基于规模化算力弹性调度的电力系统-数据中心运行优化数据集

本数据集面向真实业务场景，构建了电力系统与数据中心系统协同优化的完整数据集。研究基于规模化弹性调度技术，依次考虑通信约束、隐私保护、不确定性风险以及非常规事件的影响，依据时变电价、服务质量等因素，决策真实场景下的数据中心任务调度策略，并在PyCharm与Matlab平台进行了仿真验证。数据采集包含资源分配、运行成本等多个维度的数据，具有良好的可视化与建模支持能力。数据采集过程严格遵循科学数据质量控制流程，确保了数据的准确性、一致性与可复现性。该数据集旨在研究面向电力系统多场景多时空调节需求的规模化算力弹性调度方法，适用于任务调度、用能管理、算电协同等领域研究，为电力系统-数据中心协同优化的规模化实际应用提供了重要的实验基础。本数据集包含基于规模化算力弹性调度的电力系统－数据中心运行优化数据集－相关论文及专利和基于规模化算力弹性调度的电力系统－数据中心运行优化数据集－支撑数据两个文件夹，其中支撑数据包括5个excel文件，5个word文件；其中：基于强化学习的数据中心电力－算力－通信协同调度数据，格式为.xlxs，容量为1.23MB；基于强化学习的数据中心电力－算力－通信协同调度算例分析，格式为.docx，容量为0.29MB；基于运筹优化的数据中心电力－算力－通信协同调度数据，格式为.xlxs，容量为0.04MB；基于运筹优化的数据中心电力－算力－通信协同调度算例分析，格式为.docx，容量为11.09MB；不确定性环境下计及决策风险的电力－算力调度数据，格式为.xlxs，容量为0.02MB；不确定性环境下计及决策风险的电力－算力调度算例分析，格式为.docx，容量为0.10MB；面向电网韧性提升的算力－电力协同调度数据，格式为.xlxs，容量为0.10MB；面向电网韧性提升的算力－电力协同调度算例分析，格式为.docx，容量为0.25MB；隐私保护下的多主体数据中心任务调度数据，格式为.xlxs，容量为0.03MB；隐私保护下的多主体数据中心任务调度算例分析，格式为.docx，容量为0.24MB；相关论文及专利文件夹包括3个pdf文件：基于电力－算力－通信协同调度的数据中心用能优化方法专利，格式为.pdf，容量为1.40MB；考虑不确定性风险的数据中心任务调度方法论文，格式为.pdf，容量为0.56MB；考虑数据可用性的多地理位置数据中心能量管理方法论文，格式为.pdf，容量为0.93MB。

This dataset is developed for real-world business scenarios, and constructs a comprehensive dataset for the collaborative optimization of power systems and data center systems.

The research is based on large-scale elastic scheduling technology, sequentially considering the impacts of communication constraints, privacy protection, uncertainty risks and unconventional events. It formulates data center task scheduling strategies for real-world scenarios based on factors such as time-of-use electricity prices and Quality of Service (QoS), and carries out simulation verification on PyCharm and Matlab platforms.

The data collection covers multiple dimensions including resource allocation and operating costs, and features excellent visualization and modeling support capabilities. The entire data collection process strictly follows scientific data quality control procedures, ensuring the accuracy, consistency and reproducibility of the dataset.

This dataset is designed to study large-scale computing power elastic scheduling methods that meet the multi-scenario and multi-spatiotemporal regulation requirements of power systems. It is applicable to research in fields such as task scheduling, energy management, and computing-power collaborative optimization, providing an important experimental foundation for large-scale practical applications of power system-data center collaborative optimization.

This dataset includes two folders: Related Papers and Patents of the Power System-Data Center Operation Optimization Dataset Based on Large-Scale Computing Power Elastic Scheduling and Supporting Data of the Power System-Data Center Operation Optimization Dataset Based on Large-Scale Computing Power Elastic Scheduling. The supporting data folder contains 5 Excel files and 5 Word files, as detailed below:
1. Data on power-computing-communication collaborative scheduling in data centers based on reinforcement learning, in .xlsx format, with a size of 1.23 MB;
2. Case analysis of power-computing-communication collaborative scheduling in data centers based on reinforcement learning, in .docx format, with a size of 0.29 MB;
3. Data on power-computing-communication collaborative scheduling in data centers based on operational optimization, in .xlsx format, with a size of 0.04 MB;
4. Case analysis of power-computing-communication collaborative scheduling in data centers based on operational optimization, in .docx format, with a size of 11.09 MB;
5. Power-computing scheduling data considering decision-making risks in uncertain environments, in .xlsx format, with a size of 0.02 MB;
6. Power-computing scheduling case analysis considering decision-making risks in uncertain environments, in .docx format, with a size of 0.10 MB;
7. Computing power-power collaborative scheduling data for improving grid resilience, in .xlsx format, with a size of 0.10 MB;
8. Computing power-power collaborative scheduling case analysis for improving grid resilience, in .docx format, with a size of 0.25 MB;
9. Multi-agent data center task scheduling data under privacy protection, in .xlsx format, with a size of 0.03 MB;
10. Multi-agent data center task scheduling case analysis under privacy protection, in .docx format, with a size of 0.24 MB;

The related papers and patents folder contains 3 PDF files:
1. Patent for data center energy optimization method based on power-computing-communication collaborative scheduling, in .pdf format, with a size of 1.40 MB;
2. Paper on data center task scheduling method considering uncertainty risks, in .pdf format, with a size of 0.56 MB;
3. Paper on multi-location data center energy management method considering data availability, in .pdf format, with a size of 0.93 MB.

搜集汇总

背景与挑战

背景概述

该数据集聚焦电力系统与数据中心的协同优化，提供了基于规模化算力弹性调度的多维度实验数据，适用于任务调度、用能管理等领域研究。数据集包含强化学习和运筹优化等多种方法的数据及分析，支持电力系统-数据中心协同优化的实际应用研究。