基于短时受扰轨迹的电力系统暂态稳定评估方法
|
摘要
针对实时评估电网暂态稳定性的快速性和准确性的需求,提出一种基于故障清除后短时受扰轨迹和卷积神经网络的电力系统暂态稳定性的评估方法。该方法无需人工计算轨迹特征作为输入量,直接基于量测时序数据,利用深度学习模型的自主学习能力逐层提取隐含在短时轨迹的局部特征,构建短时受扰轨迹与稳定类别间的非线性映射关系,并引入考虑故障初期受扰程度的暂态稳定信息矩阵样本构建,以增强提取的局部特征的鲁棒性,提升模型的泛化能力,有效减少误判样本数,达到进一步提高准确率的目的。IEEE-39节点系统的仿真结果验证所提方法的有效性,并且所提方法的评估准确率比传统的暂态稳定浅层评估模型更加优越。
In order to meet the speed and accuracy requirements of real-time assessment of power grid transient stability, a method assessing transient stability of power system is proposed based on short-time disturbed trajectories after fault clearance and convolution neural network. This method does not need to calculate the trajectory characteristics as input, but directly uses the measured time series data. Based on the autonomous learning ability of the model, the local features hidden in the short-time trajectories are extracted layer by layer. A nonlinear mapping relationship between the short-time disturbed trajectories and stability type is constructed, and the sample construction of transient stability information matrix considering initial disturbance degree of the fault is introduced to enhance robustness of the extracted local features, improve the generalization ability of the model, and effectively reduce the number of misjudged samples, so that the accuracy of the model can be further improved. The proposed method is validated with the IEEE 39-bus system. Its accuracy is higher than that for conventional transient stability assessment methods.
In order to meet the speed and accuracy requirements of real-time assessment of power grid transient stability, a method assessing transient stability of power system is proposed based on short-time disturbed trajectories after fault clearance and convolution neural network. This method does not need to calculate the trajectory characteristics as input, but directly uses the measured time series data. Based on the autonomous learning ability of the model, the local features hidden in the short-time trajectories are extracted layer by layer. A nonlinear mapping relationship between the short-time disturbed trajectories and stability type is constructed, and the sample construction of transient stability information matrix considering initial disturbance degree of the fault is introduced to enhance robustness of the extracted local features, improve the generalization ability of the model, and effectively reduce the number of misjudged samples, so that the accuracy of the model can be further improved. The proposed method is validated with the IEEE 39-bus system. Its accuracy is higher than that for conventional transient stability assessment methods.
引文
[1]胡学浩.美加联合电网大面积停电事故的反思和启示[J].电网技术,2003,27(9):2-6.Hu Xuehao.Rethinking and enlightenment of large scope blackout in interconnected north America power grid[J].Power System Technology,2003,27(9):2-6(in Chinese).
[2]Itohv T,Ikuno S.Interpolating moving least-squares-based meshless time-domain method for stable simulation of electromagnetic wave propagation in complex-shaped domain[J].IEEE Transactions on Magnetics,2016,52(3):1-4.
[3]汪小明,刘涤尘,吴军,等.基于能量函数法的电网暂态稳定性分析[J].电网技术,2011,35(8):13-17.Wang Xiaoming,Liu Dichen,Wu Jun,et al.Energy function-based power system transient stability analysis[J].Power System Technology,2011,35(8):13-17(in Chinese).
[4]吴为,汤涌,孙华东,等.暂态稳定受扰轨迹预测的模型参数自适应研究[J].电网技术,2013,37(3):827-834.Wu Wei,Tang Yong,Sun Huadong,et al.Research on self-adaption of model parameters for prediction of disturbed trajectories in transient stability analysis[J].Power System Technology,2013,37(3):827-834(in Chinese).
[5]黄丹,杨秀媛,陈树勇,等.基于自忆性灰色Verhulst模型的暂态稳定受扰轨迹实时预测[J].高电压技术,2018,44(4):1285-1291.Huang Dan,Yang Xiuyuan,Chen Shuyong,et al.Prediction of disturbed trajectories in transient stability analysis based on self-memory gray verhulst model[J].High Voltage Engineering,2018,44(4):1285-1291(in Chinese).
[6]郑超,苗田,马世英.基于关键支路受扰轨迹凹凸性的暂态稳定判别及紧急控制[J].中国电机工程学报,2016,36(10):2600-2610.Zheng Chao,Miao Tian,Ma Shiying.Transient stability identification and emergency control based on the convex characteristic of the key branch’s disturbed trajectory[J].Proceedings of the CSEE,2016,36(10):2600-2610(in Chinese).
[7]杨松浩,王怀远,苏福,等.基于相轨迹凹凸性的暂态不稳定性判别方法的分析比较[J].电力自动化设备,2017,37(9):193-198.Yang Songhao,Wang Huaiyuan,Su Fu,et al.Analysis and comparison of transient instability detection methods based on convexity and concavity of phase trajectory[J].Electric Power Automation Equipment,2017,37(9):193-198(in Chinese).
[8]唐飞,王波,査晓明,等.基于双阶段并行隐马尔科夫模型的电力系统暂态稳定评估[J].中国电机工程学报,2013,33(10):90-97.Tang Fei,Wang Bo,Zha Xiaoming,et al.Power system transient stability assessment based on two-stage parallel hidden Markov model[J].Proceedings of the CSEE,2013,33(10):90-97(in Chinese).
[9]Padrón S,Hernández M,Falcón A.Reducing under-frequency load shedding in isolated power systems using neural networks.gran canaria:a case study[J].IEEE Transactions on Power Systems,2016,31(1):63-71.
[10]周艳真,吴俊勇,冀鲁豫,等.基于两阶段支持向量机的电力系统暂态稳定预测及预防控制[J].中国电机工程学报,2018,38(1):137-147.Zhou Yanzhen,Wu Junyong,Ji Luyu,et al.Two-stage support vector machines for transient stability prediction and preventive control of power systems[J].Proceedings of the CSEE,2018,38(1):137-147(in Chinese).
[11]陈厚合,王长江,姜涛,等.基于投影能量函数和Pin-SVM的电力系统暂态稳定评估[J].电工技术学报,2017,32(11):67-76.Chen Houhe,Wang Changjiang,Jiang Tao,et al.Transient stability assessment in bulk power grid using projection energy function and support vector machine with pinball loss[J].Transactions of China Electrotechnical Society,2017,32(11):67-76(in Chinese).
[12]朱利鹏,陆超,黄河,等.基于广域时序数据挖掘策略的暂态电压稳定评估[J].电网技术,2016,40(1):180-185.Zhu Lipeng,Lu Chao,Huang He,et al.Wide-area time series data mining based transient voltage stability assessment[J].Power System Technology,2016,40(1):180-185(in Chinese).
[13]霍思敏,王科,陈震海,等.基于轨迹输入特征支持向量机的湖南电网暂态稳定在线识别[J].电力系统保护与控制,2012,40(18):19-23.Huo Simin,Wang Ke,Chen Zhenhai,et al.Hunan power grid transient stability online detection based on support vector machine with with trajectory input features[J].Power System Protection and Control,2012,40(18):19-23(in Chinese).
[14]Zeiler M D,Fergus R.Stochastic Pooling for Regularization of Deep Convolutional Neural Networks[J].Eprint Arxiv,2013.
[15]苟竞,刘俊勇,Gareth Taylor,等.基于发电机对暂态势能集的电力系统暂态稳定快速评估[J].电网技术,2015,39(2):464-471.Gou Jing,Liu Junyong,Gareth Taylor,et al.Fast assessment of power system transient stability based on transient potential energy set of pairwise generators[J].Power System Technology,2015,39(2):464-471(in Chinese).
[16]魏少攀,杨明,韩学山,等.一种利用机端电压幅值轨迹MLE指标的暂态功角稳定在线辨识方法[J].中国电机工程学报,2017,37(13):3775-3786.Wei Shaopan,Yang Ming,Han Xueshan,et al.An online identification method for transient angle stability using MLE index of generator terminal voltage amplitude trajectory[J].Proceedings of the CSEE,2017,37(13):3775-3786(in Chinese).
[17]叶圣永,王晓茹,刘志刚,等.基于受扰严重机组特征及机器学习方法的电力系统暂态稳定评估[J].中国电机工程学报,2011,31(1):46-51.Ye Shengyong,Wang Xiaoru,Liu Zhigang,et al.Power system transient stability assessment based on severely disturbed generator attributes and machine learning method[J].Proceedings of the CSEE,2011,31(1):46-51(in Chinese).
[18]卢芳,于继来.基于广域相量测量的暂态稳定快速评估方法[J].电力系统自动化,2010,34(8):24-28.Lu Fang,Yu Jilai.WAMS based power system transient stability assessment[J].Automation of Electric Power Systems,2010,34(8):24-28(in Chinese).
[2]Itohv T,Ikuno S.Interpolating moving least-squares-based meshless time-domain method for stable simulation of electromagnetic wave propagation in complex-shaped domain[J].IEEE Transactions on Magnetics,2016,52(3):1-4.
[3]汪小明,刘涤尘,吴军,等.基于能量函数法的电网暂态稳定性分析[J].电网技术,2011,35(8):13-17.Wang Xiaoming,Liu Dichen,Wu Jun,et al.Energy function-based power system transient stability analysis[J].Power System Technology,2011,35(8):13-17(in Chinese).
[4]吴为,汤涌,孙华东,等.暂态稳定受扰轨迹预测的模型参数自适应研究[J].电网技术,2013,37(3):827-834.Wu Wei,Tang Yong,Sun Huadong,et al.Research on self-adaption of model parameters for prediction of disturbed trajectories in transient stability analysis[J].Power System Technology,2013,37(3):827-834(in Chinese).
[5]黄丹,杨秀媛,陈树勇,等.基于自忆性灰色Verhulst模型的暂态稳定受扰轨迹实时预测[J].高电压技术,2018,44(4):1285-1291.Huang Dan,Yang Xiuyuan,Chen Shuyong,et al.Prediction of disturbed trajectories in transient stability analysis based on self-memory gray verhulst model[J].High Voltage Engineering,2018,44(4):1285-1291(in Chinese).
[6]郑超,苗田,马世英.基于关键支路受扰轨迹凹凸性的暂态稳定判别及紧急控制[J].中国电机工程学报,2016,36(10):2600-2610.Zheng Chao,Miao Tian,Ma Shiying.Transient stability identification and emergency control based on the convex characteristic of the key branch’s disturbed trajectory[J].Proceedings of the CSEE,2016,36(10):2600-2610(in Chinese).
[7]杨松浩,王怀远,苏福,等.基于相轨迹凹凸性的暂态不稳定性判别方法的分析比较[J].电力自动化设备,2017,37(9):193-198.Yang Songhao,Wang Huaiyuan,Su Fu,et al.Analysis and comparison of transient instability detection methods based on convexity and concavity of phase trajectory[J].Electric Power Automation Equipment,2017,37(9):193-198(in Chinese).
[8]唐飞,王波,査晓明,等.基于双阶段并行隐马尔科夫模型的电力系统暂态稳定评估[J].中国电机工程学报,2013,33(10):90-97.Tang Fei,Wang Bo,Zha Xiaoming,et al.Power system transient stability assessment based on two-stage parallel hidden Markov model[J].Proceedings of the CSEE,2013,33(10):90-97(in Chinese).
[9]Padrón S,Hernández M,Falcón A.Reducing under-frequency load shedding in isolated power systems using neural networks.gran canaria:a case study[J].IEEE Transactions on Power Systems,2016,31(1):63-71.
[10]周艳真,吴俊勇,冀鲁豫,等.基于两阶段支持向量机的电力系统暂态稳定预测及预防控制[J].中国电机工程学报,2018,38(1):137-147.Zhou Yanzhen,Wu Junyong,Ji Luyu,et al.Two-stage support vector machines for transient stability prediction and preventive control of power systems[J].Proceedings of the CSEE,2018,38(1):137-147(in Chinese).
[11]陈厚合,王长江,姜涛,等.基于投影能量函数和Pin-SVM的电力系统暂态稳定评估[J].电工技术学报,2017,32(11):67-76.Chen Houhe,Wang Changjiang,Jiang Tao,et al.Transient stability assessment in bulk power grid using projection energy function and support vector machine with pinball loss[J].Transactions of China Electrotechnical Society,2017,32(11):67-76(in Chinese).
[12]朱利鹏,陆超,黄河,等.基于广域时序数据挖掘策略的暂态电压稳定评估[J].电网技术,2016,40(1):180-185.Zhu Lipeng,Lu Chao,Huang He,et al.Wide-area time series data mining based transient voltage stability assessment[J].Power System Technology,2016,40(1):180-185(in Chinese).
[13]霍思敏,王科,陈震海,等.基于轨迹输入特征支持向量机的湖南电网暂态稳定在线识别[J].电力系统保护与控制,2012,40(18):19-23.Huo Simin,Wang Ke,Chen Zhenhai,et al.Hunan power grid transient stability online detection based on support vector machine with with trajectory input features[J].Power System Protection and Control,2012,40(18):19-23(in Chinese).
[14]Zeiler M D,Fergus R.Stochastic Pooling for Regularization of Deep Convolutional Neural Networks[J].Eprint Arxiv,2013.
[15]苟竞,刘俊勇,Gareth Taylor,等.基于发电机对暂态势能集的电力系统暂态稳定快速评估[J].电网技术,2015,39(2):464-471.Gou Jing,Liu Junyong,Gareth Taylor,et al.Fast assessment of power system transient stability based on transient potential energy set of pairwise generators[J].Power System Technology,2015,39(2):464-471(in Chinese).
[16]魏少攀,杨明,韩学山,等.一种利用机端电压幅值轨迹MLE指标的暂态功角稳定在线辨识方法[J].中国电机工程学报,2017,37(13):3775-3786.Wei Shaopan,Yang Ming,Han Xueshan,et al.An online identification method for transient angle stability using MLE index of generator terminal voltage amplitude trajectory[J].Proceedings of the CSEE,2017,37(13):3775-3786(in Chinese).
[17]叶圣永,王晓茹,刘志刚,等.基于受扰严重机组特征及机器学习方法的电力系统暂态稳定评估[J].中国电机工程学报,2011,31(1):46-51.Ye Shengyong,Wang Xiaoru,Liu Zhigang,et al.Power system transient stability assessment based on severely disturbed generator attributes and machine learning method[J].Proceedings of the CSEE,2011,31(1):46-51(in Chinese).
[18]卢芳,于继来.基于广域相量测量的暂态稳定快速评估方法[J].电力系统自动化,2010,34(8):24-28.Lu Fang,Yu Jilai.WAMS based power system transient stability assessment[J].Automation of Electric Power Systems,2010,34(8):24-28(in Chinese).