulated steering system is carried out by using the particle swarm optimization algorithm. After optimization, the maximum pressure of the steering system is reduced by 13.5%, and the oscillation amplitude is reduced by 16%, so the optimization effect is obvious. The dynamic characteristics of the hydraulic steering system under different loads, such as pressure and flow rate, are obtained through field steering tests of wheel loaders. The results show that the load has an important effect on the pressure response of the system, and the causes and influencing factors of pressure and flow fluctuation are determined. The relationship between mileage and hydrogen consumption is obtained, which provides data support for vehicle control strategy. The high-pressure overflow power consumption accounts for 60% of the total work, and the work lost on the steering gear reaches 36 kJ. The test results verify the rationality and correctness of the optimization method of steering mechanism and provide data support for the improvement in steering hydraulic system.","PeriodicalId":507038,"journal":{"name":"World Electric Vehicle Journal","volume":"70 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141653130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：