曹政 程冀川 尹相國(guó) 馬磊
摘 要: 針對(duì)基于邏輯關(guān)系的電網(wǎng)故障診斷方法診斷結(jié)果不可靠的缺點(diǎn),提出基于繼電保護(hù)狀態(tài)分析的電網(wǎng)故障診斷方法。設(shè)計(jì)電網(wǎng)故障診斷模型,提出三種基于繼電保護(hù)狀態(tài)分析的診斷策略,包括電網(wǎng)狀態(tài)診斷、保護(hù)行為診斷和斷路行為診斷,以此為依據(jù)構(gòu)建模型的故障假說(shuō),獲取疑似故障數(shù)據(jù)。將疑似故障數(shù)據(jù)輸入關(guān)聯(lián)樹(shù)進(jìn)行迭代,以迭代結(jié)果作為所提方法的最終診斷結(jié)果。實(shí)驗(yàn)表明所提方法的診斷具有優(yōu)異的可靠性和靈敏性。
關(guān)鍵詞: 繼電保護(hù)狀態(tài)分析; 電網(wǎng)故障診斷; 模型; 關(guān)聯(lián)樹(shù); 診斷策略; 靈敏性
中圖分類號(hào): TN915.02?34; TM77 文獻(xiàn)標(biāo)識(shí)碼: A 文章編號(hào): 1004?373X(2018)13?0147?04
Abstract: The power grid fault diagnosis method based on logical relationship has the defect of unreliable diagnosis result. Therefore, the power grid fault diagnosis method based on analysis of relay protection state is proposed, the power grid fault diagnosis model is designed, and three diagnosis strategies (power grid state diagnosis, protection behavior diagnosis and turn?off behavior diagnosis) based on relay protection status analysis are presented. On these basis, the fault hypothesis of the model is constructed to get the suspected fault data. The suspected fault data is input into the relevance tree for iteration, whose result is deemed as the final diagnosis result of the proposed method. The experimental results show that the proposed method has high diagnosis reliability and sensitivity.
Keywords: relay protection state analysis; power grid fault diagnosis; model; relevance tree; diagnosis strategy; sensitivity
電網(wǎng)在人們生活與工作中越來(lái)越重要,維護(hù)電網(wǎng)安全、高效、可靠地運(yùn)轉(zhuǎn)是電網(wǎng)基礎(chǔ)建設(shè)的關(guān)鍵點(diǎn)。在科技發(fā)展下,用電設(shè)備不斷增多,各省市間經(jīng)濟(jì)、政策的溝通不斷加強(qiáng),我國(guó)年用電量屢創(chuàng)新高,電網(wǎng)故障帶來(lái)的經(jīng)濟(jì)損失越來(lái)越大。電網(wǎng)故障具有隨機(jī)性和不可避免性,造成的危害影響廣泛,靈敏并可靠地診斷電網(wǎng)故障可以確保電網(wǎng)故障能夠在規(guī)定時(shí)限內(nèi)得以解決,是電網(wǎng)正常運(yùn)轉(zhuǎn)的最基本需求。
電網(wǎng)故障診斷的分析時(shí)間點(diǎn)是電網(wǎng)產(chǎn)生故障之后,常運(yùn)用智能決策系統(tǒng)、設(shè)備識(shí)別系統(tǒng)幫助電網(wǎng)管理員對(duì)故障線路進(jìn)行保護(hù),實(shí)現(xiàn)分析或重建故障情景。文獻(xiàn)[1]按照繼電器行為推理及分析策略,提出一種基于用電設(shè)備?斷路器邏輯關(guān)系的電網(wǎng)故障診斷方法,具有診斷效率高的優(yōu)點(diǎn)。但是由于復(fù)雜電網(wǎng)中的故障分布規(guī)律難以確定,方法的邏輯關(guān)系被打破,經(jīng)常出現(xiàn)誤報(bào)、漏報(bào),造成故障診斷結(jié)果不可靠[2]。由此能夠看出,電網(wǎng)故障診斷不能缺少對(duì)故障數(shù)量以及類型的累積分析。因此,提出基于繼電保護(hù)狀態(tài)分析的電網(wǎng)故障診斷方法。
1.1 電網(wǎng)故障診斷模型
當(dāng)電網(wǎng)出現(xiàn)故障時(shí),斷電設(shè)備運(yùn)轉(zhuǎn)行為出現(xiàn)變化,利用互聯(lián)網(wǎng)遠(yuǎn)程上傳斷電區(qū)域,故障診斷區(qū)域主要集中在斷電區(qū)域[3],可假設(shè)斷電區(qū)域周邊電網(wǎng)的斷路元件運(yùn)轉(zhuǎn)行為具有一致性。
繼電保護(hù)狀態(tài)分析采用完全動(dòng)態(tài)分析手段評(píng)價(jià)電網(wǎng)保護(hù)設(shè)備運(yùn)行數(shù)據(jù),將所取得的邏輯動(dòng)態(tài)變量輸入電網(wǎng)故障診斷模型,得到模型表達(dá)式:
一般認(rèn)為,凡是能夠進(jìn)行連續(xù)診斷的方法都具有靈敏性,因此靈敏性只分為極優(yōu)和普通兩個(gè)標(biāo)準(zhǔn)。
2.3 結(jié)果分析
實(shí)驗(yàn)對(duì)電網(wǎng)故障診斷方法診斷結(jié)果的數(shù)據(jù)提取采用暫態(tài)數(shù)據(jù)獲取手段,分A組和B組進(jìn)行小型電網(wǎng)和復(fù)雜電網(wǎng)的故障診斷,每0.2 s采集一次診斷數(shù)據(jù),對(duì)數(shù)據(jù)進(jìn)行匯總評(píng)估。從實(shí)驗(yàn)情況來(lái)看,基于繼電保護(hù)狀態(tài)分析的電網(wǎng)故障診斷方法很少存在電網(wǎng)保護(hù)設(shè)備和斷路元件拒動(dòng)或誤動(dòng)的情況,評(píng)估結(jié)果如表2所示。其他兩種方法的靈敏性和可靠性較差,不能保證用電安全。
本文提出基于繼電保護(hù)狀態(tài)分析的電網(wǎng)故障診斷方法,繼電保護(hù)狀態(tài)分析是指以電網(wǎng)保護(hù)設(shè)備運(yùn)轉(zhuǎn)情況為依據(jù),進(jìn)行電網(wǎng)數(shù)據(jù)的監(jiān)控、保護(hù)和分析,以分析結(jié)果為邏輯關(guān)系的動(dòng)態(tài)變量來(lái)診斷電網(wǎng)故障。實(shí)驗(yàn)結(jié)果表明,繼電保護(hù)狀態(tài)分析在電網(wǎng)故障診斷中具備優(yōu)秀的靈敏性和可靠性,可大大加強(qiáng)用電安全。
[1] 孫明蔚,童曉陽(yáng),劉新宇,等.運(yùn)用時(shí)序貝葉斯知識(shí)庫(kù)的電網(wǎng)故障診斷方法[J].電網(wǎng)技術(shù),2014,38(3):715?722.
SUN Mingwei, TONG Xiaoyang, LIU Xinyu, et al. A power system fault diagnosis method using temporal Bayesian knowledge bases [J]. Power system technology, 2014, 38(3): 715?722.
[2] 李乃永,梁軍,李磊,等.基于廣域故障錄波信息的調(diào)度端電網(wǎng)故障診斷系統(tǒng)[J].電力系統(tǒng)自動(dòng)化,2014,38(16):100?104.
LI Naiyong, LIANG Jun, LI Lei, et al. Fault diagnosis system based on wide?area fault wave recording information at dispatching centre [J]. Automation of electric power systems, 2014, 38(16): 100?104.
[3] 熊國(guó)江,石東源.電網(wǎng)故障診斷改進(jìn)解析模型及其自適應(yīng)生物地理學(xué)優(yōu)化方法[J].電工技術(shù)學(xué)報(bào),2014,29(4):205?211.
XIONG Guojiang, SHI Dongyuan. An improved analytic model for fault diagnosis of power grids and its self?adaptive biogeography?based optimization method [J]. Transactions of China electrotechnical society, 2014, 29(4): 205?211.
[4] 王瑞,施偉鋒.小波變換在電網(wǎng)故障診斷中的應(yīng)用[J].電子科技,2014,27(12):128?130.
WANG Rui, SHI Weifeng. Application of wavelet transform in power grid fault diagnosis [J]. Electronic science and technology, 2014, 27(12): 128?130.
[5] 程學(xué)珍,陳強(qiáng),于永進(jìn),等.基于最大似然譯碼字的Petri網(wǎng)電網(wǎng)故障診斷方法[J].電工技術(shù)學(xué)報(bào),2015,30(15):46?52.
CHENG Xuezhen, CHEN Qiang, YU Yongjin, et al. A fault diagnosis approach of power networks based on maximum likelihood decoding Petri net models [J]. Transactions of China electrotechnical society, 2015, 30(15): 46?52.
[6] 劉道兵,顧雪平,梁海平,等.電網(wǎng)故障診斷完全解析模型的解集評(píng)價(jià)與最優(yōu)解求取[J].中國(guó)電機(jī)工程學(xué)報(bào),2014,34(31):5668?5676.
LIU Daobing, GU Xueping, LIANG Haiping, et al. Solution evaluation and optimal solution discrimination of a complete analytical model for power system fault diagnosis [J]. Proceedings of the CSEE, 2014, 34(31): 5668?5676.
[7] 劉凡齊,張秀霞,王君瑞,等.繼電保護(hù)整定計(jì)算模塊的設(shè)計(jì)[J].現(xiàn)代電子技術(shù),2015,38(18):49?52.
LIU Fanqi, ZHANG Xiuxia, WANG Junrui, et al. Design of relay protection setting calculation module [J]. Modern electronics technique, 2015, 38(18): 49?52.
[8] 翟化欣.層次分析法和神經(jīng)網(wǎng)絡(luò)的電網(wǎng)安全評(píng)估[J].現(xiàn)代電子技術(shù),2016,39(21):168?171.
ZHAI Huaxin. Power grid security assessment combining AHP with neural network [J]. Modern electronics technique, 2016, 39(21): 168?171.
[9] 薛毓強(qiáng),李宗輝.基于因果時(shí)序網(wǎng)絡(luò)的FOA?GRNN電網(wǎng)故障診斷方法[J].電力系統(tǒng)及其自動(dòng)化學(xué)報(bào),2014,26(11):72?77.
XUE Yuqiang, LI Zonghui. Study on FOA?GRNN fault diagnosis method of power grid based on causal sequential network [J]. Proceedings of the CSU?EPSA, 2014, 26(11): 72?77.
[10] 李煒,馬樂(lè),段世剛,等.中繼衛(wèi)星系統(tǒng)信道模擬器硬件電路設(shè)計(jì)[J].現(xiàn)代電子技術(shù),2016,39(1):68?71.
LI Wei, MA Le, DUAN Shigang, et al. Hardware circuit design of channel simulator for relay satellite system [J]. Modern electronics technique, 2016, 39(1): 68?71.