Pub Date : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960498
Evelise de Godoy Antunes, Pierre Haessig, Chaoyu Wang, R. Leborgne
Microgrid sizing optimization is often formulated as a black-box optimization problem. This allows modeling the microgrid with a realistic temporal simulation of the energy flows between components. Such models are usually optimized with gradient-free methods, because no analytical expression for gradient is available. However, the development of new Automatic Differentiation (AD) packages allows the efficient and exact computation of the gradient of black-box models. Thus, this work proposes to solve the optimal microgrid sizing using gradient-based algorithms with AD packages. However, physical realism of the model makes the objective function discontinuous which hinders the optimization convergence. After an appropriate smoothing, the objective is still nonconvex, but convergence is achieved for more that 90 % of the starting points. This suggest that a multi-start gradient-based algorithm can improve the state-of-the-art sizing methodologies.
{"title":"Optimal Microgrid Sizing using Gradient-based Algorithms with Automatic Differentiation","authors":"Evelise de Godoy Antunes, Pierre Haessig, Chaoyu Wang, R. Leborgne","doi":"10.1109/ISGT-Europe54678.2022.9960498","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960498","url":null,"abstract":"Microgrid sizing optimization is often formulated as a black-box optimization problem. This allows modeling the microgrid with a realistic temporal simulation of the energy flows between components. Such models are usually optimized with gradient-free methods, because no analytical expression for gradient is available. However, the development of new Automatic Differentiation (AD) packages allows the efficient and exact computation of the gradient of black-box models. Thus, this work proposes to solve the optimal microgrid sizing using gradient-based algorithms with AD packages. However, physical realism of the model makes the objective function discontinuous which hinders the optimization convergence. After an appropriate smoothing, the objective is still nonconvex, but convergence is achieved for more that 90 % of the starting points. This suggest that a multi-start gradient-based algorithm can improve the state-of-the-art sizing methodologies.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117073310","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 : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960467
Tohid Behdadnia, G. Deconinck
Malicious attacks in the cyber-physical power systems (CPPS) can eventually result in cascading failure and even widespread blackout, if not rectified in a timely manner. The probability of success of most of these attacks mainly depends on their timeliness, as the degree of system vulnerabilities varies from time to time by changing its operating state. In this paper, we propose a new denial of service (DoS) attack strategy where the attackers leverage learning capabilities of convolutional neural networks in the encrypted domain to forecast the optimal time of launching a DoS attack. In our simulations Internet Protocol Security (IPsec) is used to secure communication channels between phasor measurement units, phasor data concentrators, and the regional/national control center. It is illustrated that, despite providing confidential communication channels by IPsec-based security gateways, an attacker still can estimate the future operating state of the power system in advance. This gives an opportunity to the attackers for initiating an effective DoS attack. The proposed method is validated by the simulation results, which show a significant increase in the success rate of DoS attacks.
{"title":"A New Deep Learning-Based Strategy for Launching Timely DoS Attacks in PMU-Based Cyber-Physical Power Systems","authors":"Tohid Behdadnia, G. Deconinck","doi":"10.1109/ISGT-Europe54678.2022.9960467","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960467","url":null,"abstract":"Malicious attacks in the cyber-physical power systems (CPPS) can eventually result in cascading failure and even widespread blackout, if not rectified in a timely manner. The probability of success of most of these attacks mainly depends on their timeliness, as the degree of system vulnerabilities varies from time to time by changing its operating state. In this paper, we propose a new denial of service (DoS) attack strategy where the attackers leverage learning capabilities of convolutional neural networks in the encrypted domain to forecast the optimal time of launching a DoS attack. In our simulations Internet Protocol Security (IPsec) is used to secure communication channels between phasor measurement units, phasor data concentrators, and the regional/national control center. It is illustrated that, despite providing confidential communication channels by IPsec-based security gateways, an attacker still can estimate the future operating state of the power system in advance. This gives an opportunity to the attackers for initiating an effective DoS attack. The proposed method is validated by the simulation results, which show a significant increase in the success rate of DoS attacks.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"549 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121442191","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 : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960324
I. Lukić, Kristijan Čvek, K. Fekete, Marina Dubravac
This paper presents the analysis of voltage profile and active power losses in a low voltage distribution network (0.4 kV) with connected prosumers that have installed photovoltaic (PV) power plants in their houses. A computer model of three radial low voltage feeders is created in DIgSILENT PowerFactory software for simulation purposes. The analysis is done for various scenarios of PV active and reactive power production. Conventional as well as optimal power flows (OPF) are analyzed and compared. In two scenarios, optimal power flows are used to optimize voltage profile and losses using PVs reactive power injections. The obtained results indicate that the OPF did not achieve a significant reduction in active power losses. However, a noticeable improvement in voltage profile is accomplished regarding conventional power flow.
{"title":"Analysis of Losses and Voltages in Prosumer-Rich Distribution Feeders","authors":"I. Lukić, Kristijan Čvek, K. Fekete, Marina Dubravac","doi":"10.1109/ISGT-Europe54678.2022.9960324","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960324","url":null,"abstract":"This paper presents the analysis of voltage profile and active power losses in a low voltage distribution network (0.4 kV) with connected prosumers that have installed photovoltaic (PV) power plants in their houses. A computer model of three radial low voltage feeders is created in DIgSILENT PowerFactory software for simulation purposes. The analysis is done for various scenarios of PV active and reactive power production. Conventional as well as optimal power flows (OPF) are analyzed and compared. In two scenarios, optimal power flows are used to optimize voltage profile and losses using PVs reactive power injections. The obtained results indicate that the OPF did not achieve a significant reduction in active power losses. However, a noticeable improvement in voltage profile is accomplished regarding conventional power flow.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121661755","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 : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960404
Saqib Iqbal, K. Mehran, M. Nasir
Neighborhood level power-sharing is a key feature in DC microgrids (DCMGs) for resource balancing among distributed users having intermittent distributed generators (DGs) and varying load consumption. For resource sharing in DCMGs, power losses predominantly consist of distribution losses and power electronic conversion losses. Depending on the power scheduling matrix, under varying load consumption and DGs output, both of these losses have varying contributions to the system’s overall losses. For optimal energy sharing in a distributed DCMG, the power scheduling decisions need to be made considering the system’s overall power losses. The traditional optimal power flow algorithms do not account for the power electronic losses, therefore, they fail to guarantee the overall system loss minimization. In this work, we first presented a detailed analysis of semiconductor losses. Subsequently, the main components of converter losses which have influence on converter efficiency with varying output power are discussed. Further, we proposed a non-linear optimization framework that allows the users to share their resources (surplus generation or loads) in a DCMG network while keeping the overall system’s losses to a minimum. Distribution losses are calculated using a Newton-Raphson method and power electronic conversion losses are modeled as a non-linear function of converter output power against its nominal power rating. Both of these losses are collectively used in optimization framework to minimized the system’s losses. The proposed model is formulated in the standard form of optimization using OptimProblem available in Matlab 2020 and applied to a DCMG having multiple DGs, energy storage systems (ESS) and load consumption units. Results show that the total system losses can be significantly reduced up to 30-40% with the proposed optimization framework in comparison to the traditional standalone distribution losses based optimization framework.
{"title":"A Scheme for Resource Sharing in Distributed DC Microgrids with Minimal System Losses","authors":"Saqib Iqbal, K. Mehran, M. Nasir","doi":"10.1109/ISGT-Europe54678.2022.9960404","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960404","url":null,"abstract":"Neighborhood level power-sharing is a key feature in DC microgrids (DCMGs) for resource balancing among distributed users having intermittent distributed generators (DGs) and varying load consumption. For resource sharing in DCMGs, power losses predominantly consist of distribution losses and power electronic conversion losses. Depending on the power scheduling matrix, under varying load consumption and DGs output, both of these losses have varying contributions to the system’s overall losses. For optimal energy sharing in a distributed DCMG, the power scheduling decisions need to be made considering the system’s overall power losses. The traditional optimal power flow algorithms do not account for the power electronic losses, therefore, they fail to guarantee the overall system loss minimization. In this work, we first presented a detailed analysis of semiconductor losses. Subsequently, the main components of converter losses which have influence on converter efficiency with varying output power are discussed. Further, we proposed a non-linear optimization framework that allows the users to share their resources (surplus generation or loads) in a DCMG network while keeping the overall system’s losses to a minimum. Distribution losses are calculated using a Newton-Raphson method and power electronic conversion losses are modeled as a non-linear function of converter output power against its nominal power rating. Both of these losses are collectively used in optimization framework to minimized the system’s losses. The proposed model is formulated in the standard form of optimization using OptimProblem available in Matlab 2020 and applied to a DCMG having multiple DGs, energy storage systems (ESS) and load consumption units. Results show that the total system losses can be significantly reduced up to 30-40% with the proposed optimization framework in comparison to the traditional standalone distribution losses based optimization framework.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115302451","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 : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960347
H. Nemati, L. Sigrist, Luis Rouco Rodríguez, P. Sánchez-Martín, Álvaro Ortega
Optimization algorithms formulated to define the joint participation of Energy Storage Systems (ESSs) in energy and reserve markets often lead to unfeasibilities related to the available energy stored in the ESS, particularly if a relatively long-time horizon is considered (e.g., 24 hours). This paper addresses this issue and proposes an ESS model that assigns a specific amount of energy for up or down reserve provision according to the needs of ESS operator. Generally, ESSs do not participate on their own in the aforementioned markets, but rather, they usually operate jointly with stochastic non-dispatchable Renewable Energy Sources (RESs) in the form of a Virtual Power Plant (VPP). The proposed model allows operators to avoid possible unfeasibilities, and the potential penalties resulting from deviating from the day-ahead market (DAM) and secondary reserve market (SRM) offers. The model is implemented for a VPP consisting of a wind farm, a solar PV plant, and an ESS. The effectiveness of the model for bidding in joint markets is validated by several case studies.
{"title":"Addressing Unfeasibilities of Energy Storage Systems Participating in Energy and Reserve Markets","authors":"H. Nemati, L. Sigrist, Luis Rouco Rodríguez, P. Sánchez-Martín, Álvaro Ortega","doi":"10.1109/ISGT-Europe54678.2022.9960347","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960347","url":null,"abstract":"Optimization algorithms formulated to define the joint participation of Energy Storage Systems (ESSs) in energy and reserve markets often lead to unfeasibilities related to the available energy stored in the ESS, particularly if a relatively long-time horizon is considered (e.g., 24 hours). This paper addresses this issue and proposes an ESS model that assigns a specific amount of energy for up or down reserve provision according to the needs of ESS operator. Generally, ESSs do not participate on their own in the aforementioned markets, but rather, they usually operate jointly with stochastic non-dispatchable Renewable Energy Sources (RESs) in the form of a Virtual Power Plant (VPP). The proposed model allows operators to avoid possible unfeasibilities, and the potential penalties resulting from deviating from the day-ahead market (DAM) and secondary reserve market (SRM) offers. The model is implemented for a VPP consisting of a wind farm, a solar PV plant, and an ESS. The effectiveness of the model for bidding in joint markets is validated by several case studies.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125394720","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 : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960318
C. Silva, P. Faria, B. Canizes, Z. Vale
The Smart Cities concept is evolving from the project stage to the real world. Appliances become smarter and enable bidirectional communication – a huge step toward implementing demand response and empowering active consumers in the energy market. However, managing local communities with these new players is complex, and the entity behind them needs the right knowledge and tools. The authors thereby propose a methodology to manage the active consumers on DR events optimally. The study in the present paper is done from a real-time perspective. The distribution system operator detects a voltage violation and requests a load reduction to the Aggregator, the 96 active consumers, through load shifting. The proposed methodology was applied to eight scenarios, and the correct number of demand response participants for the case study was found.
{"title":"Real-Time Approach for Managing Power Network by Shifting Electricity Consumers Demand","authors":"C. Silva, P. Faria, B. Canizes, Z. Vale","doi":"10.1109/ISGT-Europe54678.2022.9960318","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960318","url":null,"abstract":"The Smart Cities concept is evolving from the project stage to the real world. Appliances become smarter and enable bidirectional communication – a huge step toward implementing demand response and empowering active consumers in the energy market. However, managing local communities with these new players is complex, and the entity behind them needs the right knowledge and tools. The authors thereby propose a methodology to manage the active consumers on DR events optimally. The study in the present paper is done from a real-time perspective. The distribution system operator detects a voltage violation and requests a load reduction to the Aggregator, the 96 active consumers, through load shifting. The proposed methodology was applied to eight scenarios, and the correct number of demand response participants for the case study was found.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126239712","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 : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960571
E. Schaefer, Bart Homan, Gerwin Hoogsteen, J. Hurink, R. V. Leeuwen
This paper investigates the effect of adding energy storage to a metro DC microgrid in order to recuperate otherwise dissipated energy produced through regenerative braking. For this, a power load profile is constructed for a metro station using existing measured railcar data, which is then used to derive the requirements for storing regenerative braking energy. It is found that introducing a storage can lead to an almost complete recuperation of otherwise dissipated energy. The derived requirements for the storage are then used to investigate a selection of possible scenarios more in depth. These scenarios include both smart controlled and uncontrolled scenarios, and with the usage of an ideal storage device as well as a (non-ideal) flywheel. Simulation results confirm that all braking energy can be recuperated through the addition of energy storage to the microgrid.
{"title":"Recuperation of railcar braking energy using energy storage at station level","authors":"E. Schaefer, Bart Homan, Gerwin Hoogsteen, J. Hurink, R. V. Leeuwen","doi":"10.1109/ISGT-Europe54678.2022.9960571","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960571","url":null,"abstract":"This paper investigates the effect of adding energy storage to a metro DC microgrid in order to recuperate otherwise dissipated energy produced through regenerative braking. For this, a power load profile is constructed for a metro station using existing measured railcar data, which is then used to derive the requirements for storing regenerative braking energy. It is found that introducing a storage can lead to an almost complete recuperation of otherwise dissipated energy. The derived requirements for the storage are then used to investigate a selection of possible scenarios more in depth. These scenarios include both smart controlled and uncontrolled scenarios, and with the usage of an ideal storage device as well as a (non-ideal) flywheel. Simulation results confirm that all braking energy can be recuperated through the addition of energy storage to the microgrid.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114216722","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 : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960511
Chhith Chhlonh, M. Alvarez‐Herault, V. Vai, B. Raison
This paper addresses the planning of low-voltage microgrid by comparing its structure with and without PV and storage in a rural area for over 30 years. The comparison aims to minimize the investment costs and get a reliable system. First, the loads (households) have been surveyed in the targeted areas and their power consumptions estimated. Then, using the Shortest Path (SP) algorithm, the connections from loads to electrical poles are conceived for the whole network. Next, the Mixed Integer Linear Programming (MILP) algorithm has been considered to balance the three-phase system. After that, the Genetic Algorithm (GA) is employed to determine the location and sizing of PV and decentralized batteries. A centralized battery is set to store the reversed power from the network to the grid and provide the power back when the PV modules produce less energy. Finally, an economic analysis validates the interest of such planning procedure. Located in Cambodia, three cases studies are implemented into the proposed method to get the results. Based on the results from simulation, the systems with integrated PV and storage proves to have a better performance.
{"title":"Comparative Planning of LVAC for Microgrid Topologies With PV-Storage in Rural Areas – Cases Study in Cambodia","authors":"Chhith Chhlonh, M. Alvarez‐Herault, V. Vai, B. Raison","doi":"10.1109/ISGT-Europe54678.2022.9960511","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960511","url":null,"abstract":"This paper addresses the planning of low-voltage microgrid by comparing its structure with and without PV and storage in a rural area for over 30 years. The comparison aims to minimize the investment costs and get a reliable system. First, the loads (households) have been surveyed in the targeted areas and their power consumptions estimated. Then, using the Shortest Path (SP) algorithm, the connections from loads to electrical poles are conceived for the whole network. Next, the Mixed Integer Linear Programming (MILP) algorithm has been considered to balance the three-phase system. After that, the Genetic Algorithm (GA) is employed to determine the location and sizing of PV and decentralized batteries. A centralized battery is set to store the reversed power from the network to the grid and provide the power back when the PV modules produce less energy. Finally, an economic analysis validates the interest of such planning procedure. Located in Cambodia, three cases studies are implemented into the proposed method to get the results. Based on the results from simulation, the systems with integrated PV and storage proves to have a better performance.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114793444","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 : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960569
J. S. Nightingale, Yingjie Wang, Fairouz Zobiri, M. A. Mustafa
When applied to short-term energy consumption forecasting, the federated learning framework allows for the creation of a predictive model without sharing raw data. There is a limit to the accuracy achieved by standard federated learning due to the heterogeneity of the individual clients' data, especially in the case of electricity data, where prediction of peak demand is a challenge. A set of clustering techniques has been explored in the literature to improve prediction quality while maintaining user privacy. These studies have mainly been conducted using sets of clients with similar attributes that may not reflect real-world consumer diversity. This paper explores, implements and compares these clustering techniques for privacy-preserving load forecasting on a representative electricity consumption dataset. The experimental results demonstrate the effects of electricity consumption heterogeneity on federated forecasting and a non-representative sample's impact on load forecasting.
{"title":"Effect of Clustering in Federated Learning on Non-IID Electricity Consumption Prediction","authors":"J. S. Nightingale, Yingjie Wang, Fairouz Zobiri, M. A. Mustafa","doi":"10.1109/ISGT-Europe54678.2022.9960569","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960569","url":null,"abstract":"When applied to short-term energy consumption forecasting, the federated learning framework allows for the creation of a predictive model without sharing raw data. There is a limit to the accuracy achieved by standard federated learning due to the heterogeneity of the individual clients' data, especially in the case of electricity data, where prediction of peak demand is a challenge. A set of clustering techniques has been explored in the literature to improve prediction quality while maintaining user privacy. These studies have mainly been conducted using sets of clients with similar attributes that may not reflect real-world consumer diversity. This paper explores, implements and compares these clustering techniques for privacy-preserving load forecasting on a representative electricity consumption dataset. The experimental results demonstrate the effects of electricity consumption heterogeneity on federated forecasting and a non-representative sample's impact on load forecasting.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115931640","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 : 2022-10-10DOI: 10.1109/ISGT-Europe54678.2022.9960626
Serguei Fominykh
Serbia has a great potential for the implementation of carbon capture and storage projects in the oil and gas industry. The recently adopted Climate Change Act provides the stimulation mechanism named Clean Development Mechanism (CDM) that proved its efficiency under the Kyoto Protocol. CDM would motivate the major industry players, especially Naftna Industrija Srbije j.s.c. (NIS), to focus on carbon capture and storage (CCS) projects aiming to benefit from the reduction of CO2 emissions and contribute to the overall sustainable development of Serbia. Currently, the Serbian Government has not enabled CDM implementation under the Climate Change Act via by-laws due to uncertainties in the world carbon markets. The CDM should be enabled in the nearest future, as any delay in deployment of CCS would negatively affect the Republic of Serbia’s target to reduce its GHG emissions by 9.8% below 1990 levels by 2030, and NIS’s goal to foster CCS technologies.
{"title":"The Decarbonisation Perspectives: The Implementation of Carbon Capture and Storage Projects in the Oil and Gas Industry Under the Serbian Climate Change Act","authors":"Serguei Fominykh","doi":"10.1109/ISGT-Europe54678.2022.9960626","DOIUrl":"https://doi.org/10.1109/ISGT-Europe54678.2022.9960626","url":null,"abstract":"Serbia has a great potential for the implementation of carbon capture and storage projects in the oil and gas industry. The recently adopted Climate Change Act provides the stimulation mechanism named Clean Development Mechanism (CDM) that proved its efficiency under the Kyoto Protocol. CDM would motivate the major industry players, especially Naftna Industrija Srbije j.s.c. (NIS), to focus on carbon capture and storage (CCS) projects aiming to benefit from the reduction of CO2 emissions and contribute to the overall sustainable development of Serbia. Currently, the Serbian Government has not enabled CDM implementation under the Climate Change Act via by-laws due to uncertainties in the world carbon markets. The CDM should be enabled in the nearest future, as any delay in deployment of CCS would negatively affect the Republic of Serbia’s target to reduce its GHG emissions by 9.8% below 1990 levels by 2030, and NIS’s goal to foster CCS technologies.","PeriodicalId":311595,"journal":{"name":"2022 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116069494","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：