Reno, Matthew J.; Blakely, Logan; Trevizan, Rodrigo D.; Pena, Bethany D.; Lave, Matthew S.; Azzolini, Joseph A.; Yusuf, Jubair Y.; Jones, Christian B.; Furlani Bastos, Alvaro F.; Chalamala, Rohit C.; Korkali, Mert K.; Sun, Chih-Che S.; Donadee, Jonathan D.; Stewart, Emma M.; Donde, Vaibhav D.; Peppanen, Jouni P.; Hernandez, Miguel H.; Deboever, Jeremiah D.; Rocha, Celso R.; Rylander, Matthew R.; Siratarnsophon, Piyapath S.; Grijalva, Santiago G.; Talkington, Samuel T.; Mason, Karl M.; Vejdan, Sadegh V.; Khan, Ahmad U.; Mbeleg, Jordan S.; Ashok, Kavya A.; Divan, Deepak D.; Li, Feng L.; Therrien, Francis T.; Jacques, Patrick J.; Rao, Vittal R.; Francis, Cody F.; Zaragoza, Nicholas Z.; Nordy, David N.; Glass, Jim G.; Holman, Derek H.; Mannon, Tim M.; Pinney, David P.
This report summarizes the work performed under a project funded by U.S. DOE Solar Energy Technologies Office (SETO), including some updates from the previous report SAND2022-0215, to use grid edge measurements to calibrate distribution system models for improved planning and grid integration of solar PV. Several physics-based data-driven algorithms are developed to identify inaccuracies in models and to bring increased visibility into distribution system planning. This includes phase identification, secondary system topology and parameter estimation, meter-to-transformer pairing, medium-voltage reconfiguration detection, determination of regulator and capacitor settings, PV system detection, PV parameter and setting estimation, PV dynamic models, and improved load modeling. Each of the algorithms is tested using simulation data and demonstrated on real feeders with our utility partners. The final algorithms demonstrate the potential for future planning and operations of the electric power grid to be more automated and data-driven, with more granularity, higher accuracy, and more comprehensive visibility into the system.
This report documents the Resilience Enhancements through Deep Learning Yields (REDLY) project, a three-year effort to improve electrical grid resilience by developing scalable methods for system operators to protect the grid against threats leading to interrupted service or physical damage. The computational complexity and uncertain nature of current real-world contingency analysis presents significant barriers to automated, real-time monitoring. While there has been a significant push to explore the use of accurate, high-performance machine learning (ML) model surrogates to address this gap, their reliability is unclear when deployed in high-consequence applications such as power grid systems. Contemporary optimization techniques used to validate surrogate performance can exploit ML model prediction errors, which necessitates the verification of worst-case performance for the models.
Efficiently embedding and/or integrating mechanistic information with data-driven models is essential if it is desired to simultaneously take advantage of both engineering principles and data-science. Further the opportunity for hybridization occurs in many scenarios, such as the development of a faster model of an accurate high-fidelity computer model; the correction of a mechanistic model that does not fully-capture the physical phenomena of the system; or the integration of a data-driven component approximating an unknown correlation within a mechanistic model. At the same time, different techniques have been proposed and applied in different literatures to achieve this hybridization, such as hybrid modeling, physics-informed Machine Learning (ML) and model calibration. In this paper we review the methods, challenges, applications and algorithms of these three research areas and discuss them in the context of the different hybridization scenarios. Moreover, we provide a comprehensive comparison of the hybridization techniques with respect to their differences and similarities, as well as advantages and limitations and future perspectives. Finally, we apply and illustrate hybrid modeling, physics-informed ML and model calibration via a chemical reactor case study.
The protection systems (circuit breakers, relays, reclosers, and fuses) of the electric grid are the primary component responding to resilience events, ranging from common storms to extreme events. The protective equipment must detect and operate very quickly, generally <0.25 seconds, to remove faults in the system before the system goes unstable or additional equipment is damaged. The burden on protection systems is increasing as the complexity of the grid increases; renewable energy resources, particularly inverter-based resources (IBR) and increasing electrification all contribute to a more complex grid landscape for protection devices. In addition, there are increasing threats from natural disasters, aging infrastructure, and manmade attacks that can cause faults and disturbances in the electric grid. The challenge for the application of AI into power system protection is that events are rare and unpredictable. In order to improve the resiliency of the electric grid, AI has to be able to learn from very little data. During an extreme disaster, it may not be important that the perfect, most optimal action is taken, but AI must be guaranteed to always respond by moving the grid toward a more stable state during unseen events.
Reno, Matthew J.; Blakely, Logan; Trevizan, Rodrigo D.; Pena, Bethany D.; Lave, Matthew S.; Azzolini, Joseph A.; Yusuf, Jubair Y.; Jones, Christian B.; Furlani Bastos, Alvaro F.; Chalamala, Rohit C.; Korkali, Mert K.; Sun, Chih-Che S.; Donadee, Jonathan D.; Stewart, Emma M.; Donde, Vaibhav D.; Peppanen, Jouni P.; Hernandez, Miguel H.; Deboever, Jeremiah D.; Rocha, Celso R.; Rylander, Matthew R.; Siratarnsophon, Piyapath S.; Grijalva, Santiago G.; Talkington, Samuel T.; Gomez-Peces, Cristian G.; Mason, Karl M.; Vejdan, Sadegh V.; Khan, Ahmad U.; Mbeleg, Jordan S.; Ashok, Kavya A.; Divan, Deepak D.; Li, Feng L.; Therrien, Francis T.; Jacques, Patrick J.; Rao, Vittal S.; Francis, Cody F.; Zaragoza, Nicholas Z.; Nordy, David N.; Glass, Jim G.
This report summarizes the work performed under a project funded by U.S. DOE Solar Energy Technologies Office (SETO) to use grid edge measurements to calibrate distribution system models for improved planning and grid integration of solar PV. Several physics-based data-driven algorithms are developed to identify inaccuracies in models and to bring increased visibility into distribution system planning. This includes phase identification, secondary system topology and parameter estimation, meter-to-transformer pairing, medium-voltage reconfiguration detection, determination of regulator and capacitor settings, PV system detection, PV parameter and setting estimation, PV dynamic models, and improved load modeling. Each of the algorithms is tested using simulation data and demonstrated on real feeders with our utility partners. The final algorithms demonstrate the potential for future planning and operations of the electric power grid to be more automated and data-driven, with more granularity, higher accuracy, and more comprehensive visibility into the system.
High penetration of solar photovoltaics can have a significant impact on the power flows and voltages in distribution systems. In order to support distribution grid planning, control and optimization, it is imperative for utilities to maintain an accurate database of the locations and sizes of PV systems. This paper extends previous work on methods to estimate the location of PV systems based on knowledge of the distribution network model and availability of voltage magnitude measurement streams. The proposed method leverages the expected impact of solar injection variations on the circuit voltage and takes into account the operation and impact of changes in voltage due to discrete voltage regulation equipment (VRE). The estimation model enables determining the most likely location of PV systems, as well as voltage regulator tap and switching capacitors state changes. The method has been tested for individual and multiple PV system, using the Chi-Square test as a metric to evaluate the goodness of fit. Simulations on the IEEE 13-bus and IEEE 123-bus distribution feeders demonstrate the ability of the method to provide consistent estimations of PV locations as well as VRE actions.
2021 IEEE Power and Energy Conference at Illinois, PECI 2021
Blakely, Logan; Reno, Matthew J.; Jones, C.B.; Furlani-Bastos, Alvaro; Nordy, David
The use of grid-edge sensing in distribution model calibration is a significant aid in reducing the time and cost associated with finding and correcting errors in the models. This work proposes a novel method for the phase identification task employing correlation coefficients on residential advanced metering infrastructure (AMI) combined with additional sensors on the medium-voltage distribution system to enable utilities to effectively calibrate the phase classification in distribution system models algorithmically. The proposed method was tested on a real utility feeder of ∼800 customers that includes 15-min voltage measurements on each phase from IntelliRupters® and 15-min AMI voltage measurements from all customers. The proposed method is compared with a standard phase identification method using voltage correlations with the substation and shows significantly improved results. The final phase predictions were verified to be correct in the field by the utility company.
Distribution system model accuracy is increasingly important and using advanced metering infrastructure (AMI) data to algorithmically identify and correct errors can dramatically reduce the time required to correct errors in the models. This work proposes a data-driven, physics-based approach for grouping residential meters downstream of the same service transformer. The proposed method involves a two-stage approach that first uses correlation coefficient analysis to identify transformers with errors in their customer grouping then applies a second stage, using a linear regression formulation, to correct the errors. This method achieved >99% accuracy in transformer groupings, demonstrated using EPRI's Ckt 5 model containing 1379 customers and 591 transformers.
Broderick, Robert J.; Reno, Matthew J.; Lave, Matthew S.; Azzolini, Joseph A.; Blakely, Logan; Galtieri, Jason G.; Mather, Barry M.; Weekley, Andrew W.; Hunsberger, Randolph H.; Chamana, Manohar C.; Li, Qinmiao L.; Zhang, Wenqi Z.; Latif, Aadil L.; Zhu, Xiangqi Z.; Grijalva, Santiago G.; Zhang, Xiaochen Z.; Deboever, Jeremiah D.; Qureshi, Muhammad U.; Therrien, Francis T.; Lacroix, Jean-Sebastien L.; Li, Feng L.; Belletête, Marc B.; Hébert, Guillaume H.; Montenegro, Davis M.; Dugan, Roger D.
The rapid increase in penetration of distributed energy resources on the electric power distribution system has created a need for more comprehensive interconnection modeling and impact analysis. Unlike conventional scenario-based studies, quasi-static time-series (QSTS) simulations can realistically model time-dependent voltage controllers and the diversity of potential impacts that can occur at different times of year. However, to accurately model a distribution system with all its controllable devices, a yearlong simulation at 1-second resolution is often required, which could take conventional computers a computational time of 10 to 120 hours when an actual unbalanced distribution feeder is modeled. This computational burden is a clear limitation to the adoption of QSTS simulations in interconnection studies and for determining optimal control solutions for utility operations. The solutions we developed include accurate and computationally efficient QSTS methods that could be implemented in existing open-source and commercial software used by utilities and the development of methods to create high-resolution proxy data sets. This project demonstrated multiple pathways for speeding up the QSTS computation using new and innovative methods for advanced time-series analysis, faster power flow solvers, parallel processing of power flow solutions and circuit reduction. The target performance level for this project was achieved with year-long high-resolution time series solutions run in less than 5 minutes within an acceptable error.
The recent growth of sensing devices on the distribution system, such as smart meter deployment, has enabled a wide variety of data-driven distribution system model calibration algorithms. A challenge associated with developing algorithms for model calibration tasks is the determination of parameters for a particular algorithm. This work proposes a method for parameter selection utilizing silhouette score analysis that allows these parameters to be tuned on a per-feeder basis. This method leverages cluster analysis and the distance matrices often produced by phase identification methods. The proposed method was tested on 5 feeders from 2 different utilities to select the number of clusters used in a spectral clustering phase identification algorithm. A synthetic dataset was then used to validate the method with the phase identification algorithm performing with 100% accuracy.
In power grid operation, optimal power flow (OPF) problems are solved several times per day to find economically optimal generator setpoints that balance given load demands. Ideally, we seek an optimal solution that is also “N-1 secure”, meaning the system can absorb contingency events such as transmission line or generator failure without loss of service. Current practice is to solve the OPF problem and then check a subset of contingencies against heuristic values, resulting in, at best, suboptimal solutions. Unfortunately, online solution of the OPF problem including the full N-1 contingencies (i.e., two-stage stochastic programming formulation) is intractable for even modest sized electrical grids. To address this challenge, this work presents an efficient method to embed N-1 security constraints into the solution of the OPF by using Neural Network (NN) models to represent the security boundary. Our approach introduces a novel sampling technique, as well as a tuneable parameter to allow operators to balance the conservativeness of the security model within the OPF problem. Our results show that we are able to solve contingency formulations of larger size grids than reported in literature using non-linear programming (NLP) formulations with embedded NN models to local optimality. Solutions found with the NN constraint have marginally increased computational time but are more secure to contingency events.
Distribution system models play a critical role in the modern grid, driving distributed energy resource integration through hosting capacity analysis and providing insight into critical areas of interest such as grid resilience and stability. Thus, the ability to validate and improve existing distribution system models is also critical. This work presents a method for identifying service transformers which contain errors in specifying the customers connected to the low-voltage side of that transformer. Pairwise correlation coefficients of the smart meter voltage time series are used to detect when a customer is not in the transformer grouping that is specified in the model. The proposed method is demonstrated both on synthetic data as well as a real utility feeder, and it successfully identifies errors in the transformer labeling in both datasets.
Calibrating distribution system models to aid in the accuracy of simulations such as hosting capacity analysis is increasingly important in the pursuit of the goal of integrating more distributed energy resources. The recent availability of smart meter data is enabling the use of machine learning tools to automatically achieve model calibration tasks. This research focuses on applying machine learning to the phase identification task, using a co-association matrix-based, ensemble spectral clustering approach. The proposed method leverages voltage time series from smart meters and does not require existing or accurate phase labels. This work demonstrates the success of the proposed method on both synthetic and real data, surpassing the accuracy of other phase identification research.
Timeseries power and voltage data recorded by electricity smart meters in the US have been shown to provide immense value to utilities when coupled with advanced analytics. However, Advanced Metering Infrastructure (AMI) has diverse characteristics depending on the utility implementing the meters. Currently, there are no specific guidelines for the parameters of data collection, such as measurement interval, that are considered optimal, and this continues to be an active area of research. This paper aims to review different grid edge, delay tolerant algorithms using AMI data and to identify the minimum granularity and type of data required to apply these algorithms to improve distribution system models. The primary focus of this report is on distribution system secondary circuit topology and parameter estimation (DSPE).
Smart grid technologies and wide-spread installation of advanced metering infrastructure (AMI) equipment present new opportunities for the use of machine learning algorithms paired with big data to improve distribution system models. Accurate models are critical in the continuing integration of distributed energy resources (DER) into the power grid, however the low-voltage models often contain significant errors. This paper proposes a novel spectral clustering approach for validating and correcting customer electrical phase labels in existing utility models using the voltage timeseries produced by AMI equipment. Spectral clustering is used in conjunction with a sliding window ensemble to improve the accuracy and scalability of the algorithm for large datasets. The proposed algorithm is tested using real data to validate or correct over 99% of customer phase labels within the primary feeder under consideration. This is over a 94% reduction in error given the 9% of customers predicted to have incorrect phase labels.
High-resolution, quasi-static time series (QSTS) simulations are essential for modeling modern distribution systems with high-penetration of distributed energy resources (DER) in order to accurately simulate the time-dependent aspects of the system. Presently, QSTS simulations are too computationally intensive for widespread industry adoption. This paper proposes to simulate a portion of the year with QSTS and to use decision tree machine learning methods, random forests and boosting ensembles, to predict the voltage regulator tap changes for the remainder of the year, accurately reproducing the results of the time-consuming, brute-force, yearlong QSTS simulation. This research uses decision tree ensemble machine learning, applied for the first time to QSTS simulations, to produce high-accuracy QSTS results, up to 4x times faster than traditional methods.