Computation offloading is an effective strategy for leveraging thecomputing resources of edge-side devices, enhancing performance to swiftly respond to real-time computation tasks whileensuring successful completion...
详细信息
ISBN:
(纸本)9798350375794;9798350375800
Computation offloading is an effective strategy for leveraging thecomputing resources of edge-side devices, enhancing performance to swiftly respond to real-time computation tasks whileensuring successful completion using limited resources. This paper proposes an offloading method that takes into account the collaboration of multipleedge-side devices for real-time computation tasks in distribution networks. The data transmission process is analyzed in detail, while characteristics of real-time computation tasks, logic correlations between each split part of real-time computation tasks, and hardware resources for edgecomputing devices are considered as complex constraints. The proposed method reduces the overall completion delay and improves the successful completion rate of real-time tasks by splitting these tasks properly, transmitting computation data orderly, and allocating resources reasonably. Theeffectiveness of the proposed method is verified by a typical case.
Theescalating cyber-attacks targeting power infrastructure underscore the critical importance of smart grid security. However, existing solutions often struggle with the challenge of balancing security and performanc...
详细信息
ISBN:
(纸本)9798350318562;9798350318555
Theescalating cyber-attacks targeting power infrastructure underscore the critical importance of smart grid security. However, existing solutions often struggle with the challenge of balancing security and performance overhead, leading to suboptimal protection or increased operational latency. To address this, we propose an intrusion detection system (IDS) designed to operate within P4-based programmable network devices, enabling real-time identification of critical attacks like distributed denial-of-service (DDoS) and false data injection (FDI). Central to our approach is a novel data structure optimized for time series data, capturing key information such as packet timing and data payload distribution. Leveraging decision trees, a robust machine learning technique, enables effective anomaly detection and prediction. Additionally, we integrate data compression techniques to reduce device memory usage while maintaining detection accuracy. Our evaluation results demonstrate minimal overhead in packet processing speed with 1 to 20 nanoseconds differences per packet, and enhanced data storageefficiency with compression ratios reaching up to 60.9%. Despite these optimizations, there is only a slight decrease in detection accuracy, such as a 2.81% drop in detecting false data injection attack (FDIA).
With the rapid development of renewableenergy, photovoltaic power generation has become a current research hotspot. This paper proposes a photovoltaic power generation forecasting method and system based on data mini...
详细信息
existing security and privacy preserving schemes have too much wastage in token-based prepaid smart grids. In this paper, we propose a novel token concept and construct a security and privacy preserving prepaid scheme...
详细信息
ISBN:
(纸本)9781665475785
existing security and privacy preserving schemes have too much wastage in token-based prepaid smart grids. In this paper, we propose a novel token concept and construct a security and privacy preserving prepaid scheme based on the power request model. In the proposed scheme, honest consumers keep their anonymity, while consumers who send used tokens can be tracked for law enforcement and maintenance purposes. In addition to security and privacy preservation, the computational cost of verification is independent of the number of tokens based on the novel token concept in the power request model. Therefore, the number of points can be the basic unit in order to minimize the wastage. Finally, we prove the security features and demonstrate the performances of our scheme.
Smart grid technology is advancing bravely in the tide of the development of the times, and edgecomputing plays an increasingly prominent role in the real-time optimal control of smart distribution networks. This art...
详细信息
The addition of digital technology to energy systems, smart grids have been created, which make power delivery more reliable, efficient, and long-lasting. But this progress comes with big security risks, especially in...
详细信息
The development of communication technology has made the need to build more reliable and efficient smart power grid systems imminent. Theemergence of edgecomputing has significantly alleviated the pressure of data t...
详细信息
With the rapid development of flexible interconnection technology in active distribution networks, many power electronic devices have been employed to improve system operational performance. As a novel fully-controlle...
详细信息
Application of blockchain in financial services has opened new ways of efficiency in transaction processing, assets management and security. The application of parallel, distributed, and gridcomputing with blockchain...
详细信息
The prediction of transformer oil temperature is a time series prediction task, which is of great significance for the maintenance of substation equipment. In the process of multivariate prediction, not all variables ...
详细信息
ISBN:
(纸本)9798350389166;9798350389173
The prediction of transformer oil temperature is a time series prediction task, which is of great significance for the maintenance of substation equipment. In the process of multivariate prediction, not all variables are strongly correlated with the prediction of the transformer oil temperature. This paper proposes a transformer oil temperature prediction method based on Feature Selection and Smooth Residual blocks, named FSSR. A two-stage feature selection method is applied to select the related features. Residual connection is applied to convolutional neural network in smooth residual blocks. experimental study is performed on theeTTh1 and eTTh2 datasets. With different number of selected features, the best one can reduce the average MSe (mean squared error) by at least 20% than the worst one. Compared to the baseline models, FSSR has the best accuracy. Compared with LSTM (Long Short-Term Memory network), the average MSe of FSSR decreased by 27.62% on theeTTh1 dataset, and 39.34% on theeTTh2 dataset.
暂无评论