基于自然语言处理技术的智能配电网调度优化方法研究  被引量:2

Research on Intelligent Distribution Network Scheduling Optimization Method Based on Natural Language Processing Technology

在线阅读下载全文

作  者:岳宝强 袁森 彭静 王军 亓富军 YUE Baoqiang;YUAN Sen;PENG Jing;WANG Jun;QI Fujun(Linyi Power Supply Compary of State Grid Shandong Electric Power Company,Linyi 276000,China)

机构地区:[1]国网山东省电力公司临沂供电公司,山东临沂276000

出  处:《微型电脑应用》2023年第11期55-59,共5页Microcomputer Applications

基  金:国网山东省电力公司科技项目资助(2023A-0614)。

摘  要:由于历史电力数据中许多不确定性特征被忽略,造成调度优化方案无法解决电网节点电压越限,为此提出一种基于自然语言处理技术的智能配电网调度优化方法。采用包含注意力机制的编码器-解码器框架,并利用自然语言处理技术深入挖掘电网历史运行数据,获取负荷参数特征;在此基础上,建立一个优化调度模型,其中考虑有功调度成本最低的目标;为了求解这个优化调度模型,采用改进的随机搜索和启发式搜索算法,通过这些方法,得到一个最优的智能配电网调度方案。算例结果表明,使用所提出的调度优化方法后,智能配电网节点电压保持在0.95~1.05 pu的优质电压范围内,并且电压越上限的情况完全消失。Because many uncertain features in historical power data are ignored,the generated scheduling optimization scheme can not solve the problem of overvoltage of grid nodes.This paper presents an intelligent distribution network scheduling optimization method based on natural language processing technology.The encoder-decoder framework including attention mechanism is adopted,and natural language processing technology is used to mine deeply into the historical operation data to obtain the characteristics of load parameters.On this basis,an optimal scheduling model is established,in which the lowest cost of active scheduling is considered.In order to solve the optimal scheduling model,an optimal scheduling scheme of intelligent distribution network is obtained by improved random search and heuristic search algorithms.The calculation results show that by the proposed scheduling optimization method,the node voltage of the intelligent distribution network is kept within the high quality voltage range of 0.95~1.05 pu,and the situation of voltage exceeding the upper limit completely disappears.

关 键 词:自然语言处理技术(NLP) 电网数据 负荷预测 优化调度 改进的和声搜索算法 

分 类 号:TM731[电气工程—电力系统及自动化]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象