Phasor measurement units(PMUs) can provide real-time measurement data to construct the ubiquitous electric of the Internet of Things. However, due to complex factors on site, PMU data can be easily compromised by inte...Phasor measurement units(PMUs) can provide real-time measurement data to construct the ubiquitous electric of the Internet of Things. However, due to complex factors on site, PMU data can be easily compromised by interference or synchronization jitter. It will lead to various levels of PMU data quality issues, which can directly affect the PMU-based application and even threaten the safety of power systems. In order to improve the PMU data quality, a data-driven PMU bad data detection algorithm based on spectral clustering using single PMU data is proposed in this paper. The proposed algorithm does not require the system topology and parameters. Firstly, a data identification method based on a decision tree is proposed to distinguish event data and bad data by using the slope feature of each data. Then, a bad data detection method based on spectral clustering is developed. By analyzing the weighted relationships among all the data, this method can detect the bad data with a small deviation. Simulations and results of field recording data test illustrate that this data-driven method can achieve bad data identification and detection effectively. This technique can improve PMU data quality to guarantee its applications in the power systems.展开更多
With more data-driven applications introduced in wide-area monitoring systems(WAMS),data quality of phasor measurement units(PMUs)becomes one of the fundamental requirements for ensuring reliable WAMS applications.Thi...With more data-driven applications introduced in wide-area monitoring systems(WAMS),data quality of phasor measurement units(PMUs)becomes one of the fundamental requirements for ensuring reliable WAMS applications.This paper proposes a doubly-fed deep learning method for bad data identification in linear state estimation,which can:(1)identify bad data under both steady states and contingencies;(2)achieve higher accuracy than conventional pre-filtering approaches;(3)reduce iteration burden for linear state estimation;(4)efficiently identify bad data in a parallelizable scheme.The proposed method consists of four key steps:(1)preprocessing filter;(2)online training of short-term deep neural network;(3)offline training of long-term deep neural network;(4)a decision merger.Through delicate design and comprehensive training,the proposed method can effectively differentiate the bad data from event data without relying on real-time topology information.An IEEE 39-bus system simulated by DSATools TSAT and a provincial electric power system with real PMU data collected are used to verify the proposed method.Multiple test scenarios are applied,which include steady states,three-phase-to-ground faults with(un)successful auto-reclosing,low-frequency oscillation,and low-frequency oscillation with simultaneous threephase-to-ground faults.The proposed method demonstrates satisfactory performance during both the training session and the testing session.展开更多
Due to its high accuracy and ease of calculation,synchrophasor-based linear state estimation(LSE)has attracted a lot of attention in the last decade and has formed the cornerstone of many wide area monitor system(WAMS...Due to its high accuracy and ease of calculation,synchrophasor-based linear state estimation(LSE)has attracted a lot of attention in the last decade and has formed the cornerstone of many wide area monitor system(WAMS)applications.However,an increasing number of data quality concerns have been reported,among which bad data can significantly undermine the performance of LSE and many other WAMS applications it supports.Bad data filtering can be difficult in practice due to a variety of issues such as limited processing time,non-uniform and changing patterns,and etc.To pre-process phasor measurement unit(PMU)measurements for LSE,we propose an improved denoising autoencoder(DA)-aided bad data filtering strategy in this paper.Bad data is first identified by the classifier module of the proposed DA and then recovered by the autoencoder module.Two characteristics distinguish the proposed methodology:1)The approach is lightweight and can be implemented at individual PMU level to achieve maximum parallelism and high efficiency,making it suited for real-time processing;2)the system not only identifies bad data but also recovers it,especially for critical measurements.We use numerical experiments employing both simulated and real-world phasor data to validate and illustrate the effectiveness of the proposed method.展开更多
The problems of recovering the state of power systems and detecting the instances of bad data have been widely studied in literature.Nevertheless,these two operations have been designed and optimized for the most part...The problems of recovering the state of power systems and detecting the instances of bad data have been widely studied in literature.Nevertheless,these two operations have been designed and optimized for the most part in isolation.Specifically,state estimators are optimized based on the minimum mean-square error criteria,which is only optimal when the source of distortions in the data is Gaussian random noise.Hence,the state estimators fail to perform optimality when the data is further contaminated by bad data,which cannot necessarily be modeled by additive Gaussian terms.The problem of power state estimation has been studied extensively.But the fundamental performance limits and the attendant decision rules are unknown when the data is potentially compromised by random bad data(due to sensor failures)or structured bad data(due to cyber attacks,which are also referred to false data injection attacks).This paper provides a general framework that formalizes the underlying connection between state estimation and bad data detection routines.We aim to carry out the combined tasks of detecting the presence of random and structured bad data,and form accurate estimations for the state of power grid.This paper characterizes the optimal detectors and estimators.Furthermore,the gains with respect to the existing state estimators and bad data detectors are established through numerical evaluations.展开更多
In experimental tests, besides data in range of allowable error, the experimenters usually get some unexpected wrong data called bad points. In usual experimental data processing, the method of bad points exclusion ba...In experimental tests, besides data in range of allowable error, the experimenters usually get some unexpected wrong data called bad points. In usual experimental data processing, the method of bad points exclusion based on automatic programming is seldom taken into consideration by researchers. This paper presents a new method to reject bad points based on Hough transform, which is modified to save computational and memory consumptions. It is fit for linear data processing and can be extended to process data that is possible to be transformed into and from linear form; curved lines, which can be effectively detected by Hough transform. In this paper, the premise is the distribution of data, such as linear distribution and exponential distribution, is predetermined. Steps of the algorithm start from searching for an approximate curve line that minimizes the sum of parameters of data points. The data points, whose parameters are above a self-adapting threshold, will be deleted. Simulation experiments have manifested that the method proposed in this paper performs efficiently and robustly.展开更多
配电网环境复杂,配电网同步相量测量装置(distribution network synchronous phasor measurement unit, D-PMU)容易受到干扰而产生坏数据,进一步影响基于测量数据的应用效果。为了提高D-PMU数据质量,提出一种不依赖系统拓扑的基于密度...配电网环境复杂,配电网同步相量测量装置(distribution network synchronous phasor measurement unit, D-PMU)容易受到干扰而产生坏数据,进一步影响基于测量数据的应用效果。为了提高D-PMU数据质量,提出一种不依赖系统拓扑的基于密度的噪场应用空间聚类(density-based spatial clustering of applications with noise, DBSCAN)的配电网同步测量坏数据检测方法。首先利用基于密度的聚类算法DBSCAN进行异常数据检测。通过轮廓系数和邓恩指数对DBSCAN的聚类结果进行综合评价。利用麻雀搜索算法实现自适应参数调整,解决检测时需要预先处理训练、标记数据的问题。在此基础上,将时间序列聚类的K-Medoids算法和动态时间规整算法相结合,通过衡量不同时间序列之间的相似性,解决了D-PMU在电气联系较弱时对扰动数据与坏数据的区分问题,增强了数据处理的准确性与噪声环境下的稳健性。仿真和实际数据的测试结果表明,所提方法能有效区分真实扰动数据并准确识别D-PMU坏数据。展开更多
为提高状态估计的抗差性,提出一种基于最大指数绝对值目标函数的状态估计(maximum exponential absolute value state estimation,MEAV)方法。首先给出了MEAV的基本模型,并介绍了其理论基础和数学性质。由于MEAV基本模型的目标函数并非...为提高状态估计的抗差性,提出一种基于最大指数绝对值目标函数的状态估计(maximum exponential absolute value state estimation,MEAV)方法。首先给出了MEAV的基本模型,并介绍了其理论基础和数学性质。由于MEAV基本模型的目标函数并非处处可导,因而无法利用基于梯度的方法进行求解。为此,给出了MEAV基本模型的等价模型,并详细推导了基于原-对偶内点算法的MEAV等价模型的求解方法。算例分析表明,MEAV在估计过程中可自动抑制多个强相关不良数据,显示了良好的抗差性和较高的计算效率,因而具有良好的工程应用前景。展开更多
基金supported by the National Key R&D Program (No.2017YFB0902901)the National Natural Science Foundation of China (No.51627811,No.51725702,and No.51707064)。
文摘Phasor measurement units(PMUs) can provide real-time measurement data to construct the ubiquitous electric of the Internet of Things. However, due to complex factors on site, PMU data can be easily compromised by interference or synchronization jitter. It will lead to various levels of PMU data quality issues, which can directly affect the PMU-based application and even threaten the safety of power systems. In order to improve the PMU data quality, a data-driven PMU bad data detection algorithm based on spectral clustering using single PMU data is proposed in this paper. The proposed algorithm does not require the system topology and parameters. Firstly, a data identification method based on a decision tree is proposed to distinguish event data and bad data by using the slope feature of each data. Then, a bad data detection method based on spectral clustering is developed. By analyzing the weighted relationships among all the data, this method can detect the bad data with a small deviation. Simulations and results of field recording data test illustrate that this data-driven method can achieve bad data identification and detection effectively. This technique can improve PMU data quality to guarantee its applications in the power systems.
基金supported by the Science and Technology Program of State Grid Corporation of China under project“AI based oscillation detection and control”(No.SGJS0000DKJS1801231)
文摘With more data-driven applications introduced in wide-area monitoring systems(WAMS),data quality of phasor measurement units(PMUs)becomes one of the fundamental requirements for ensuring reliable WAMS applications.This paper proposes a doubly-fed deep learning method for bad data identification in linear state estimation,which can:(1)identify bad data under both steady states and contingencies;(2)achieve higher accuracy than conventional pre-filtering approaches;(3)reduce iteration burden for linear state estimation;(4)efficiently identify bad data in a parallelizable scheme.The proposed method consists of four key steps:(1)preprocessing filter;(2)online training of short-term deep neural network;(3)offline training of long-term deep neural network;(4)a decision merger.Through delicate design and comprehensive training,the proposed method can effectively differentiate the bad data from event data without relying on real-time topology information.An IEEE 39-bus system simulated by DSATools TSAT and a provincial electric power system with real PMU data collected are used to verify the proposed method.Multiple test scenarios are applied,which include steady states,three-phase-to-ground faults with(un)successful auto-reclosing,low-frequency oscillation,and low-frequency oscillation with simultaneous threephase-to-ground faults.The proposed method demonstrates satisfactory performance during both the training session and the testing session.
基金This work was supported by SGCC Science and Technology Program under project“AI-based oscillation detection and control”(SGJS0000DKJS1801231)。
文摘Due to its high accuracy and ease of calculation,synchrophasor-based linear state estimation(LSE)has attracted a lot of attention in the last decade and has formed the cornerstone of many wide area monitor system(WAMS)applications.However,an increasing number of data quality concerns have been reported,among which bad data can significantly undermine the performance of LSE and many other WAMS applications it supports.Bad data filtering can be difficult in practice due to a variety of issues such as limited processing time,non-uniform and changing patterns,and etc.To pre-process phasor measurement unit(PMU)measurements for LSE,we propose an improved denoising autoencoder(DA)-aided bad data filtering strategy in this paper.Bad data is first identified by the classifier module of the proposed DA and then recovered by the autoencoder module.Two characteristics distinguish the proposed methodology:1)The approach is lightweight and can be implemented at individual PMU level to achieve maximum parallelism and high efficiency,making it suited for real-time processing;2)the system not only identifies bad data but also recovers it,especially for critical measurements.We use numerical experiments employing both simulated and real-world phasor data to validate and illustrate the effectiveness of the proposed method.
基金supported by the US NationalScience Foundation(No.ECCS-1554482).
文摘The problems of recovering the state of power systems and detecting the instances of bad data have been widely studied in literature.Nevertheless,these two operations have been designed and optimized for the most part in isolation.Specifically,state estimators are optimized based on the minimum mean-square error criteria,which is only optimal when the source of distortions in the data is Gaussian random noise.Hence,the state estimators fail to perform optimality when the data is further contaminated by bad data,which cannot necessarily be modeled by additive Gaussian terms.The problem of power state estimation has been studied extensively.But the fundamental performance limits and the attendant decision rules are unknown when the data is potentially compromised by random bad data(due to sensor failures)or structured bad data(due to cyber attacks,which are also referred to false data injection attacks).This paper provides a general framework that formalizes the underlying connection between state estimation and bad data detection routines.We aim to carry out the combined tasks of detecting the presence of random and structured bad data,and form accurate estimations for the state of power grid.This paper characterizes the optimal detectors and estimators.Furthermore,the gains with respect to the existing state estimators and bad data detectors are established through numerical evaluations.
文摘In experimental tests, besides data in range of allowable error, the experimenters usually get some unexpected wrong data called bad points. In usual experimental data processing, the method of bad points exclusion based on automatic programming is seldom taken into consideration by researchers. This paper presents a new method to reject bad points based on Hough transform, which is modified to save computational and memory consumptions. It is fit for linear data processing and can be extended to process data that is possible to be transformed into and from linear form; curved lines, which can be effectively detected by Hough transform. In this paper, the premise is the distribution of data, such as linear distribution and exponential distribution, is predetermined. Steps of the algorithm start from searching for an approximate curve line that minimizes the sum of parameters of data points. The data points, whose parameters are above a self-adapting threshold, will be deleted. Simulation experiments have manifested that the method proposed in this paper performs efficiently and robustly.
文摘配电网环境复杂,配电网同步相量测量装置(distribution network synchronous phasor measurement unit, D-PMU)容易受到干扰而产生坏数据,进一步影响基于测量数据的应用效果。为了提高D-PMU数据质量,提出一种不依赖系统拓扑的基于密度的噪场应用空间聚类(density-based spatial clustering of applications with noise, DBSCAN)的配电网同步测量坏数据检测方法。首先利用基于密度的聚类算法DBSCAN进行异常数据检测。通过轮廓系数和邓恩指数对DBSCAN的聚类结果进行综合评价。利用麻雀搜索算法实现自适应参数调整,解决检测时需要预先处理训练、标记数据的问题。在此基础上,将时间序列聚类的K-Medoids算法和动态时间规整算法相结合,通过衡量不同时间序列之间的相似性,解决了D-PMU在电气联系较弱时对扰动数据与坏数据的区分问题,增强了数据处理的准确性与噪声环境下的稳健性。仿真和实际数据的测试结果表明,所提方法能有效区分真实扰动数据并准确识别D-PMU坏数据。
文摘为提高状态估计的抗差性,提出一种基于最大指数绝对值目标函数的状态估计(maximum exponential absolute value state estimation,MEAV)方法。首先给出了MEAV的基本模型,并介绍了其理论基础和数学性质。由于MEAV基本模型的目标函数并非处处可导,因而无法利用基于梯度的方法进行求解。为此,给出了MEAV基本模型的等价模型,并详细推导了基于原-对偶内点算法的MEAV等价模型的求解方法。算例分析表明,MEAV在估计过程中可自动抑制多个强相关不良数据,显示了良好的抗差性和较高的计算效率,因而具有良好的工程应用前景。
