학술논문

Cost-Effective Server Placement and Network Slice Provisioning for Virtual Power Plants
Document Type
Conference
Source
GLOBECOM 2023 - 2023 IEEE Global Communications Conference Global Communications Conference, GLOBECOM 2023 - 2023 IEEE. :2263-2268 Dec, 2023
Subject
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Engineering Profession
General Topics for Engineers
Power, Energy and Industry Applications
Signal Processing and Analysis
Costs
Heuristic algorithms
Network slicing
Bandwidth
Virtual power plants
Servers
Reliability
Language
ISSN
2576-6813
Abstract
Effective communication service for a Virtual Power Plant (VPP) is a major challenge as a VPP has very stringent latency and reliability requirements for communication with the Distributed Energy Resources (DERs). Network slicing is a promising solution to this challenge as network slices can be customized and provisioned to satisfy the requirements of the VPP's communication traffic. While determining the resources required for the network slice, it is important to consider the location of the VPP server and the paths along which the traffic is routed as significant latency can be experienced if the DERs are spread across a large area. The server's placement and traffic paths chosen can also impact the cost incurred as it would impact the required bandwidth on the links. We formulate a VPP Server Placement and Provisioning Problem that determines i) placement of VPP server, ii) paths for traffic distribution and iii) required bandwidth on each link in a wide area network connecting the DERs and the server while satisfying latency constraints and minimizing cost of reserving bandwidth on the links. We develop a heuristic algorithm for the formulation to obtain an effective solution in reasonable time. We also present a modified, convex version of the formulation under certain constraints through which an optimal solution can be obtained. We demonstrate the performance of our heuristic algorithm by comparing it with the optimal solution for the convex problem and another heuristic approach.