Bayesian State Estimation for Unobservable Distribution Systems via Deep Learning
Open accessThe DOI matches its Crossref record; 56 of the 58 references it cites were found in a registry or an index.
Abstract
The problem of state estimation for unobservable distribution systems is considered. A deep learning approach to Bayesian state estimation is proposed for real-time applications. The proposed technique consists of distribution learning of stochastic power injection, a Monte Carlo technique for the training of a deep neural network for state estimation, and a Bayesian bad-data detection and filtering algorithm. Structural characteristics of the deep neural networks are investigated. Simulations illustrate the accuracy of Bayesian state estimation for unobservable systems and demonstrate the benefit of employing a deep neural network. Numerical results show the robustness of Bayesian state estimation against modeling and estimation errors and the presence of bad and missing data. Comparing with pseudo-measurement techniques, direct Bayesian state estimation via deep learning neural network outperforms existing benchmarks.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“The second method is not affected by the location or type of failure.”
From A Comprehensive Review on the Role of Artificial Intelligence in Power System Stability, Control, and Protection: Insights and Future Directions (2024) · cites this paper as [91]
In the citing paper
“An ANN-based approach was recently discussed for a generalized neuron-based stabilizer [90] and a multilayer perceptron for transient stability assessment [91].”
From A Comprehensive Review on the Role of Artificial Intelligence in Power System Stability, Control, and Protection: Insights and Future Directions (2024) · cites this paper as [91]
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 2×
“The Adam algorithm [43] designed for non-stationary objectives and noisy measurements is particularly suitable.”
MethodDelving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.123 · Cited by 19,741 · cited 1×
“To select an initial point for the optimization, He’s normal method [52] was used.”
Method(2016) · Cited by 11,679
Zimmerman, R. D., Murillo-Sanchez, C. E., & Thomas, R. J. (2011) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2010.2051168 · Cited by 6,756 · cited 1×
“a) Systems Simulated: The simulations were performed on two systems defined in the MATPOWER toolbox [52].”
MethodEfficient Processing of Deep Neural Networks: A Tutorial and Survey
Sze, V., Chen, Y.-H., Yang, T.-J., & Emer, J. S. (2017) · Proceedings of the IEEE · doi:10.1109/jproc.2017.2761740 · Cited by 4,215 · cited 1×
“Special hardware that exploits massive parallelism can greatly speed up the computation [49].”
BackgroundAbur, A., & Expósito, A. G. (2004) · CRC Press eBooks · doi:10.1201/9780203913673 · Cited by 2,974 · cited 2×
“Bad-data detection and identification has been studied extensively [3], [4].”
BackgroundNeural networks for short-term load forecasting: a review and evaluation
Hippert, H. S., Pedreira, C. E., & Souza, R. C. (2001) · IEEE Transactions on Power Systems · doi:10.1109/59.910780 · Cited by 2,221
Power System Static-State Estimation, Part I: Exact Model
Schweppe, F., & Wildes, J. (1970) · IEEE Transactions on Power Apparatus and Systems · doi:10.1109/tpas.1970.292678 · Cited by 1,383
Distribution System Modeling and Analysis
Kersting, W. (2012) · Electric Power Engineering Series · doi:10.1201/b12056-31 · Cited by 1,241
All of Statistics: A Concise Course in Statistical Inference
Wasserman, L. (2004) · Springer Texts in Statistics · doi:10.1007/978-0-387-21736-9 · Cited by 1,090 · cited 3×
“All these estimators produced results well within the non-parametric confidence bounds obtained using the Dvoretzky-Kiefer-Wolfowitz inequality [45].”
BackgroundMethodAll of Nonparametric Statistics
(2006) · Springer Texts in Statistics · doi:10.1007/0-387-30623-4 · Cited by 993
Efficient algorithms for agglomerative hierarchical clustering methods
Day, W. H. E., & Edelsbrunner, H. (1984) · Journal of Classification · doi:10.1007/bf01890115 · Cited by 971 · cited 1×
“In particular, hierarchical agglomerative clustering [44] allows us to control the number and the sizes of clusters.”
BackgroundMalicious Data Attacks on the Smart Grid
Kosut, O., Jia, L., Thomas, R. J., & Tong, L. (2011) · IEEE Transactions on Smart Grid · doi:10.1109/tsg.2011.2163807 · Cited by 784 · cited 2×
“State estimation relies on data, and the quality of data depends on the types of sensors, the quality of the collection and communications, and the possibility of cyber data attacks where measurement data are manipulated by an attacker [5], [51].”
BackgroundSelf-citationState Estimation in Electric Power Systems
Monticelli, A. (1999) · Springer eBooks · doi:10.1007/978-1-4615-4999-4 · Cited by 713 · cited 2×
“Indeed, there is no general method to detect, identify, and remove bad data effectively [2].”
BackgroundMethodAll of Nonparametric Statistics
Fotopoulos, S. B. (2007) · Technometrics · doi:10.1198/tech.2007.s454 · Cited by 562 · cited 1×
“Distribution learning can be parametric or non-parametric [45], [46].”
BackgroundState estimation for real-time monitoring of distribution systems
Baran, M. E., & Kelley, A. W. (1994) · IEEE Transactions on Power Systems · doi:10.1109/59.336098 · Cited by 474
Distribution system state estimation
Lu, C. N., Teng, J. H., & Liu, W.-H. E. (1995) · IEEE Transactions on Power Systems · doi:10.1109/59.373946 · Cited by 449
Manitsas, E., Singh, R., Pal, B. C., & Strbac, G. (2012) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2012.2187804 · Cited by 371 · cited 1×
“a) ASE Performance: We compared the proposed Bayesian state estimation with deep neural network (herein abbreviated as BSEdnn) with two WLS-based pseudo-measurement methods in the literature: 1) WLSp: referred to as WLS with pseudo measurements, WLSp generates injection pseudo measurements by averaging the energy cons…”
MethodStatistical Representation of Distribution System Loads Using Gaussian Mixture Model
Singh, R., Pal, B. C., & Jabr, R. A. (2010) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2009.2030271 · Cited by 353
Neural network with unbounded activation functions is universal approximator
Sonoda, S., & Murata, N. (2017) · Applied and Computational Harmonic Analysis · doi:10.1016/j.acha.2015.12.005 · Cited by 303 · cited 1×
“, [41]) has established that a neural network with a single hidden layer is sufficient to approximate an arbitrary continuous function.”
BackgroundWasserman, L. (2021) · Cited by 282
Real-Time Power System State Estimation and Forecasting via Deep Unrolled Neural Networks
Zhang, L., Wang, G., & Giannakis, G. B. (2019) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2019.2926023 · Cited by 257
Deep Learning-Based Interval State Estimation of AC Smart Grids Against Sparse Cyber Attacks
Wang, H., Ruan, J., Wang, G., Zhou, B., Liu, Y., Fu, X., & Peng, J. (2018) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2018.2804669 · Cited by 237
Forecasting-Aided State Estimation—Part I: Panorama
Brown Do Coutto Filho, M., & de Souza, J. C. S. (2009) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2009.2030295 · Cited by 234 · cited 1×
“measurement techniques are part of the so-called forecastingaided state estimation [13], [14].”
BackgroundDistribution System Modeling and Analysis
Kersting, W. H. (2006) · CRC Press eBooks · doi:10.1201/9781420006162 · Cited by 231
State estimation for electric power distribution systems in quasi real-time conditions
Roytelman, I., & Shahidehpour, S. M. (1993) · IEEE Transactions on Power Delivery · doi:10.1109/61.248315 · Cited by 230 · cited 1×
“In some of the earliest contributions [9]–[12], it was well rec-”
BackgroundSingh, R., Pal, B. C., & Jabr, R. A. (2010) · IET Generation, Transmission & Distribution · doi:10.1049/iet-gtd.2009.0167 · Cited by 205 · cited 2×
“Bayesian methods for state estimation can be classified into two categories: Bayesian pseudo-measurements [4]–[7] and Bayesian state estimation [8]–[15]; the former uses probability distributions to generate pseudo-measurements so that conventional techniques such as WLS can be applied.”
MethodDistribution circuit state estimation using a probabilistic approach
Ghosh, A. K., Lubkeman, D. L., Downey, M. J., & Jones, R. H. (1997) · IEEE Transactions on Power Systems · doi:10.1109/59.574922 · Cited by 201 · cited 1×
“In some of the earliest contributions [9]–[12], it was well rec-”
BackgroundSubspace Methods for Data Attack on State Estimation: A Data Driven Approach
Kim, J., Tong, L., & Thomas, R. J. (2015) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2014.2385670 · Cited by 199 · cited 1×
“State estimation relies on data, and the quality of data depends on the types of sensors, the quality of the collection and communications, and the possibility of cyber data attacks where measurement data are manipulated by an attacker [5], [51].”
BackgroundSelf-citationDistribution System State Estimation Based on Nonsynchronized Smart Meters
Alimardani, A., Therrien, F., Atanackovic, D., Jatskevich, J., & Vaahedi, E. (2015) · IEEE Transactions on Smart Grid · doi:10.1109/tsg.2015.2429640 · Cited by 197 · cited 1×
“Attempts have been made to incorporate smart meter data for state estimation [6]–[8].”
Data-Driven Learning-Based Optimization for Distribution System State Estimation
Zamzam, A. S., Fu, X., & Sidiropoulos, N. D. (2019) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2019.2909150 · Cited by 197
Probabilistic Forecast of PV Power Generation Based on Higher Order Markov Chain
Sanjari, M. J., & Gooi, H. B. (2017) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2016.2616902 · Cited by 196 · cited 1×
“Because of the historical data for injections are limited, we estimate the injection distribution based on the parametric model of Gaussian mixtures commonly used to model load and renewable generations [47]–[49].”
MethodA robust state estimator for medium voltage distribution networks
Wu, J., He, Y., & Jenkins, N. (2013) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2012.2215927 · Cited by 158
Load modeling for distribution circuit state estimation
Ghosh, A. K., Lubkeman, D. L., & Jones, R. H. (1997) · IEEE Transactions on Power Delivery · doi:10.1109/61.584427 · Cited by 146
Dynamic state estimation including anomaly detection and identification for power systems
Nishiya, K.-I., Hasegawa, J., & Koike, T. (1982) · IEE Proceedings C Generation Transmission and Distribution · doi:10.1049/ip-c.1982.0032 · Cited by 124
Impact of Pseudo-Measurements From New Power Profiles on State Estimation in Low-Voltage Grids
Angioni, A., Schlösser, T., Ponci, F., & Monti, A. (2015) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2015.2454673 · Cited by 123
Particle Filter Approach to Dynamic State Estimation of Generators in Power Systems
Emami, K., Fernando, T. L., Iu, H. H., Trinh, H. M., & Wong, K. P. (2015) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2014.2366196 · Cited by 108 · cited 1×
“In our approach, instead of using Monte Carlo sampling to calculating the conditional mean directly as in [31], [32], Monte Carlo sampling is used to train a neural network that, in real-time, computes the MMSE estimate directly from the measurements.”
MethodState estimation of medium voltage distribution networks using smart meter measurements
Al-Wakeel, A., Wu, J., & Jenkins, N. (2016) · Applied Energy · doi:10.1016/j.apenergy.2016.10.010 · Cited by 103
Bayesian linear state estimation using smart meters and PMUs measurements in distribution grids
Luca Schenato, B., Barchi, G., Macii, D., Arghandeh, R., Poolla, K. R., & von Meier, A. (2014) · IEEE International Conference on Smart Grid Communications (SmartGridComm) · doi:10.1109/smartgridcomm.2014.7007708 · Cited by 96 · cited 1×
“Another approach is based on a linear approximation of the AC power flow [30].”
BackgroundPower System Tracking State Estimation and Bad Data Processing
Falcao, D. M., Cooke, P. A., & Brameller, A. (1982) · IEEE Transactions on Power Apparatus and Systems · doi:10.1109/tpas.1982.317110 · Cited by 87 · cited 1×
“Several such techniques [33]–[36] are based on exploiting a dynamic model to predicted the current measurement using past measurements, from which the prediction error becomes test statistics for bad-data detection.”
BackgroundA Belief Propagation Based Power Distribution System State Estimator
Hu, Y., Kuh, A., Yang, T., & Kavcic, A. (2011) · IEEE Computational Intelligence Magazine · doi:10.1109/mci.2011.941589 · Cited by 87 · cited 1×
“One approach to calculating conditional statistics is based on a graphical model of the distribution system from which belief propagation techniques are used to generate state estimates [28], [29].”
BackgroundForecasting-Aided State Estimation—Part II: Implementation
Filho, M. B. D. C., de Souza, J. C. S., & Freund, R. S. (2009) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2009.2030297 · Cited by 83 · cited 1×
“measurement techniques are part of the so-called forecastingaided state estimation [13], [14].”
BackgroundNeural networks and pseudo-measurements for real-time monitoring of distribution systems
Bernieri, A., Betta, G., Liguori, C., & Losi, A. (1996) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/19.492803 · Cited by 68
Modelling of pseudo-measurements for distribution system state estimation
Manitsas, E., Singh, R., Pal, B. C., & Štrbac, G. (2008) · CIRED Seminar 2008: SmartGrids for Distribution · doi:10.1049/ic:20080447 · Cited by 56
Salehfar, H., & Zhao, R. D. (1995) · Electric Power Systems Research · doi:10.1016/0378-7796(95)00966-7 · Cited by 53 · cited 2×
“†The authors of [37] did not use the autoencoder concept to explain their approach.”
BackgroundMethodDistributed Power System State Estimation Using Factor Graphs
Chavali, P., & Nehorai, A. (2015) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2015.2413297 · Cited by 46
A pre-estimation filtering process of bad data for linear power systems state estimators using PMUs
Pignati, M., Zanni, L., Sarri, S., Cherkaoui, R., Boudec, J. L., & Paolone, M. (2014) · Power Systems Computation Conference · doi:10.1109/pscc.2014.7038329 · Cited by 34 · cited 1×
“Several such techniques [33]–[36] are based on exploiting a dynamic model to”
BackgroundAdinolfi, F., D’Agostino, F., Morini, A., Saviozzi, M., & Silvestro, F. (2014) · IEEE PES Innovative Smart Grid Technologies Europe · doi:10.1109/isgteurope.2014.7028770 · Cited by 33
State Estimation for Unobservable Distribution Systems via Deep Neural Networks
Mestav, K. R., Luengo-Rozas, J., & Tong, L. (2018) · IEEE Power Engineering Society General Meeting · doi:10.1109/pesgm.2018.8586649 · Cited by 33 · cited 1×
“See [1] for details of network configuration.”
Self-citationDistribution state estimation: A necessary requirement for the smart grid
Lefebvre, S., Prevost, J., & Lenoir, L. (2014) · doi:10.1109/pesgm.2014.6939030 · Cited by 32 · cited 1×
“A compelling case can be made that effective state estimation is essential to optimize DER in real-time operations [2].”
BackgroundA three-phase state estimation in unbalanced distribution networks with switch modelling
Majumdar, A., & Pal, B. C. (2016) · IEEE International Conference on Control, Measurement and Instrumentation · doi:10.1109/cmi.2016.7413793 · Cited by 29 · cited 1×
“The specific forms and parameters of (1) with different levels of modeling details can be found in [38], [39].”
BackgroundState estimation for unbalanced electric power distribution systems using AMI data
Gao, Y., & Yu, N. (2017) · IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT) · doi:10.1109/isgt.2017.8085999 · Cited by 23 · cited 1×
“Attempts have been made to incorporate smart meter data for state estimation [6]–[8].”
Abur, A., Keyhani, A., & Bakhtiari, H. (1987) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.1987.4335167 · Cited by 22
Abdel-Majeed, A., Kattmann, C., Tenbohlen, S., & Saur, R. (2014) · IEEE Power Engineering Society General Meeting · doi:10.1109/pesgm.2014.6938843 · Cited by 19
Reduced Model for Power System State Estimation Using Artificial Neural Networks
Onwuachumba, A., Wu, Y., & Musavi, M. T. (2013) · IEEE Green Technologies Conference (GreenTech) · doi:10.1109/greentech.2013.69 · Cited by 14
Neural networks and pseudo-measurements for real-time monitoring of distribution systems
Bernieri, A., Liguori, C., & Losi, A. (2002) · Conference proceedings - IEEE Instrumentation/Measurement Technology Conference · doi:10.1109/imtc.1995.515112 · Cited by 9 · cited 1×
“Bayesian methods for state estimation can be classified into two categories: Bayesian pseudo-measurements [4]–[7] and Bayesian state estimation [8]–[15]; the former uses probability distributions to generate pseudo-measurements so that conventional techniques such as WLS can be applied.”
Method
Research works printed but not linked: 1
- mestav, 2019, Bayesian state estimation for unobservable distribution systems via deep neural networks, IEEE Trans Power Syst, 34, 4910, 10.1109/TPWRS.2019.2919157
Cited by
Physics-Aware Neural Networks for Distribution System State Estimation
Zamzam, A. S., & Sidiropoulos, N. D. (2020) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2020.2988352 · Cited by 242
“The fully connected deep NN model proposed in [14] is used in the first and second baselines.”
MethodAlhamrouni, I., Abdul Kahar, N. H., Salem, M., Swadi, M., Zahroui, Y., Kadhim, D. J., Mohamed, F. A., & Alhuyi Nazari, M. (2024) · Applied Sciences · doi:10.3390/app14146214 · Cited by 217
“An ANN-based approach was recently discussed for a generalized neuron-based stabilizer [90] and a multilayer perceptron for transient stability assessment [91].”
MethodDeep learning in power systems research: A review
(2020) · CSEE Journal of Power and Energy Systems · doi:10.17775/cseejpes.2020.02700 · Cited by 193
A Survey of Power System State Estimation Using Multiple Data Sources: PMUs, SCADA, AMI, and Beyond
Cheng, G., Lin, Y., Abur, A., Gómez-Expósito, A., & Wu, W. (2024) · IEEE Transactions on Smart Grid · doi:10.1109/tsg.2023.3286401 · Cited by 186
“By contrast, data-driven SE is typically developed based on probabilistic inference-based methods [57], [185], [186], [187], [188] or machine learning-based methods [112], [189], [190], [191] without the need of accurate or complete physical models.”
BackgroundMethodInfluentialArtificial intelligence techniques for enabling Big Data services in distribution networks: A review
Barja-Martinez, S., Aragüés-Peñalba, M., Munné-Collado, Í., Lloret-Gallego, P., Bullich-Massagué, E., & Villafafila-Robles, R. (2021) · Renewable and Sustainable Energy Reviews · doi:10.1016/j.rser.2021.111459 · Cited by 174
Metadata
- Authors
- Mestav, K. R., Luengo-Rozas, J., & Tong, L.
- Affiliations
- Cornell University (US)
- Venue
- IEEE Transactions on Power Systems
- Published
- 2019-11
- Volume
- 34
- Issue
- 6
- Pages
- 4910-4920
- ISSN
- 0885-8950 (ISSN-L)
- 1558-0679
- OpenAlex
- W2963590654
- Cited by
- 216
- References
- 58
- Citations by year
- 2026: 30 · 2025: 25 · 2024: 34 · 2023: 49 · 2022: 30 · 2021: 29 · 2020: 14 · 2019: 5
- FWCI
- 7.01
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Power Systems (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Science Foundation (1809830, 1816387)
- Power System Engineering Research Center
- Fulbright Scholar Program, and Iberdrola Foundation
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Power System Optimization and Stability
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Optimal Power Flow Distribution
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Smart Grid and Power Systems
- Keywords
- Bayesian state estimation, deep neural network, bad data detection, missing data, real-time state estimation, distribution learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.