基于非線性因子的改進(jìn)鳥(niǎo)群算法在動(dòng)態(tài)能耗管理中的應(yīng)用

羅鈞; 劉澤偉; 張平; 劉學(xué)明; 柳政

doi:10.11999/JEIT190264

基于非線性因子的改進(jìn)鳥(niǎo)群算法在動(dòng)態(tài)能耗管理中的應(yīng)用

doi: 10.11999/JEIT190264

1.
重慶大學(xué)光電技術(shù)及系統(tǒng)教育部重點(diǎn)實(shí)驗(yàn)室重慶 400030
2.
兵器工業(yè) 5011 區(qū)域計(jì)量站重慶 400050

基金項(xiàng)目: 國(guó)防科工局十二五(跨十三五)技術(shù)基礎(chǔ)科研項(xiàng)目(JSJL2014209B004, JSJL2014209B005)

詳細(xì)信息

作者簡(jiǎn)介:
羅鈞：男，1963年生，教授，博士生導(dǎo)師，研究方向?yàn)槟Ｊ阶R(shí)與人工智能、精密機(jī)械及測(cè)試計(jì)量、智能信息處理

劉澤偉：男，1994年生，碩士生，研究方向?yàn)榍度胧较到y(tǒng)、精密儀器及機(jī)械、測(cè)試計(jì)量技術(shù)及儀器

張平：男，1970年生，碩士生，研究方向?yàn)榫軆x器及機(jī)械、測(cè)試計(jì)量技術(shù)及儀器

劉學(xué)明：男，1963年生，碩士生，研究方向?yàn)榫軆x器及機(jī)械、測(cè)試計(jì)量技術(shù)及儀器

通訊作者:
羅鈞　luojun@cqu.edu.cn

中圖分類號(hào): TP316.7
計(jì)量
- 文章訪問(wèn)數(shù): 2873
- HTML全文瀏覽量: 786
- PDF下載量: 56
- 被引次數(shù): 0
出版歷程
- 收稿日期: 2019-04-18
- 修回日期: 2019-10-08
- 網(wǎng)絡(luò)出版日期: 2019-10-16
- 刊出日期: 2020-03-19

Application of Improved Bird Swarm Algorithm Based on Nonlinear Factor in Dynamic Energy Management

1.
Key Laboratory of Optoelectronic Technology and System of Ministry of Education, Chongqing University, Chongqing 400030, China
2.
5011 District Measurement Station of Weapon Industry, Chongqing 400050, China

Funds: The Science, Technology and Industry Bureau for National Defense 12th Five-year (13th Five-year) Basic Technology Research Projects (JSJL2014209B004, JSJL2014209B005)

摘要

摘要:
針對(duì)實(shí)時(shí)系統(tǒng)能耗管理中動(dòng)態(tài)電壓調(diào)節(jié)(DVS)技術(shù)的應(yīng)用會(huì)導(dǎo)致系統(tǒng)可靠性下降的問(wèn)題，該文提出一種基于改進(jìn)鳥(niǎo)群(IoBSA)算法的動(dòng)態(tài)能耗管理法。首先，采用佳點(diǎn)集原理均勻地初始化種群，從而提高初始解的質(zhì)量，有效增強(qiáng)種群多樣性；其次，為了更好地平衡BSA算法的全局和局部搜索能力，提出非線性動(dòng)態(tài)調(diào)整因子；接著，針對(duì)嵌入式實(shí)時(shí)系統(tǒng)中處理器頻率可以動(dòng)態(tài)調(diào)整的特點(diǎn)，建立具有時(shí)間和可靠性約束的功耗模型；最后，在保證實(shí)時(shí)性和穩(wěn)定性的前提下，利用提出的IoBSA算法，尋求最小能耗的解決方案。通過(guò)實(shí)驗(yàn)結(jié)果表明，與傳統(tǒng)BSA等常見(jiàn)算法相比，改進(jìn)鳥(niǎo)群算法在求解最小能耗上有著很強(qiáng)的優(yōu)勢(shì)及較快的處理速度。
- 能耗管理 /
- 實(shí)時(shí)系統(tǒng) /
- 動(dòng)態(tài)電壓調(diào)節(jié) /
- 改進(jìn)鳥(niǎo)群算法
Abstract:
The application of Dynamic Voltage Scaling (DVS) technique in real-time system energy management will result in the decrease of system reliability. A dynamic energy management method based on Improved Bird Swarm Algorithm (IoBSA) is proposed in this paper. Firstly, the population is initialized uniformly with the principle of good point set, so as to improve the quality of initial solution and increase the diversity of population effectively. Secondly, in order to balance better the global and local search ability of BSA algorithm, the nonlinear dynamic adjustment factor is proposed. Then, a power consumption model with time and reliability constraints is established for the dynamic adjustment of processor frequency in embedded real-time systems. On the premise of ensuring real-time performance and stability, the proposed IoBSA algorithm is used to find the solution with minimum energy consumption. The experimental results show that compared with the traditional BSA algorithm and other common algorithms, the improved bird swarm algorithm has a strong advantage in solving the minimum energy consumption and a fast processing speed energy management.
- Energy management /
- Real-time system /
- Dynamic Voltage Scaling (DVS) /
- Improved Bird Swarm Algorithm (IoBSA)

HTML全文

圖 1 兩種方法初始化點(diǎn)圖

下載: 全尺寸圖片幻燈片

圖 2 頻率故障率

下載: 全尺寸圖片幻燈片

圖 3 收斂曲線圖

下載: 全尺寸圖片幻燈片

表 1 部分算法參數(shù)列表

算法	參數(shù)設(shè)置
BSA	$C = S = 1.5,{a_1} = {a_2} = 1,{\rm FQ} = 5,P \in [0.8,\,1]$ ${\rm FL} \in [0.5,\,0.9]$
LSABSA	${a_1} = {a_2} = 1,{\rm FQ} = 5,P \in [0.8,\,1],{\rm FL} \in [0.5,\,0.9]$ ${C_{\rm{e}}} = {S_{\rm{s}}} = 0,5,{C_{\rm{s}}} = {S_{\rm{e}}} = 2.5$
本文	${a_1} = {a_2} = 1,{\rm FQ} = 5,P \in [0.8,1],{\rm FL} \in [0.5,\,0.9]$
IoBSA	${C_{\rm{e}}} = {S_{\rm{s}}} = 0,5,{C_{\rm{s}}} = {S_{\rm{e}}} = 2$
CBSA	${Q_{\min }} = 0,{Q_{\max }} = 2,A = 0.7,r = 0.4,{P_\alpha } = 0.25$
CJADE	$F = 0.8,{C_r} = 0.5,c = 0.1,p = 0.05$
文獻(xiàn)[10]	${\rm{limit}} = 50$

下載: 導(dǎo)出CSV

表 2 實(shí)驗(yàn)參數(shù)列表

參數(shù)名	值	參數(shù)名	值
種群數(shù)	60	任務(wù)量	10 30 50
歸一化頻率	0.1～1.0	截止時(shí)間	20～220
WCET	20～50	迭代次數(shù)	1000
運(yùn)行次數(shù)	20	懲罰因子	5000

下載: 導(dǎo)出CSV

表 3 任務(wù)量為10的優(yōu)化結(jié)果

NPM-Val	St.	BSA	本文IoBSA	LSABSA	CSBA	GWO	CJADE	文獻(xiàn)[10]
375.57	Best	853.45	821.52	896.57	1040.55	830.83	904.09	1187.05
(min)	Worst	1110.96	1040.01	1090.47	1178.55	1053.47	1123.84	1061.25
3427.05	Mean	967.95	913.04	1005.06	1105.57	964.94	1035.83	1147.21
(max)	Std.Dev	58.18	57.66	60.36	34.85	50.46	53.92	55.25

下載: 導(dǎo)出CSV

表 4 任務(wù)量為30的優(yōu)化結(jié)果

NPM-Val	St.	BSA	本文IoBSA	LSABSA	CSBA	GWO	CJADE	文獻(xiàn)[10]
1126.70	Best	4355.13	3642.20	4197.41	4048.74	4353.49	4382.29	4881.90
(min)	Worst	5158.38	4936.64	5175.33	5033.73	5234.853	5021.29	5470.92
10281.15	Mean	4771.52	4368.30	4739.58	4519.13	4681.22	4677.56	4928.57
(max)	Std.Dev	215.87	345.31	269.02	238.77	223.95	150.11	304.62

下載: 導(dǎo)出CSV

表 5 任務(wù)量為50的優(yōu)化結(jié)果

NPM-Val	St.	BSA	本文IoBSA	LSABSA	CSBA	GWO	CJADE	文獻(xiàn)[10]
1877.83	Best	8572.38	8281.54	8610.62	無(wú)效	8384.88	8416.94	無(wú)效
(min)	Worst	10442.74	10023.18	10149.21	無(wú)效	無(wú)效	無(wú)效	無(wú)效
17135.25	Mean	9557.82	9319.57	9513.31	無(wú)效	無(wú)效	無(wú)效	無(wú)效
(max)	Std.Dev	587.00	535.50	520.50	643448.64	529852.01	75029.97	1147609.95

下載: 導(dǎo)出CSV

參考文獻(xiàn)(20)

SALEHI M E, SAMADI M, NAJIBI M, et al. Dynamic voltage and frequency scheduling for embedded processors considering power/performance tradeoffs[J]. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2011, 19(10): 1931–1935. doi: 10.1109/tvlsi.2010.2057520

TERZOPOULOS G and KARATZA H. Performance evaluation and energy consumption of a real-time heterogeneous grid system using DVS and DPM[J]. Simulation Modelling Practice and Theory, 2013, 36: 33–43. doi: 10.1016/j.simpat.2013.04.006

ERNST D, DAS S, LEE S, et al. Razor: Circuit-level correction of timing errors for low-power operation[J]. IEEE Micro, 2004, 24(6): 10–20. doi: 10.1109/MM.2004.85

RONG Peng, PEDRAM M. Energy-aware task scheduling and dynamic voltage scaling in a real-time system[J]. Journal of Low Power Electronics, 2008, 4(1): 1–10. doi: 10.1166/jolpe.2008.154

韓文雅, 王雷. 基于混合任務(wù)模型的動(dòng)態(tài)電壓調(diào)度在無(wú)線傳感器網(wǎng)絡(luò)中的應(yīng)用[J]. 計(jì)算機(jī)應(yīng)用, 2010, 30(9): 2522–2525. doi: 10.3724/SP.J.1087.2010.02522

HAN Wenya and WANG Lei. Application of dynamic voltage scaling based on hybrid-task model in wireless sensor network[J]. Journal of Computer Applications, 2010, 30(9): 2522–2525. doi: 10.3724/SP.J.1087.2010.02522

ZHAO Baoxian, AYDIN H, and ZHU Dakai. On maximizing reliability of real-time embedded applications under hard energy constraint[J]. IEEE Transactions on Industrial Informatics, 2010, 6(3): 316–328. doi: 10.1109/tii.2010.2051970

晏福, 徐建中, 李奉書(shū). 混沌灰狼優(yōu)化算法訓(xùn)練多層感知器[J]. 電子與信息學(xué)報(bào), 2019, 41(4): 872–879. doi: 10.11999/JEIT180519

YAN Fu, XU Jianzhong, and LI Fengshu. Training multi-layer perceptrons using chaos grey wolf optimizer[J]. Journal of Electronics &Information Technology, 2019, 41(4): 872–879. doi: 10.11999/JEIT180519

張興明, 殷從月, 魏帥, 等. 基于雙仲裁機(jī)制和田口正交法的貓群優(yōu)化任務(wù)調(diào)度算法[J]. 電子與信息學(xué)報(bào), 2018, 40(10): 2521–2528. doi: 10.11999/JEIT180215

ZHANG Xingming, YIN Congyue, WEI Shuai, et al. Cat swarm optimization task scheduling algorithm based on double arbitration mechanism and Taguchi orthogonal method[J]. Journal of Electronics &Information Technology, 2018, 40(10): 2521–2528. doi: 10.11999/JEIT180215

肖樂(lè)意, 歐陽(yáng)紅林, 范朝冬. 基于記憶分子動(dòng)理論優(yōu)化算法的多目標(biāo)截面投影Otsu圖像分割[J]. 電子與信息學(xué)報(bào), 2018, 40(1): 189–199. doi: 10.11999/JEIT170301

XIAO Leyi, OUYANG Honglin, and FAN Chaodong. Multi-objective cross section projection Otsu's method based on memory knetic-molecular theory optimization algorithm[J]. Journal of Electronics &Information Technology, 2018, 40(1): 189–199. doi: 10.11999/JEIT170301

羅鈞, 劉永鋒, 付麗. 能耗限制的實(shí)時(shí)周期任務(wù)可靠性感知調(diào)度[J]. 重慶大學(xué)學(xué)報(bào), 2011, 34(8): 86–89. doi: 10.11835/j.issn.1000-582x.2011.08.015

LUO Jun, LIU Yongfeng, and FU Li. Reliability-aware schedule of periodic tasks in energy-constrained real-time systems[J]. Journal of Chongqing University, 2011, 34(8): 86–89. doi: 10.11835/j.issn.1000-582x.2011.08.015

MENG Xianbing, GAO X Z, LU Lihua, et al. A new bio-inspired optimisation algorithm: bird swarm algorithm[J]. Journal of Experimental & Theoretical Artificial Intelligence, 2016, 28(4): 673–687. doi: 10.1080/0952813X.2015.1042530

楊文榮, 馬曉燕, 邊鑫磊. 基于Levy飛行策略的自適應(yīng)改進(jìn)鳥(niǎo)群算法[J]. 河北工業(yè)大學(xué)學(xué)報(bào), 2017, 46(5): 10–16. doi: 10.14081/j.cnki.hgdxb.2017.05.002

YANG Wenrong, MA Xiaoyan, and BIAN Xinlei. Adaptive improved bird swarm algorithm based on Levy flight strategy[J]. Journal of Hebei University of Technology, 2017, 46(5): 10–16. doi: 10.14081/j.cnki.hgdxb.2017.05.002

李延延, 萬(wàn)仁霞. 一種改進(jìn)算的鳥(niǎo)群算法[J]. 微電子學(xué)與計(jì)算機(jī), 2018, 35(9): 79–84.

LI Yanyan and WAN Renxia. An improved algorithm for bird swarm optimization[J]. Microelectronics &Computer, 2018, 35(9): 79–84.

吳軍, 王龍龍. 基于雙鳥(niǎo)群混沌優(yōu)化的Otsu圖像分割算法[J]. 微電子學(xué)與計(jì)算機(jī), 2018, 35(12): 119–124. doi: 10.19304/j.cnki.issn1000-7180.2018.12.024

WU Jun and WANG Longlong. An Otsu image segmentation algorithm based on chaos optimization of two BSA[J]. Microelectronics &Computer, 2018, 35(12): 119–124. doi: 10.19304/j.cnki.issn1000-7180.2018.12.024

王進(jìn)成, 高岳林. 基于改進(jìn)的鳥(niǎo)群算法求解農(nóng)產(chǎn)品冷鏈物流配送路徑優(yōu)化問(wèn)題[J]. 安徽農(nóng)業(yè)科學(xué), 2018, 46(25): 1–4. doi: 10.13989/j.cnki.0517-6611.2018.25.001

WANG Jincheng and GAO Yuelin. Optimization problem of cold chain logistics distribution path of agricultural products based on improved algorithm of bird swarm optimization[J]. Journal of Anhui Agricultural Sciences, 2018, 46(25): 1–4. doi: 10.13989/j.cnki.0517-6611.2018.25.001

謝國(guó)民, 干毅軍, 丁會(huì)巧. 基于佳點(diǎn)集的蝙蝠定位算法在WSN中應(yīng)用[J]. 傳感技術(shù)學(xué)報(bào), 2017, 30(8): 1252–1257. doi: 10.3969/j.issn.1004-1699.2017.08.021

XIE Guomin, GAN Yijun, and DING Huiqiao. A positioning algorithm based on bat algorithm and good-point setsin the application of WSN[J]. Chinese Journal of Sensors and Actuators, 2017, 30(8): 1252–1257. doi: 10.3969/j.issn.1004-1699.2017.08.021

ZHU D, MELHEM R, and CHILDERS B. Scheduling with dynamic voltage/speed adjustment using slack reclamation in multi-processor real-time systems[C]. Proceedings of the 22nd IEEE Real-time Systems Symposium, London, UK, 2001: 84–94.

SHEHAB M, KHADER A T, LAOUCHEDI M, et al. Hybridizing cuckoo search algorithm with bat algorithm for global numerical optimization[J]. The Journal of Supercomputing, 2019, 75(5): 2395–2422. doi: 10.1007/s11227-018-2625-x

MIRJALILI S, MIRJALILI S M, and LEWIS A. Grey wolf optimizer[J]. Advances in Engineering Software, 2014, 69: 46–61. doi: 10.1016/j.advengsoft.2013.12.007

羅鈞, 楊永松, 侍寶玉. 基于改進(jìn)的自適應(yīng)差分演化算法的二維Otsu多閾值圖像分割[J]. 電子與信息學(xué)報(bào), 2019, 41(8): 2017–2024. doi: 10.11999/JEIT180949

LUO Jun, YANG Yongsong, and SHI Baoyu. multi-threshold image segmentation of 2D Otsu based on improved adaptive differential evolution algorithm[J]. Journal of Electronics &Information Technology, 2019, 41(8): 2017–2024. doi: 10.11999/JEIT180949

相關(guān)文章

施引文獻(xiàn)

資源附件(0)

訪問(wèn)統(tǒng)計(jì)