特高压输电线路周围三维电场并行计算  被引量:8

Three-dimensional electric field's parallel calculation around UHV transmission line

在线阅读下载全文

作  者:周宏威[1] 孙丽萍[1] 包文泉[1] 李奔亮[1] 

机构地区:[1]东北林业大学机电工程学院,黑龙江哈尔滨150040

出  处:《电机与控制学报》2013年第12期76-80,88,共6页Electric Machines and Control

基  金:黑龙江省自然科学基金(QC2012C055);中央高校基金项目(DL12BB02);东北林业大学创新创业训练项目(201210225152)

摘  要:考虑到传统的解析方法计算电场难以获得准确的结果,而应用新兴的数值方法计算三维电场耗费时间过多,所以提出一种适用于特高压输电线电场计算的OpenMP多核并行方法。多核并行计算可以共享内存,比分布式并行减少了机器间数据传输的环节,更具时间优势。它采用离散计算区域,给处理器划分并分配任务和信息交换等方式对模拟电荷法进行多核并行处理。本文以我国晋东南-南阳-荆门特高压输电工程的猫头真型塔输电线路为实例,使用单核处理器和8核处理器分别计算,二者所得结果一致,证明了多核并行计算的准确性。最后对多核并行计算的效能实施评测,验证了多核并行计算可以实现三维电场的快速求解,但因为通信开销等原因,并行加速比不会随着处理器的增多而无限增加,同时并行效率也越来越低。As the result that the traditional analytical method for calculating the three-dimensional electric transmission lines is difficult to obtain accurately and the new numerical method costs too much time-con- suming, a calculation for UHV transmission line based on OpenMP multi-core parallel method is pro- posed. The muhi-core parallel computing shares memory, decreases the data transmission links between machines and it is more efficiency than the distributed parallel. Discrete computational domain was uti- lized to divide and allocate tasks and information exchange for processors, which are to process multi-core parallel on the charge simulation method. The realization process of using OpenMP multi-core parallel to compute three-dimensional electric field was presented. Taking the Maotou towers in Chinese Jindongnan- Nanyang-Jingmen UHV transmission project as an example, the results calculated by single-core processor or eight-core processors are the same demonstrating that the multi-core parallel computing has high effi- ciency and accuracy. At the same time, it proves that the parallel speedup ratio does not increase as the processors increase linearly.

关 键 词:OPENMP 并行计算 特高压 电场强度 模拟电荷法 

分 类 号:TM72[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