《電子技術應用》
您所在的位置:首頁 > 其他 > 設計應用 > 基于最小割劃分的數(shù)?;旌戏抡嫦到y(tǒng)通信性能優(yōu)化方法
基于最小割劃分的數(shù)?;旌戏抡嫦到y(tǒng)通信性能優(yōu)化方法
2022年電子技術應用第1期
李億淵1,穆 清2,薛 巍1
1.清華大學 計算機科學與技術系,,北京100084,;2.中國電力科學研究院,,北京100192
摘要: 數(shù)?;旌戏抡媸抢斫庹鎸嶋娋W(wǎng)運行情況,支撐電網(wǎng)安全保障的重要手段。復雜的電網(wǎng)拓撲與硬實時的仿真需求對其計算性能提出了很高的要求,。目前數(shù)?;旌戏抡娑嗖捎貌⑿杏嬎慵夹g提高計算性能。隨著處理器和集群技術的發(fā)展,,異構集群系統(tǒng)逐漸成為高性能計算系統(tǒng)的主要構建方式,。針對多層次的系統(tǒng)架構,已有的電網(wǎng)劃分方式無法充分利用集群計算能力,。如何應對多層次核間通信延遲變化問題,,及引入設備交互導致的節(jié)點資源不對稱問題是數(shù)模混合仿真任務劃分與映射的新挑戰(zhàn),。針對中國電力科學研究院自研電磁暫態(tài)仿真系統(tǒng)ADPSS,,基于最小割劃分設計了兩階段的電網(wǎng)劃分與進程映射一體化優(yōu)化算法,在計算負載均衡和最小化通信上取得更好的平衡,,進一步降低了電磁暫態(tài)仿真的通信時間,。同時,該算法有效解決了集群節(jié)點資源不對稱情況下的任務優(yōu)化映射問題,。通過在西北和華東真實電網(wǎng)算例上的模擬測試,,所提出算法較ADPSS默認劃分與映射算法取得了平均40%和50%的通信性能提升,平均10%和12%的總體計算性能提升,。
中圖分類號: TP391
文獻標識碼: A
DOI:10.16157/j.issn.0258-7998.212436
中文引用格式: 李億淵,,穆清,,薛巍. 基于最小割劃分的數(shù)模混合仿真系統(tǒng)通信性能優(yōu)化方法[J].電子技術應用,,2022,,48(1):2-11.
英文引用格式: Li Yiyuan,Mu Qing,,Xue Wei. Communication optimization method of digital-analog hybrid simulation system based on min-cut partition[J]. Application of Electronic Technique,,2022,48(1):2-11.
Communication optimization method of digital-analog hybrid simulation system based on min-cut partition
Li Yiyuan1,,Mu Qing2,,Xue Wei1
1.Department of Computer Science and Technology,,Tsinghua University,,Beijing 100084,China,; 2.China Electric Power Research Institute,,Beijing 100192,China
Abstract: Digital-analog hybrid simulation is essential for understanding the real power grid and supporting power grid security. Complex power network topology and hard real-time simulation put forward high requirements for computing performance. At present, digital-analog hybrid simulation mainly uses parallel computing technology to improve computing performance. With the development of processor and cluster technology, heterogeneous cluster systems have gradually become the primary construction method of high-performance computing systems. For the multi-level system architecture, the existing power grid division methods can not fully use the cluster computing power. Dealing with the high latency of cross-layer communication and the unequal number of available processor cores on each computing node due to heterogeneous acceleration equipment is the main challenge of the partitioning and mapping algorithm. Aiming at the electromagnetic transient simulation system ADPSS developed by China Electric Power Research Institute, this paper designs a two-stage integrated optimization algorithm of power grid division and process mapping, which achieves a better load balance and minimizing communication, and further reduces the calculation time of the electromagnetic transient simulation. The algorithm is based on the min-cut partition and effectively solves the optimal mapping of sub-networks of unequal sizes on heterogeneous cluster systems. The simulation test was realized on the real power grid in Northwest and East China, compared with the ADPSS default partition and mapping algorithm, the proposed algorithm achieves an average communication performance improvement of 40% and 50% and an average overall computing performance improvement of 10% and 12%.
Key words : digital-analog hybrid simulation;graph partition,;min-cut,;process mapping;heterogeneous cluster system

0 引言

    近年來我國經(jīng)濟不斷發(fā)展,,社會對能源的需求不斷上升,,電力的消耗也隨之上升。我國電網(wǎng)的特高壓工程持續(xù)投運以滿足日益增長的用電負荷需求,。電力系統(tǒng)整體規(guī)模的擴大也給整個系統(tǒng)的穩(wěn)定和可靠運行帶來了更高的安全風險,。

    電力系統(tǒng)仿真是分析電網(wǎng)特征、分析電網(wǎng)穩(wěn)定性最重要的量化手段,。電力系統(tǒng)仿真分為穩(wěn)態(tài)仿真和動態(tài)仿真兩類,。動態(tài)仿真更關注電力系統(tǒng)的動態(tài)變化行為,主要包括機電暫態(tài)和電磁暫態(tài)兩種,。電磁暫態(tài)仿真建模更加精細,,是動態(tài)安全評估的重要工具。通過電磁暫態(tài)仿真,,研究人員能更好地理解電網(wǎng)在實際運行中的工作狀態(tài)及其變化,,從而在運行中有效調(diào)整控制方案,確保電力系統(tǒng)的安全,、穩(wěn)定運行,。

    電磁暫態(tài)是指電磁從一個穩(wěn)定狀態(tài)到另一個穩(wěn)定狀態(tài)中所經(jīng)歷的過程。在電力系統(tǒng)運行過程中,,通常由于電子元件的開關切換,、偶發(fā)的交直流故障以及雷擊等干擾,造成電磁暫態(tài)過程的快速變化[1],。模擬電磁暫態(tài)現(xiàn)象一般通過電力系統(tǒng)的時域建模來完成,。其目標是求出系統(tǒng)中各個時刻所有節(jié)點的電壓值和電流值,核心算法是將連續(xù)的微分系統(tǒng)離散化,,并使用迭代法隱式求解,。仿真步長代表離散時間點間隔,步長越短就能模擬更高頻的電網(wǎng)行為,,故步長大小是衡量電磁暫態(tài)仿真系統(tǒng)質(zhì)量的重要指標,。




本文詳細內(nèi)容請下載:http://forexkbc.com/resource/share/2000003901




作者信息:

李億淵1,,穆  清2,,薛  巍1

(1.清華大學 計算機科學與技術系,北京100084,;2.中國電力科學研究院,,北京100192)




wd.jpg

此內(nèi)容為AET網(wǎng)站原創(chuàng),未經(jīng)授權禁止轉(zhuǎn)載。