首页|面向新型电力系统跨域分层算力网络的协同调度关键技术研究及应用

面向新型电力系统跨域分层算力网络的协同调度关键技术研究及应用

扫码查看
针对目前电力系统跨域分层算力网络的协同调度能力不足的问题,研究提出将遗传算法和模拟退火算法相结合得到融合算法,并基于融合算法构建分层算力网络的新型协同调度技术.在提出融合算法的性能测试中发现,该融合算法的损失值和拟合度分别为 0.005 和 0.965 3,显著优于遗传算法、人工鱼群算法以及松鼠搜索算法.之后对协同调度技术进行对比分析,结果显示,研究提出调度技术的调度效果和响应时间分别为 0.92 和 82.2 ms,显著优于对比技术.上述结果说明,研究提出的协同调度关键技术的实际应用效果较强,可促进电力系统的智能化发展.
Research and application of key technology of cooperative scheduling for new power system cross-domain hierarchical computing network
Aiming at the problem of insufficient cooperative scheduling ability of cross-domain hierarchical computing network of power system,a new cooperative scheduling technology combining genetic algorithm and simulated annealing algorithm to obtain fu-sion algorithm was proposed,and based on the fusion algorithm,hierarchical computing network was constructed.In the performance test of the proposed fusion algorithm,it is found that the loss value and fit degree of the fusion algorithm are 0.005 and 0.9653,re-spectively,which is significantly better than the genetic algorithm,artificial fish swarm algorithm and squirrel search algorithm.The results show that the scheduling effect and response time of the proposed scheduling technology are 0.92 ms and 82.2ms respectively,which is significantly better than the comparison technology.The above results show that the key technology of cooperative scheduling proposed in this study has strong practical application effect and can promote the intelligent development of power system.

computing power networkcooperative schedulingSA-GAelectric power system

李济伟、王洋、宋季彤、陈紫儿、柳瑞春

展开 >

国家电网有限公司信息通信分公司,北京 100053

算力网络 协同调度 SA-GA 电力系统

2024

自动化与仪器仪表
重庆工业自动化仪表研究所,重庆市自动化与仪器仪表学会

自动化与仪器仪表

CSTPCD
影响因子:0.327
ISSN:1001-9227
年,卷(期):2024.(11)