Suzhou Electric Appliance Research Institute
期刊號: CN32-1800/TM| ISSN1007-3175

Article retrieval

文章檢索

首頁 >> 文章檢索 >> 往年索引

基于SSA-DELM配電網光伏發電接納能力研究

來源:電工電氣發布時間:2025-01-07 15:07 瀏覽次數:114

基于SSA-DELM配電網光伏發電接納能力研究

楊群力1,蘇樂2,顧晨2,周鵬2,潘學萍2
(1 江蘇省戰略與發展研究中心,江蘇 南京 210036;
2 河海大學 能源與電氣學院,江蘇 南京 211100)
 
    摘 要:針對配電網拓撲以及參數難以獲取,數學建模方法無法應用于實際分析的困難,提出基于深度極限學習機(DELM)網絡的配電網光伏發電接納能力數據驅動分析方法。對配電網潮流分析數學模型與 DELM 網絡計算流程的相似性進行了對比,闡述了采用 DELM 網絡進行配電網數據建模的可行性;提出采用麻雀搜索算法(SSA)對 DELM 網絡進行優化,來提升 DELM 網絡的建模精度;給出了節點功率-節點電壓的非機理建模策略,并據此外推配電網對單點或多點接入下的光伏發電接納能力?;谙到y仿真及某實際低壓配電網,研究了電壓安全約束下配電網對光伏發電的接納能力,驗證了所提算法的有效性和優越性。
    關鍵詞: 配電網;電壓安全;光伏發電接納能力;麻雀搜索算法;深度極限學習機;電壓靈敏度
    中圖分類號:TM615 ;TM711     文獻標識碼:A     文章編號:1007-3175(2024)12-0034-08
 
Research on Photovoltaic Power Generation Acceptance Capacity of
Distribution Network Based on SSA-DELM
 
YANG Qun-li1, SU Le2, GU Chen2, ZHOU Peng2, PAN Xue-ping2
(1 Jiangsu Strategy and Development Research Center, Nanjing 210036, China;
2 College of Energy and Electrical Engineering, Hohai University, Nanjing 211100, China)
 
    Abstract: With the difficulty of attaining topology and parameters of distribution network, mathematical modeling methods can not be applied to practical analysis difficulties. Therefore, a data-driven analysis method for analyzing the acceptance capacity of the distribution network for photovoltaic (PV) power is proposed based on deep extreme learning machine(DELM) network. Firstly, the similarity between the mathematical model of power flow analysis of distribution network and the calculation process of DELM network is compared, and the feasibility of using DELM network for distribution network data modeling is expounded. Then the sparrow search algorithm (SSA) is proposed to optimize the DELM network to improve the data modeling accuracy by the DELM network. A non-mechanistic modeling strategy of node power-node voltage is given and based on this, the acceptance capacity of the distribution grid for PV power generation under single-point or multi-point access is deduced. Based on the system simulation and an actual low-voltage distribution network, the acceptance capacity of the distribution network for photovoltaic power generation under the constraint of voltage safety is studied, and the effectiveness and superiority of the proposed algorithm are verified.
    Key words: distribution network; voltage safety; photovoltaic power acceptance capacity; sparrow search algorithm; deep extreme learning machine; voltage sensitivity
 
參考文獻
[1] 高志遠,張晶,莊衛金,等. 關于新型電力系統部分特點的思考[J]. 電力自動化設備,2023,43(6) :137-143.
[2] 杜曉東,趙建利,劉科研,等. 基于數字孿生的光伏高比例配電網過載風險預警方法[J]. 電力系統保護與控制,2022,50(9) :136-144.
[3] HOKE A, BUTLER R, HAMBRICK J, et al.Steady-state analysis of maximum photovoltaic penetration levels on typical distribution feeders[J].IEEE Transactions on Sustainable Energy,2013,4(2) :350-357.
[4] AYRES H M, FREITAS W, DE ALMEIDA M C, et al.Method for determining the maximum allowable penetration level of distributed generation without steady-state voltage violations[J].IET Generation, Transmission & Distribution,2010,4(4) :495-508.
[5] AL-SAADI H , ZIVANOVIC R , AL-SARAWI S F.Probabilistic hosting capacity for active distribution networks[J].IEEE Transactions on Industrial Informatics,2017,13(5) :2519-2532.
[6] MOHAMMAD S S A , MA J , ZHANG D , et al .Probabilistic assessment of hosting capacity in radial distribution systems[J].IEEE Transactions on Sustainable Energy,2018,9(4) :1935-1947.
[7] 薛禹勝,賴業寧. 大能源思維與大數據思維的融合:(一) 大數據與電力大數據[J] . 電力系統自動化,2016,40(1) :1-8.
[8] 黃蔓云,衛志農,孫國強,等. 數據挖掘在配電網態勢感知中的應用:模型、算法和挑戰[J]. 中國電機工程學報,2022,42(18) :6588-6598.
[9] 巨云濤,楊明友,吳文傳. 適用于配電網三相優化潮流的數據物理融合驅動線性化方法[J]. 電力系統自動化,2022,46(13) :43-52.
[10] WENG Y, LIAO Y, RAJAGOPAL R.Distributed energy resources topology identification via graphical modeling[J].IEEE Transactions on Power Systems,2017,32(4) :2682-2694.
[11] United States Energy Information Administration.How many smart meters are installed in the United States, and who has them?[EB/OL].(2023-10-20)[2024-10-28].https://www.eia.gov/tools/faqs/faq.php?id=108&t=3.
[12] European Commission.Smart Metering Deployment in the European Union[EB/OL].(2023-10-24)[2024-10-28].http://ses.jrc.ec.europa.eu/smart-meteringdeployment-european-union.
[13] YU J, WENG Y, RAJAGOPAL R.Robust mapping rule estimation for power flow analysis in distribution grids[C]//North American Power Symposium(NAPS),2017.
[14] PERTL M, HEUSSEN K, GEHRKE O, et al.Voltage estimation in active distribution grids using neural networks[C]//IEEE Power and Energy Society General Meeting(PESGM),2016.
[15] MICHAEL P, PHILIP J D, KAI H, et al.Validation of a robust neural real-time voltage estimator for active distribution grids on field data[J].Electric Power Systems Research,2018,154(8):182-192.
[16] IMEN L, DJAMEL L.Power flow variation based on extreme learning machine algorithm in power system[J].International Journal of Power Electronics and Drive Systems,2019,10(3) :1244.
[17] BAGHAEE H R, MIRSALIM M, GHAREHPETIAN G B.Power calculation using RBF neural networks to improve power sharing of hierarchical control scheme in multi-DER microgrids[J].IEEE Journal of Emerging and Selected Topics in Power Electronics,2016,4(4) :1217-1225.
[18] YANG Y, YANG Z F, YU J, et al.Fast calculation of probabilistic power flow: A model-based deep learning approach[J].IEEE Transactions on Smart Grid,2020,11(3) :2235-2244.
[19] 張天策,王劍曉,李庚銀,等. 面向高比例新能源接入的配電網電壓時空分布感知方法[J]. 電力系統自動化,2021,45(2) :37-45.
[20] 曾亮,雷舒敏,王珊珊,等. 基于 OVMD-SSA-DELM-GM 模型的超短期風電功率預測方法[J] . 電網技術,2021,45(12) :4701-4710.
[21] 楊淑霞,韓奇,徐琳茜,等. 基于魚群算法優化 BP 神經網絡的電力客戶滿意度綜合評價方法[J] . 電網技術,2011,35(5) :146-151.
[22] 梁恩豪,孫軍偉,王延峰. 基于自適應樽海鞘算法優化 BP 的風光互補并網發電功率預測[J] . 電力系統保護與控制,2021,49(24) :114-120.
[23] 張甲甲,萬定生. 基于混合 GA 優化 LSTM 的中小流域流量預測研究[J]. 計算機仿真,2022,39(2) :283-287.
[24] 薛建凱. 一種新型的群智能優化技術的研究與應用[D].上海:東華大學,2020.
[25] 麻秀范. 含分布式電源的配電網規劃與優化運行研究[D].北京:華北電力大學,2013.

 

精品无码国产一区二区12p