基于溯因推理网络的电网故障预测方法研究
|
摘要
电网诊断通常都是故障发生后,根据故障后产生的信息来推断故障,为了能在故障发生前进行预防,提出了模型预测(ModelPrediction,MP)和溯因推理网络(AbductiveReasoning Network,ARN)方法预测电网故障,模型预测利用电力系统中历史数据来预测电网无故障运行时的数据,与电网实际运行时的数据作对比,计算差值。溯因推理网络(ARN)使用带有简单低阶多项式的节点函数的几层网络,能够处理预测数据和相应的候选故障之间的复杂关系。模型预测和溯因推理网络方法相结合,能在保护装置和断路器动作前进行故障定位,具有故障预警功能。仿真结果表明该诊断系统与神经网络方法相比较可以获得快速、准确的诊断结果。
Fault diagnosis is usually judged according to the information after fault happens. In order to prevent the fault before the occurrence of the fault, model prediction(MP) and abductive reasoning network(ARN) are proposed to predict the power grid fault. MP predicts the trouble-free operation of the data of the power grid using the historical data, compares with the actual grid runtime data and calculates the difference. ARN handles complicated relationships between data processing and the corresponding candidate fault section using a hierarchical network with several layers of function nodes of simple low-order polynomials. The combination of model prediction and abductive reasoning network can locate the fault before the protection device and circuit breaker acts, and has the function of fault early warning. The simulation results show that the diagnosis system can obtain rapid and accurate diagnosis results compared with the neural network method.
Fault diagnosis is usually judged according to the information after fault happens. In order to prevent the fault before the occurrence of the fault, model prediction(MP) and abductive reasoning network(ARN) are proposed to predict the power grid fault. MP predicts the trouble-free operation of the data of the power grid using the historical data, compares with the actual grid runtime data and calculates the difference. ARN handles complicated relationships between data processing and the corresponding candidate fault section using a hierarchical network with several layers of function nodes of simple low-order polynomials. The combination of model prediction and abductive reasoning network can locate the fault before the protection device and circuit breaker acts, and has the function of fault early warning. The simulation results show that the diagnosis system can obtain rapid and accurate diagnosis results compared with the neural network method.
引文
[1]文福拴,邱家驹,韩祯祥.利用断路器信息诊断电力系统故障的高级遗传算法[J].电工技术学报,1996,11(2):58-64.Wen F S,Qiu J J,Han Z X.A tabusearch based approach to fault section estimation and state identification of unobserved protective relays in power systems using information from protective relays and circuit breakers[J].Transactions of China Electrotechnical Society,1996,11(2):58-64.
[2]文福拴,钱源平,韩祯祥,等.利用保护和断路器信息的电力系统故障诊断与不可观测的保护的状态识别的模型与Tabu搜索方法[J].电工技术学报,1998,13(5):1-8.Wen F S,Qian Y P,Han Z X,et al.A tabu Search Based Approach to Fault Sect ion Estimation and State Identification of Unobserved Protective Relays in Power Systems Using Information from Protective Relays and Circuit Breakers[J].Transactions of China Electrotechnical Society,1998,13(5):1-8.
[3]文福拴,韩祯祥.计及警报信息时间特性的故障诊断模型[J].电力系统自动化,1999,23(17):6-9.Wen F S,Han Z X.A new fault diagnosis model capable of dealing with the temporal information of alarm messages.Automation of Electric Power Systems,1999,23(17):6-9.
[4]文福拴,张勇,张岩,等.电力系统故障诊断与不可观测保护状态识别的解析模型[J].华南理工大学学报,2012,40(11):19-28.Wen F S,Zhang Y,Zhang Y,et al.Analytical Model for Fault Section Estimation and State Identification of Unobserved Protective Relays in Power Systems[J].Journal of South China University of Technology,2012,40(11):19-28.
[5]易妍,范辉,顾雪平,等.考虑不确定性的基于解析模型与信息理论的电网故障诊断方法[J].华北电力,2013,41(1):110-114.Yi Y,Fan H,Gu X P,et al.A Method Considering the Uncertainty Based on Analytical Model and Information Theory for Power System Fault Diagnosis[J].East China Electric Power.2013,41(1):110-114.
[6]梅念,石东源,李银红,等.计及信息畸变影响的电网故障诊断分级优化方法[J].电工技术学报,2009,24(9):178-190.Mei N,Shi D Y,Li Y H,et al.Power System Fault Diagnosis Based on Gradual Optimization in Consideration of Alarm Information Aberrance[J].Transactions of china electrotechnical society,2009,24(9):178-190.
[7]王磊,陈青,李天友,等.停电区域边界断路器信息缺失时的电网故障诊断[J].中国电机工程学报,2013,33(22):127-134.WANG L,CHEN Q,LI T Y,et al.Power System Fault Diagnosis Considering Absence of Alarm Messages of Circuit Breakers on the Border of Outage Area[J].Proceedings of the CSEE,2013,33(22):127-134.
[8]郭创新,游家训,彭明伟,等.基于面向元件神经网络与模糊积分融合技术的电网故障智能诊断[J].电工技术学报,2010,25(9):183-190.Guo C X,You J X,Peng M W,et al.A Fault Intelligent Diagnosis Approach Based on Element-Oriented Artificial Neural Networks and Fuzzy Integral Fusion[J].Transactions of china electrotechnical society,2010,25(9):183-190.
[9]吴亚联,李戈,龙辉.考虑相位失配量的电力系统动态状态估计方法[J].湘潭大学自然科学学报,2015,37(1):87-91.WuY L,Li G,Long H.Power System Dynamic State Estimation Considering Phase Mismatch[J].Natural Science Journal of Xiangtan University,2015,37(1):87-91.
[10]Tebianian H,Jeyasurya B.Dynamic state estimation in power systems:Modeling,and challenges[J].Electric Power Systems Research,2015,121(1):109-114.
[11]施微,黄磊,徐箭,等.基于有限同步量测数据的电力系统故障检测、辨识与定位[J].电网技术,2015,39(4):1067-1074.Shi W,Huang L,Xu J,et al.Fault Detection,Identification and Location Based on Measurement Data From Finite Synchronized Phasor Measurement[J].Power System Technology,2015,39(4):1067-1074.
[12]赵洪山,田甜.基于自适应无迹卡尔曼滤波的电力系统动态状态估计[J].电网技术,2014,38(1),188-192.Zhao H S,Tian T.Dynamic State Estimation for Power System Based on An Adaptive Unscented Kalman Filter[J].Power System Technology,2014,38(1):188-192.
[13]刘向杰,孔小兵.电力工业复杂系统模型预测控制-现状与发展[J].中国电机工程报,2013,33(5):79-85.Liu X J,Kong X B.Present Situation and Prospect of Model Predictive Control Application in Complex Power Industrial Process[J].Proceedings of the CSEE,2013,33(5):79-85.
[14]杨罡,刘明光,屈志坚.基于MPC算法的电力系统负荷频率控制[J].北京交通大学学报,2012,36(2):105-110.Yang G,Liu M G,Qu Z J.Load frequency control of power systems based on MPC algorithm[J].JOURNAL OF BEIJING JIAOTONGUNIVERSITY,2012,36(2):105-110.
[15]刘嘉璐,刘光,李海东.一种新型遗传算法与人工神经网络相结合的模型在电力系统短期负荷预测中的应用研究[J].电气传动自动化,2005,27(6):28-30.Liu J L,Liu G,Li H D.Study of a novel model combined genetic algor ithm with ar tificial neur al network for the short-term load for ecasting in power system[J].ELECTRIC DRIVE AUTOMATION,2005,27(6):28-30.
[16]CORDIER M O,DAGUE P,DUMAS M,et a1.A comparative analysis of AI and control theory approaches to model-based diagnosis[C].Proceedings of the 14th European Conference on Artificial Intelligence,August 20-25,Berlin,Germany,2000:36-140.
[17]郭挺,谢敏,刘明波.模型预测暂态电压稳定紧急控制的简化空间算法[J].中国电机工程报,2012,32(16):53-61.Guo T,Xie M,Liu M B.Reduced Space Algorithm for Model Predictive Transient Voltage Stability[J].Proceedings of the CSEE,2012,32(16):53-61.
[18]刘水平,刘明波,谢敏.基于轨迹灵敏度的协同分布式模型预测电压控制[J].华南理工大学学报,2013,41(4):14-20.Liu S P,Liu M B,Xie M.Cooperatively-Distributed Model Predictive Control of Voltage Based on Trajectory Sensitivity[J].Journal of South China University of Technology,2013,41(4):14-20.
[19]G.J.Montgomery,K.C.Drake.Aductive reasoning networks[J].Neuro comput,1991,3(2):97-104.
[20]Huang Y C.Abductive reasoning network based diagnosis system for fault section estimation in power systems[J].IEEE Transactions on Power Delivery,2002,17(2):369-374.
[21]Farlow,S.J.,Ed.Self-Organizing Methods in Modeling[M].New York:Marcel Dekker,1984.
[22]F.Wen,Z.Han.Fault section estimation in power systems using a genetic algorithm[J].Elect.Power Syst,Res.,1995,34(3):165-172.
[2]文福拴,钱源平,韩祯祥,等.利用保护和断路器信息的电力系统故障诊断与不可观测的保护的状态识别的模型与Tabu搜索方法[J].电工技术学报,1998,13(5):1-8.Wen F S,Qian Y P,Han Z X,et al.A tabu Search Based Approach to Fault Sect ion Estimation and State Identification of Unobserved Protective Relays in Power Systems Using Information from Protective Relays and Circuit Breakers[J].Transactions of China Electrotechnical Society,1998,13(5):1-8.
[3]文福拴,韩祯祥.计及警报信息时间特性的故障诊断模型[J].电力系统自动化,1999,23(17):6-9.Wen F S,Han Z X.A new fault diagnosis model capable of dealing with the temporal information of alarm messages.Automation of Electric Power Systems,1999,23(17):6-9.
[4]文福拴,张勇,张岩,等.电力系统故障诊断与不可观测保护状态识别的解析模型[J].华南理工大学学报,2012,40(11):19-28.Wen F S,Zhang Y,Zhang Y,et al.Analytical Model for Fault Section Estimation and State Identification of Unobserved Protective Relays in Power Systems[J].Journal of South China University of Technology,2012,40(11):19-28.
[5]易妍,范辉,顾雪平,等.考虑不确定性的基于解析模型与信息理论的电网故障诊断方法[J].华北电力,2013,41(1):110-114.Yi Y,Fan H,Gu X P,et al.A Method Considering the Uncertainty Based on Analytical Model and Information Theory for Power System Fault Diagnosis[J].East China Electric Power.2013,41(1):110-114.
[6]梅念,石东源,李银红,等.计及信息畸变影响的电网故障诊断分级优化方法[J].电工技术学报,2009,24(9):178-190.Mei N,Shi D Y,Li Y H,et al.Power System Fault Diagnosis Based on Gradual Optimization in Consideration of Alarm Information Aberrance[J].Transactions of china electrotechnical society,2009,24(9):178-190.
[7]王磊,陈青,李天友,等.停电区域边界断路器信息缺失时的电网故障诊断[J].中国电机工程学报,2013,33(22):127-134.WANG L,CHEN Q,LI T Y,et al.Power System Fault Diagnosis Considering Absence of Alarm Messages of Circuit Breakers on the Border of Outage Area[J].Proceedings of the CSEE,2013,33(22):127-134.
[8]郭创新,游家训,彭明伟,等.基于面向元件神经网络与模糊积分融合技术的电网故障智能诊断[J].电工技术学报,2010,25(9):183-190.Guo C X,You J X,Peng M W,et al.A Fault Intelligent Diagnosis Approach Based on Element-Oriented Artificial Neural Networks and Fuzzy Integral Fusion[J].Transactions of china electrotechnical society,2010,25(9):183-190.
[9]吴亚联,李戈,龙辉.考虑相位失配量的电力系统动态状态估计方法[J].湘潭大学自然科学学报,2015,37(1):87-91.WuY L,Li G,Long H.Power System Dynamic State Estimation Considering Phase Mismatch[J].Natural Science Journal of Xiangtan University,2015,37(1):87-91.
[10]Tebianian H,Jeyasurya B.Dynamic state estimation in power systems:Modeling,and challenges[J].Electric Power Systems Research,2015,121(1):109-114.
[11]施微,黄磊,徐箭,等.基于有限同步量测数据的电力系统故障检测、辨识与定位[J].电网技术,2015,39(4):1067-1074.Shi W,Huang L,Xu J,et al.Fault Detection,Identification and Location Based on Measurement Data From Finite Synchronized Phasor Measurement[J].Power System Technology,2015,39(4):1067-1074.
[12]赵洪山,田甜.基于自适应无迹卡尔曼滤波的电力系统动态状态估计[J].电网技术,2014,38(1),188-192.Zhao H S,Tian T.Dynamic State Estimation for Power System Based on An Adaptive Unscented Kalman Filter[J].Power System Technology,2014,38(1):188-192.
[13]刘向杰,孔小兵.电力工业复杂系统模型预测控制-现状与发展[J].中国电机工程报,2013,33(5):79-85.Liu X J,Kong X B.Present Situation and Prospect of Model Predictive Control Application in Complex Power Industrial Process[J].Proceedings of the CSEE,2013,33(5):79-85.
[14]杨罡,刘明光,屈志坚.基于MPC算法的电力系统负荷频率控制[J].北京交通大学学报,2012,36(2):105-110.Yang G,Liu M G,Qu Z J.Load frequency control of power systems based on MPC algorithm[J].JOURNAL OF BEIJING JIAOTONGUNIVERSITY,2012,36(2):105-110.
[15]刘嘉璐,刘光,李海东.一种新型遗传算法与人工神经网络相结合的模型在电力系统短期负荷预测中的应用研究[J].电气传动自动化,2005,27(6):28-30.Liu J L,Liu G,Li H D.Study of a novel model combined genetic algor ithm with ar tificial neur al network for the short-term load for ecasting in power system[J].ELECTRIC DRIVE AUTOMATION,2005,27(6):28-30.
[16]CORDIER M O,DAGUE P,DUMAS M,et a1.A comparative analysis of AI and control theory approaches to model-based diagnosis[C].Proceedings of the 14th European Conference on Artificial Intelligence,August 20-25,Berlin,Germany,2000:36-140.
[17]郭挺,谢敏,刘明波.模型预测暂态电压稳定紧急控制的简化空间算法[J].中国电机工程报,2012,32(16):53-61.Guo T,Xie M,Liu M B.Reduced Space Algorithm for Model Predictive Transient Voltage Stability[J].Proceedings of the CSEE,2012,32(16):53-61.
[18]刘水平,刘明波,谢敏.基于轨迹灵敏度的协同分布式模型预测电压控制[J].华南理工大学学报,2013,41(4):14-20.Liu S P,Liu M B,Xie M.Cooperatively-Distributed Model Predictive Control of Voltage Based on Trajectory Sensitivity[J].Journal of South China University of Technology,2013,41(4):14-20.
[19]G.J.Montgomery,K.C.Drake.Aductive reasoning networks[J].Neuro comput,1991,3(2):97-104.
[20]Huang Y C.Abductive reasoning network based diagnosis system for fault section estimation in power systems[J].IEEE Transactions on Power Delivery,2002,17(2):369-374.
[21]Farlow,S.J.,Ed.Self-Organizing Methods in Modeling[M].New York:Marcel Dekker,1984.
[22]F.Wen,Z.Han.Fault section estimation in power systems using a genetic algorithm[J].Elect.Power Syst,Res.,1995,34(3):165-172.