Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102729
M. Farajollahi, Aslan Mojallal, M. R. Dadash Zadeh
Active power control is one of the crucial functions run by the microgrid controller to schedule distributed energy resources (DER) in a microgrid. This schedule can be obtained through solving an optimization problem, so-called optimal dispatch, to minimize the cost of microgrid operation. To match all possible cases that can occur during real-time operation as well as to ensure that the optimal dispatch always reaches a feasible solution, some of the constraints associated with optimal dispatch problem are defined in terms of soft constraints, which introduces penalty factors into the problem. The user needs to tune these penalty factors, which might be confusing and cumbersome. In this regard, there is a considerable value in easing the hassle of tuning penalty factors. This paper first proposes a method to reduce the number of penalty factors used in the microgrid optimal dispatch problem. Additionally, the penalty factors are introduced in such a form that users can easily tune them all together with minimum effort.
{"title":"Effective Microgrid Optimal Dispatch Settings","authors":"M. Farajollahi, Aslan Mojallal, M. R. Dadash Zadeh","doi":"10.1109/GridEdge54130.2023.10102729","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102729","url":null,"abstract":"Active power control is one of the crucial functions run by the microgrid controller to schedule distributed energy resources (DER) in a microgrid. This schedule can be obtained through solving an optimization problem, so-called optimal dispatch, to minimize the cost of microgrid operation. To match all possible cases that can occur during real-time operation as well as to ensure that the optimal dispatch always reaches a feasible solution, some of the constraints associated with optimal dispatch problem are defined in terms of soft constraints, which introduces penalty factors into the problem. The user needs to tune these penalty factors, which might be confusing and cumbersome. In this regard, there is a considerable value in easing the hassle of tuning penalty factors. This paper first proposes a method to reduce the number of penalty factors used in the microgrid optimal dispatch problem. Additionally, the penalty factors are introduced in such a form that users can easily tune them all together with minimum effort.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125168662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102719
Utkarsh Kumar, Fei Ding
This paper presents a resilience-oriented cellular grid formation approach to achieve scalable and reconfigurable community microgrid operations for distribution systems with behind-the-meter distributed energy resources. A set of interconnected solar photovoltaics, energy storage systems, and load is termed as a cell, implying a subset of the grid that can operate independently using its own resources. Cells are identified such that each cell inherently has sufficient energy resources to black start and can provide a certain level of backup power for its load under the loss of utility power supply. The proposed cell formation approach builds on a unique self-organizing map-based method (SomRes) to quantify a system’s resilience. Using SomRes and a non-dominated sorting-based genetic algorithm (NSGA-II), a fast and efficient cell formation algorithm is developed to identify cells in a distribution system that are resilient against extreme events. The efficacy of the proposed approach is demonstrated on a numerical model of a real distribution feeder in Colorado, United States.
{"title":"A Novel Resilience-Oriented Cellular Grid Formation Approach for Distribution Systems with Behind-the-Meter Distributed Energy Resources","authors":"Utkarsh Kumar, Fei Ding","doi":"10.1109/GridEdge54130.2023.10102719","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102719","url":null,"abstract":"This paper presents a resilience-oriented cellular grid formation approach to achieve scalable and reconfigurable community microgrid operations for distribution systems with behind-the-meter distributed energy resources. A set of interconnected solar photovoltaics, energy storage systems, and load is termed as a cell, implying a subset of the grid that can operate independently using its own resources. Cells are identified such that each cell inherently has sufficient energy resources to black start and can provide a certain level of backup power for its load under the loss of utility power supply. The proposed cell formation approach builds on a unique self-organizing map-based method (SomRes) to quantify a system’s resilience. Using SomRes and a non-dominated sorting-based genetic algorithm (NSGA-II), a fast and efficient cell formation algorithm is developed to identify cells in a distribution system that are resilient against extreme events. The efficacy of the proposed approach is demonstrated on a numerical model of a real distribution feeder in Colorado, United States.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"22 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117006841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102750
J. M. Moloney, S. Williamson, Cameron L. Hall
Stability of power grids in the presence of small fluctuations in frequency is important for the reliable and robust transmission of electricity. Recent research suggests damping parameters can be optimized for each generator in the grid to ensure the strong stability of the network. However, results are typically demonstrated on systems with aggregated “generators” and “loads” as opposed to the full power grid. We demonstrate that, under many circumstances, the response of aggregated systems is very different to that of the corresponding full system. In the cases considered, optimizing damping parameters based on an aggregated system can lead to comparatively poor stability performance when these parameters are applied to the full system.
{"title":"Generator Aggregation and Power Grid Stability","authors":"J. M. Moloney, S. Williamson, Cameron L. Hall","doi":"10.1109/GridEdge54130.2023.10102750","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102750","url":null,"abstract":"Stability of power grids in the presence of small fluctuations in frequency is important for the reliable and robust transmission of electricity. Recent research suggests damping parameters can be optimized for each generator in the grid to ensure the strong stability of the network. However, results are typically demonstrated on systems with aggregated “generators” and “loads” as opposed to the full power grid. We demonstrate that, under many circumstances, the response of aggregated systems is very different to that of the corresponding full system. In the cases considered, optimizing damping parameters based on an aggregated system can lead to comparatively poor stability performance when these parameters are applied to the full system.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"167 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121363665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102708
S. S. Varghese, G. Joós, S. Q. Ali
Ultra fast charging stations enable the charging of electric vehicles under 15 minutes. Integration of the charging station with renewable source of energy and energy storage system alleviates the stress on the distribution grid during each charging process. The proposed energy management system ensures the coordinated response of the charging station assets to meet the electric vehicle demand while reducing the charging impacts on the grid. A day ahead rolling horizon energy management algorithm that updates the dispatch every five minutes is proposed which follows the change in the load and the renewable generation. Strategies and case studies considering variable renewable energy availability and electric vehicle load based on performance indices such as net energy imported from the grid, energy not met ratio, number of grid power violation events, and metrics for independence and self-consumption of renewable resources are evaluated, and data is presented.
{"title":"Energy Management of Ultra Fast Charging Stations","authors":"S. S. Varghese, G. Joós, S. Q. Ali","doi":"10.1109/GridEdge54130.2023.10102708","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102708","url":null,"abstract":"Ultra fast charging stations enable the charging of electric vehicles under 15 minutes. Integration of the charging station with renewable source of energy and energy storage system alleviates the stress on the distribution grid during each charging process. The proposed energy management system ensures the coordinated response of the charging station assets to meet the electric vehicle demand while reducing the charging impacts on the grid. A day ahead rolling horizon energy management algorithm that updates the dispatch every five minutes is proposed which follows the change in the load and the renewable generation. Strategies and case studies considering variable renewable energy availability and electric vehicle load based on performance indices such as net energy imported from the grid, energy not met ratio, number of grid power violation events, and metrics for independence and self-consumption of renewable resources are evaluated, and data is presented.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114840729","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102741
Mazhar Ali, X. Gao, A. Rahman, Musabbir Hossain, Wei Sun
A cyber-physical-system (CPS) framework serves as the foundation for the modern architecture of critical infrastructures, including electrical power grids, oil and natural gas distribution, transportation systems, and many more. Although the CPS design has assisted in the reliable, efficient, and robust service of physical systems, concurrently it has raised tremendous concern about secure and resilient operation due to the coupling and the presence of several hardware and software products among different CPS layers. These vulnerabilities have opened up a new avenue for adversaries to deploy coordinated cyber-physical attacks, with difficult detection, isolation, and recovery of the CPS. In this paper, we use smart grids and telecommunication networks as an example to elaborate on emerging challenges and analyze new types of coordinated cyber-physical attacks, such as multistage and multiwave attacks. A general model is presented for multistage and multiwave attacks along with adaptive restoration strategies for resilient recovery. The proposed framework is applicable to other CPS as well.
{"title":"Emerging Coordinated Cyber-Physical-Systems Attacks and Adaptive Restoration Strategies","authors":"Mazhar Ali, X. Gao, A. Rahman, Musabbir Hossain, Wei Sun","doi":"10.1109/GridEdge54130.2023.10102741","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102741","url":null,"abstract":"A cyber-physical-system (CPS) framework serves as the foundation for the modern architecture of critical infrastructures, including electrical power grids, oil and natural gas distribution, transportation systems, and many more. Although the CPS design has assisted in the reliable, efficient, and robust service of physical systems, concurrently it has raised tremendous concern about secure and resilient operation due to the coupling and the presence of several hardware and software products among different CPS layers. These vulnerabilities have opened up a new avenue for adversaries to deploy coordinated cyber-physical attacks, with difficult detection, isolation, and recovery of the CPS. In this paper, we use smart grids and telecommunication networks as an example to elaborate on emerging challenges and analyze new types of coordinated cyber-physical attacks, such as multistage and multiwave attacks. A general model is presented for multistage and multiwave attacks along with adaptive restoration strategies for resilient recovery. The proposed framework is applicable to other CPS as well.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128336887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102706
Xin Xu, C. Mishra, Chen Wang, Kevin D. Jones, J. Starling, R. Gardner, L. Vanfretti
Existing methods for synchrophasor data analysis focus on oscillations that can be interpreted as a sum of sinusoids with or without damping, however, they are more difficult to apply when considering other types of waveforms. This paper investigates how to apply spectral estimation analysis methods when analyzing a periodic voltage sag whose waveform is comparable to a periodic pulse train. A methodology is proposed where the estimated power spectral density and spectrograms are used to detect the area impacted by a periodic sag and to identify its dominant propagation path, which helps with the localization of the disturbance’s spread. The proposed methodology is applied to measurement data obtained from a region in Dominion Energy’s service territory that is geographically bounded by other utilities, which validates the effectiveness of the proposed method in a real-world utility setting.
{"title":"Tracking Periodic Voltage Sags via Synchrophasor Data in a Geographically Bounded Service Territory","authors":"Xin Xu, C. Mishra, Chen Wang, Kevin D. Jones, J. Starling, R. Gardner, L. Vanfretti","doi":"10.1109/GridEdge54130.2023.10102706","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102706","url":null,"abstract":"Existing methods for synchrophasor data analysis focus on oscillations that can be interpreted as a sum of sinusoids with or without damping, however, they are more difficult to apply when considering other types of waveforms. This paper investigates how to apply spectral estimation analysis methods when analyzing a periodic voltage sag whose waveform is comparable to a periodic pulse train. A methodology is proposed where the estimated power spectral density and spectrograms are used to detect the area impacted by a periodic sag and to identify its dominant propagation path, which helps with the localization of the disturbance’s spread. The proposed methodology is applied to measurement data obtained from a region in Dominion Energy’s service territory that is geographically bounded by other utilities, which validates the effectiveness of the proposed method in a real-world utility setting.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122692644","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102705
Alex Nassif, K. Wheeler
Inverter-based renewable generation resources are proliferating as a response to environmental policy. Along with these variable forms of generation comes the application of battery energy storage systems that are necessary to level off generation as well as provide system support that in many jurisdictions can include ramp rate regulation. They can also enable high levels of renewable penetration by contributing to system inertia, ancillary services near critical facilities, reducing transmission security violations, and orderly islanding, with the objective of improving system resilience. It is well known that the costs of renewable generation and energy storage have been following a descending trend which has led to a gradually higher adoption level. These inverter-based resources, however, create new problems for electrical utilities planners and engineers. One such issue, which has been studied recently, is how to measure, test, and manage load rejection overvoltage. This phenomenon takes place upon sudden islanding of a power system area such that it becomes supported by grid-following inverter-based resources only. This paper presents background, practical methods to test the behavior, as well as two case studies of utility-scale generation and energy storage connected to a distribution feeder.
{"title":"Modeling and Measurement of Load Rejection Overvoltage of Inverter-Based Resources Interconnected to Distribution Feeders","authors":"Alex Nassif, K. Wheeler","doi":"10.1109/GridEdge54130.2023.10102705","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102705","url":null,"abstract":"Inverter-based renewable generation resources are proliferating as a response to environmental policy. Along with these variable forms of generation comes the application of battery energy storage systems that are necessary to level off generation as well as provide system support that in many jurisdictions can include ramp rate regulation. They can also enable high levels of renewable penetration by contributing to system inertia, ancillary services near critical facilities, reducing transmission security violations, and orderly islanding, with the objective of improving system resilience. It is well known that the costs of renewable generation and energy storage have been following a descending trend which has led to a gradually higher adoption level. These inverter-based resources, however, create new problems for electrical utilities planners and engineers. One such issue, which has been studied recently, is how to measure, test, and manage load rejection overvoltage. This phenomenon takes place upon sudden islanding of a power system area such that it becomes supported by grid-following inverter-based resources only. This paper presents background, practical methods to test the behavior, as well as two case studies of utility-scale generation and energy storage connected to a distribution feeder.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128579004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102724
Lola Botman, J. Lago, Thijs Becker, O. Agudelo, K. Vanthournout, B. De Moor
Short-term individual household load forecasting is relevant for several applications and low voltage grid (LVG) stakeholders, e.g., for grid simulations, operation planning, congestion anticipation or advance payments. Electrical consumption at the household level is highly stochastic, point forecasting cannot capture this efficiently. To have insights about the uncertainty of the prediction, probabilistic methods should be developed. We propose a method to predict the half-hourly consumption of individual households one day ahead, based on a neural network, enhanced with empirical quantiles based on the point forecasts errors. The method is scalable thanks to its low computational requirements. Additionally, it requires only historical data and calendar features. Finally, the method is evaluated in a case study where it achieves state-of-the-art accuracy.
{"title":"A scalable method for probabilistic short-term forecasting of individual households consumption in low voltage grids","authors":"Lola Botman, J. Lago, Thijs Becker, O. Agudelo, K. Vanthournout, B. De Moor","doi":"10.1109/GridEdge54130.2023.10102724","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102724","url":null,"abstract":"Short-term individual household load forecasting is relevant for several applications and low voltage grid (LVG) stakeholders, e.g., for grid simulations, operation planning, congestion anticipation or advance payments. Electrical consumption at the household level is highly stochastic, point forecasting cannot capture this efficiently. To have insights about the uncertainty of the prediction, probabilistic methods should be developed. We propose a method to predict the half-hourly consumption of individual households one day ahead, based on a neural network, enhanced with empirical quantiles based on the point forecasts errors. The method is scalable thanks to its low computational requirements. Additionally, it requires only historical data and calendar features. Finally, the method is evaluated in a case study where it achieves state-of-the-art accuracy.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125980468","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102735
Ranyu Shi, Ali Menati, Le Xie
With increasing activities in cryptocurrencies and their fast-growing global mining demand comes new opportunities and challenges facing the electric energy systems. From the electric grid perspective, the key challenges are how to properly monitor and predict cryptocurrency mining demand at wholesale and retail levels. While large-scale mining companies connected to the transmission level can use directly instrument sensors to monitor their mining demand, how to monitor behind-the-meter cryptocurrency mining demand is still an open question. In this paper, we propose an edge-level distribution level Bitcoin mining detection scheme that utilizes smart meter data to detect the on/off status of the mining machines and estimates the power consumption magnitude of the mining load in each house. We investigate the performance of our algorithm with different Bitcoin load variations representing a wide range of possible mining devices and behaviors. Numerical results suggest that the proposed algorithm can detect both the on/off status of these loads with above 94% accuracy and calculate its load magnitude with less than 16% error for common ASIC miners. Building upon this method, aggregators could coordinate individual household mining loads for participation in demand response programs that help reduce peak demand and increase social welfare.
{"title":"Non-intrusive Monitoring of Edge-level Cryptocurrency Mining in Power Distribution Grids","authors":"Ranyu Shi, Ali Menati, Le Xie","doi":"10.1109/GridEdge54130.2023.10102735","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102735","url":null,"abstract":"With increasing activities in cryptocurrencies and their fast-growing global mining demand comes new opportunities and challenges facing the electric energy systems. From the electric grid perspective, the key challenges are how to properly monitor and predict cryptocurrency mining demand at wholesale and retail levels. While large-scale mining companies connected to the transmission level can use directly instrument sensors to monitor their mining demand, how to monitor behind-the-meter cryptocurrency mining demand is still an open question. In this paper, we propose an edge-level distribution level Bitcoin mining detection scheme that utilizes smart meter data to detect the on/off status of the mining machines and estimates the power consumption magnitude of the mining load in each house. We investigate the performance of our algorithm with different Bitcoin load variations representing a wide range of possible mining devices and behaviors. Numerical results suggest that the proposed algorithm can detect both the on/off status of these loads with above 94% accuracy and calculate its load magnitude with less than 16% error for common ASIC miners. Building upon this method, aggregators could coordinate individual household mining loads for participation in demand response programs that help reduce peak demand and increase social welfare.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126168839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-10DOI: 10.1109/GridEdge54130.2023.10102751
A. Crain, E. Rebello, Adam Sherwood, Darren Jang
A simple neural network state-of-charge predictor trained on one-year of energy storage system data is presented. The model uses the active power command and the state-of-charge for the current time-step, and implements a nonlinear auto-regressive network with exogenous inputs to predict the state-of-charge at the subsequent time-step. The neural network training algorithm is written in the Julia programming language, independent of any existing machine learning platforms; the resulting model is compared to one developed using Python/TensorFlow. The simulation performance was validated with data collected from the energy storage system that was dispatched to follow a standard frequency regulation duty cycle not used as part of the training data. The mean-absolute-error between the predicted state of charge and the validation data is shown to be less then 1%, despite the limited data and lack of physical information about the system.
{"title":"Development of a NARX State-of-Charge Predictor based on Active Power Demand","authors":"A. Crain, E. Rebello, Adam Sherwood, Darren Jang","doi":"10.1109/GridEdge54130.2023.10102751","DOIUrl":"https://doi.org/10.1109/GridEdge54130.2023.10102751","url":null,"abstract":"A simple neural network state-of-charge predictor trained on one-year of energy storage system data is presented. The model uses the active power command and the state-of-charge for the current time-step, and implements a nonlinear auto-regressive network with exogenous inputs to predict the state-of-charge at the subsequent time-step. The neural network training algorithm is written in the Julia programming language, independent of any existing machine learning platforms; the resulting model is compared to one developed using Python/TensorFlow. The simulation performance was validated with data collected from the energy storage system that was dispatched to follow a standard frequency regulation duty cycle not used as part of the training data. The mean-absolute-error between the predicted state of charge and the validation data is shown to be less then 1%, despite the limited data and lack of physical information about the system.","PeriodicalId":377998,"journal":{"name":"2023 IEEE PES Grid Edge Technologies Conference & Exposition (Grid Edge)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127016423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：