The expected increase in distributed energy resource (DER) penetration at residential levels is promoting new local market frameworks to manage the use of these resources efficiently and improve users' and energy ...
详细信息
ISBN:
(数字)9798350381740
ISBN:
(纸本)9798350381757
The expected increase in distributed energy resource (DER) penetration at residential levels is promoting new local market frameworks to manage the use of these resources efficiently and improve users' and energy communities' welfare. In this regard, Peer-to-peer (P2P) energy trading and the flexibility market emerge as tentative solutions to address this purpose, empowering the users and energy communities' role in the electricity markets and in the energy transition. This article introduces a deterministic three-stage optimization model to self-manage congestion arising from the high penetration of DERs within an energy community through a local P2P and flexibility market framework. Thus, under a day-ahead time framework, the initial stage considers that each user minimizes their own objective function, which is the energy bought from the grid. Then, in the second stage, the DSO receives the scheduling dispatch from each user through the local market operator and solves an optimal power flow problem to identify possible congestion line issues. If any congestion appears in the second stage, it is managed in the third stage using the flexibility provided by the DERs and the P2P energy trading, considering the user's preferences derived from the first stage problem. The model has been tested in a modified version of the IEEE 33 bus system and shows the capability to mitigate congestion line issues using the flexibility from the storage and PV systems under a P2P energy trading scheme.
GEMM with the small size of input matrices is becoming widely used in many fields like HPC and machine learning. Although many famous BLAS libraries already supported small GEMM, they cannot achieve near-optimal perfo...
详细信息
As to address the impact of heterogeneity in distributed Deep Learning (DL) systems, most previous approaches focus on prioritizing the contribution of fast workers and reducing the involvement of slow workers, incurr...
详细信息
ISBN:
(纸本)9781728170022
As to address the impact of heterogeneity in distributed Deep Learning (DL) systems, most previous approaches focus on prioritizing the contribution of fast workers and reducing the involvement of slow workers, incurring the limitations of workload imbalance and computation inefficiency. We reveal that grouping workers into communities, an abstraction proposed by us, and handling parameter synchronization in community level can conquer these limitations and accelerate the training convergence progress. The inspiration of community comes from our exploration of prior knowledge about the similarity between workers, which is often neglected by previous work. These observations motivate us to propose a new synchronization mechanism named Community-aware Synchronous parallel (CSP), which uses the Asynchronous Advantage Actor-Critic (A3C), a Reinforcement Learning (RL) based algorithm, to intelligently determine community configuration and fully improve the synchronization performance. The whole idea has been implemented in a system called Petrel that achieves a good balance between convergence efficiency and communication overhead. The evaluation under different benchmarks demonstrates our approach can effectively accelerate the training convergence speed and reduce synchronization traffic.
in response to the global energy crisis and environmental problems, the State grid has put forward a “new power system”, which mainly increases the proportion of new energy in the traditional power system. Among the...
详细信息
ISBN:
(数字)9781665471978
ISBN:
(纸本)9781665471985
in response to the global energy crisis and environmental problems, the State grid has put forward a “new power system”, which mainly increases the proportion of new energy in the traditional power system. Among them, the new energy based on photovoltaic+energy storage at the distribution network terminal will be the most promising form of power generation. Optical storage can not only solve the shortcomings of randomness, intermittence and fluctuation of sub-photovoltaic power generation, provide stable power, but also undertake the task of off-grid power supply after grid failure or in remote areas. Therefore, this paper studies the optical storage system under the distribution network from the perspective of optical storage model and grid-connected control, and builds the optical storage system under the distribution network for simulation verification, which provides a solid foundation for the subsequent coordinated control of load and storage in the source network.
The design technique based on the well-established concept of PV-STATCOM is implemented. The main task towards the achievement of more reliable increased integration of distributed energy resources on to the grid requ...
详细信息
In today's big data era, data and information are the main capital. High Energy Physics (HEP) as a data-intensive field of research drives significant demands for managing large datasets. Physics data need to be s...
详细信息
ISBN:
(纸本)9789532900996
In today's big data era, data and information are the main capital. High Energy Physics (HEP) as a data-intensive field of research drives significant demands for managing large datasets. Physics data need to be stored, accessed, and properly understood. This paper deals with the increasing resource requirements of the Large Hadron Collider (LHC) experiments at European Laboratory for Particle Physics (CERN). We analyze the unique requirements of the A Large Ion Collider Experiment (ALICE). This paper proposes a software-defined storage (SDS) model based on cloud and edge computing to optimize and organize the data storage of the current ALICE gridcomputing model. Also, it analyses how software-defined storage and distributedcomputing technologies could potentially be used to address the challenges and unique requirements faced by the ALICE experiment. The proposed model represents the basis for future research in which the parameters will be further examined, with an emphasis on the scalability of storage and computing resources.
In order to further optimize the output current harmonic suppression effect of photovoltaic grid-connected inverters, a composite control strategy of LCL type photovoltaic grid-connected inverter output current is pro...
详细信息
Smart grid functionalities, such as real-time monitoring and load balancing, require smart metering data collection at frequent time intervals. There are several threats to this data collection process, including pass...
详细信息
ISBN:
(纸本)9781665494052
Smart grid functionalities, such as real-time monitoring and load balancing, require smart metering data collection at frequent time intervals. There are several threats to this data collection process, including passive threats to customers' privacy and active threats to metering data integrity. Customer privacy can be breached with the fine grained metering data collection, and without appropriate measures for integrity verification, the metering data can be exploited to hamper the smart grid functionalities. distributed privacy-preserving frameworks are more robust than centralized frameworks against privacy threats. Several distributed privacy-preserving frameworks for smart metering data exist in the literature. However, these frameworks assume a semi-honest threat model that does not consider threats to data integrity. This paper introduces a distributed framework under a malicious adversarial model. The proposed framework is capable of verifying metering data's integrity while maintaining customer privacy. We evaluate our framework's performance via simulation and show its feasibility for real-world deployments. We also evaluate the framework's resilience to active and passive attacks, followed by a comparative analysis with existing related frameworks in the literature.
Differential evolution (DE) algorithms face performance challenges, which lean on improving solutions quality, speed-up, and exploitation of computational resources. parallelism represents a suitable paradigm for over...
详细信息
ISBN:
(数字)9798350397543
ISBN:
(纸本)9798350397550
Differential evolution (DE) algorithms face performance challenges, which lean on improving solutions quality, speed-up, and exploitation of computational resources. parallelism represents a suitable paradigm for overcoming the DE challenges. The ensemble of differential evolution variants (EDEV) algorithm is a recent DE algorithm. EDEV constitutes three DE variants (JADE, CoDE, and EPSDE), which may decrease its speedup. In this paper, a multi-population parallel ensemble of differential evolution variants (MPPEDEV) is proposed based on the synchronous master/slave parallel model. The performance of the proposed MPPEDEV is tested using a constrained real parameter problem proposed in CEC 2006. Compared to four state-of-the-art DE algorithms, which are JADE, CoDE, EPSDE, and EDEV, the results show that MPPEDEV outperforms EDEV in terms of execution time and solutions quality, depending on the population size as a control parameter. Furthermore, MPPEDEV and EDEV outperform JADE, CoDE, and EPSDE in terms of solutions' quality.
Accurate and efficient airway segmentation is essential for evaluating pulmonary diseases, aiding diagnosis, reducing the preoperative burden of airway identification, and minimizing patient discomfort during prolonge...
详细信息
ISBN:
(数字)9798350386226
ISBN:
(纸本)9798350386233
Accurate and efficient airway segmentation is essential for evaluating pulmonary diseases, aiding diagnosis, reducing the preoperative burden of airway identification, and minimizing patient discomfort during prolonged surgeries. However, current pulmonary airway reconstruction techniques are hindered by two major challenges: difficulty in accurately reconstructing fine airway branches due to the tendency to overlook small targets, and insufficient structural connectivity leading to frequent branch discontinuities within the airway tree. These limitations directly affect the clinical applicability of reconstructed airways. To overcome these challenges, a novel 3D pulmonary airway segmentation multi-task framework is proposed, designed to enhance the performance of existing backbone models. This approach integrates Anatomical Prior-Based Multi-Task Learning (AP-MTL) through the use of Gaussian-constructed connectivity-enhanced isosurfaces, significantly improving the network’s ability to maintain airway continuity. Additionally, a Class-Balanced CT Density Distribution Reconstruction mechanism (DDR-CB) is introduced, further refining the model’s capability to detect and segment fine airway branches. As a result of these enhancements, the model demonstrates a 11.5% average improvement in segmentation accuracy and connectivity compared to the baseline. The source code is publicly accessible at https://***/inexhaustible419/APMTLAirwaySegment.
暂无评论