Pub Date : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587517
B. Zeng, Xuan Wei, Jiahuan Feng
Under the smart-grid environment, demand response (DR) provides an equivalent reserve resource to mitigate operational uncertainties, in addition to the supply-side solutions. Thus, identifying the effect of DR to service reliability turns to be essential for strategic planning decisions. In this paper, a novel data-driven dispatching approach for sustainable exploitation of DR capabilities in future smart-grids is proposed. Differing to existing studies, the user willingness factor attended with DR is especially focused in this work. To achieve this, we develop a two-term DR model, wherein the compliance of customers is characterized as a dynamic self-optimizing process that specified by the regret measure regarding historical payoffs. On this basis, a data-driven-based DR scheduling model is formulated from the grid’s point of view. It could permit desired tradeoffs between the system reliability target and sustainability of DR provision. To verify the effectiveness of the proposed approach, a hybrid algorithm embedded with sequential Monte-Carlo simulations is developed. Numerical experiments are conducted to illustrate the performance of the proposed method based on a real-world distribution network.
{"title":"A Data-Driven Dispatching Approach for Sustainable Exploitation of Demand Response Resources","authors":"B. Zeng, Xuan Wei, Jiahuan Feng","doi":"10.1109/SmartGridComm.2018.8587517","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587517","url":null,"abstract":"Under the smart-grid environment, demand response (DR) provides an equivalent reserve resource to mitigate operational uncertainties, in addition to the supply-side solutions. Thus, identifying the effect of DR to service reliability turns to be essential for strategic planning decisions. In this paper, a novel data-driven dispatching approach for sustainable exploitation of DR capabilities in future smart-grids is proposed. Differing to existing studies, the user willingness factor attended with DR is especially focused in this work. To achieve this, we develop a two-term DR model, wherein the compliance of customers is characterized as a dynamic self-optimizing process that specified by the regret measure regarding historical payoffs. On this basis, a data-driven-based DR scheduling model is formulated from the grid’s point of view. It could permit desired tradeoffs between the system reliability target and sustainability of DR provision. To verify the effectiveness of the proposed approach, a hybrid algorithm embedded with sequential Monte-Carlo simulations is developed. Numerical experiments are conducted to illustrate the performance of the proposed method based on a real-world distribution network.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117224434","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 : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587531
Ashfaq Ahmad, J. Khan
Self consumption and user comfort are two important metrics to evaluate efficiency and quality-of-service (QoS) of an energy management technique in stand-alone distributed photovoltaic (PV) systems. Prior work focuses on a joint problem of maximizing the two metrics, however, every user demand is variable and uncertain, and PV output power is highly vulnerable to weather variations. In consequence, the joint problem has non linearities at a given instant, on a given day and in a given weather condition. The extent of these non linearities increases with the consideration of high temporal resolution. If these non linearities are well addressed, would lead to significant improvement in system efficiency and user QoS. In this paper, we propose an artificial neural network (ANN) based technique to solve the joint optimization problem with inherent non linearities. Our proposed technique is scalable to user tasks, and adaptable to temporal resolution and the non linearities. Simulation results validate effectiveness of the proposed technique in terms of the selected performance metrics.
{"title":"Stand-Alone Distributed PV Systems: Maximizing Self Consumption and User Comfort using ANNs","authors":"Ashfaq Ahmad, J. Khan","doi":"10.1109/SmartGridComm.2018.8587531","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587531","url":null,"abstract":"Self consumption and user comfort are two important metrics to evaluate efficiency and quality-of-service (QoS) of an energy management technique in stand-alone distributed photovoltaic (PV) systems. Prior work focuses on a joint problem of maximizing the two metrics, however, every user demand is variable and uncertain, and PV output power is highly vulnerable to weather variations. In consequence, the joint problem has non linearities at a given instant, on a given day and in a given weather condition. The extent of these non linearities increases with the consideration of high temporal resolution. If these non linearities are well addressed, would lead to significant improvement in system efficiency and user QoS. In this paper, we propose an artificial neural network (ANN) based technique to solve the joint optimization problem with inherent non linearities. Our proposed technique is scalable to user tasks, and adaptable to temporal resolution and the non linearities. Simulation results validate effectiveness of the proposed technique in terms of the selected performance metrics.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"19 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121009610","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 : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587554
Hyungeun Choi, Seunghyoung Ryu, Hongseok Kim
Recent development of artificial intelligence (AI) makes AI applicable to diverse fields, and the smart grid is not an exception. In particular, there have been extensive researches on load forecasting using deep learning. Most existing studies have been conducted on deep neural network (DNN) and recurrent neural network (RNN). Very recently, CNN with shallow network has been studied for short-term load forecasting (STLF). In this paper, we propose a novel framework based on ResNet/LSTM combined model. The proposed model has two steps. First, ResNet extracts latent features of daily and weekly load data. Then, LSTM is applied to train the encoded feature vector with dynamics, and make prediction suitable for volatile load data. By leveraging ResNet and LSTM, the proposed model has the advantage of forecasting load data that has both regularity and inconsistency. To demonstrate the performance, we compare the proposed model with other deep learning models: multi-layer perceptron (MLP), ResNet, LSTM and ResNet/MLP combined model. The results show that the proposed ResNet/LSTM combined model has 21.3% of MAPE improvement in overall, and 25.8% of MAPE improvement for the bottom 25% group in terms of MAPE compared to MLP.
{"title":"Short-Term Load Forecasting based on ResNet and LSTM","authors":"Hyungeun Choi, Seunghyoung Ryu, Hongseok Kim","doi":"10.1109/SmartGridComm.2018.8587554","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587554","url":null,"abstract":"Recent development of artificial intelligence (AI) makes AI applicable to diverse fields, and the smart grid is not an exception. In particular, there have been extensive researches on load forecasting using deep learning. Most existing studies have been conducted on deep neural network (DNN) and recurrent neural network (RNN). Very recently, CNN with shallow network has been studied for short-term load forecasting (STLF). In this paper, we propose a novel framework based on ResNet/LSTM combined model. The proposed model has two steps. First, ResNet extracts latent features of daily and weekly load data. Then, LSTM is applied to train the encoded feature vector with dynamics, and make prediction suitable for volatile load data. By leveraging ResNet and LSTM, the proposed model has the advantage of forecasting load data that has both regularity and inconsistency. To demonstrate the performance, we compare the proposed model with other deep learning models: multi-layer perceptron (MLP), ResNet, LSTM and ResNet/MLP combined model. The results show that the proposed ResNet/LSTM combined model has 21.3% of MAPE improvement in overall, and 25.8% of MAPE improvement for the bottom 25% group in terms of MAPE compared to MLP.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121141747","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 : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587577
Pietro Danzi, Sarah Hambridge, Č. Stefanović, P. Popovski
In the power grid, the Balance Responsible Parties (BRPs) purchase energy based on a forecast of the user consumption. The forecasts are imperfect, and the corrections of their real-time deviations are managed by a System Operator (SO), which charges the BRPs for the procured imbalances. Flexible consumers, associated with a BRP, can be involved in a demand response (DR) program to reduce the imbalance costs. However, running the DR program requires the BRP to invest resources in the infrastructure and increases its operating costs. To limit the intervention of BRP, we implement the DR via a blockchain smart contract. Moreover, to reduce the delay of publication of the imbalance price, caused by the inefficient accounting process of the current balancing markets, a second blockchain is adopted at the SO layer, procuring a fast and auditable credit settlements. The feasibility of the proposed architecture is evaluated over an Ethereum blockchain platform. The results show that block chains can enable a high automation of the balancing market, by providing (i) the implementation of aggregators with low operating cost and (ii) the timely and transparent access to the balancing information, thus fostering new business models for the BRPs.
{"title":"Blockchain-Based and Multi-Layered Electricity Imbalance Settlement Architecture","authors":"Pietro Danzi, Sarah Hambridge, Č. Stefanović, P. Popovski","doi":"10.1109/SmartGridComm.2018.8587577","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587577","url":null,"abstract":"In the power grid, the Balance Responsible Parties (BRPs) purchase energy based on a forecast of the user consumption. The forecasts are imperfect, and the corrections of their real-time deviations are managed by a System Operator (SO), which charges the BRPs for the procured imbalances. Flexible consumers, associated with a BRP, can be involved in a demand response (DR) program to reduce the imbalance costs. However, running the DR program requires the BRP to invest resources in the infrastructure and increases its operating costs. To limit the intervention of BRP, we implement the DR via a blockchain smart contract. Moreover, to reduce the delay of publication of the imbalance price, caused by the inefficient accounting process of the current balancing markets, a second blockchain is adopted at the SO layer, procuring a fast and auditable credit settlements. The feasibility of the proposed architecture is evaluated over an Ethereum blockchain platform. The results show that block chains can enable a high automation of the balancing market, by providing (i) the implementation of aggregators with low operating cost and (ii) the timely and transparent access to the balancing information, thus fostering new business models for the BRPs.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125327522","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 : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587494
Marcus Voss, Christian Bender-Saebelkampf, S. Albayrak
Low aggregations of electric load profiles are more fluctuating, relative forecast errors are comparatively high, and it has been shown that different forecast models and feature configurations may be best suitable for specific households or buildings. However, at low aggregations, the monetary incentive for manual feature engineering and model selection is low, as benefits from forecast improvements are small. Convolutional Neural Networks (CNN) have proven to achieve high accuracy in an end-to-end fashion with minimal effort for manual feature selection. WaveNet, a CNN-based approach, has been developed to handle noisy time-series data for speech recognition and synthesis. In this work we explore if WaveNet is suitable for short-term forecasts of lowly aggregated electric loads. We find that WaveNet performs similarly to, and slightly better than, typical benchmark models for individual households, at the cost of higher model complexity. Preliminary experiments show that transfer learning can further improve results and decrease training times for individual households, as a pattern such as the correlation between outside temperature and load can be learned as general features. For aggregations of 10–200 households WaveNet improves most over the benchmarks, e.g., 13% compared to vanilla Artificial Neural Networks at 200 households, making it possibly suitable for aggregated load forecasting.
{"title":"Residential Short-Term Load Forecasting Using Convolutional Neural Networks","authors":"Marcus Voss, Christian Bender-Saebelkampf, S. Albayrak","doi":"10.1109/SmartGridComm.2018.8587494","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587494","url":null,"abstract":"Low aggregations of electric load profiles are more fluctuating, relative forecast errors are comparatively high, and it has been shown that different forecast models and feature configurations may be best suitable for specific households or buildings. However, at low aggregations, the monetary incentive for manual feature engineering and model selection is low, as benefits from forecast improvements are small. Convolutional Neural Networks (CNN) have proven to achieve high accuracy in an end-to-end fashion with minimal effort for manual feature selection. WaveNet, a CNN-based approach, has been developed to handle noisy time-series data for speech recognition and synthesis. In this work we explore if WaveNet is suitable for short-term forecasts of lowly aggregated electric loads. We find that WaveNet performs similarly to, and slightly better than, typical benchmark models for individual households, at the cost of higher model complexity. Preliminary experiments show that transfer learning can further improve results and decrease training times for individual households, as a pattern such as the correlation between outside temperature and load can be learned as general features. For aggregations of 10–200 households WaveNet improves most over the benchmarks, e.g., 13% compared to vanilla Artificial Neural Networks at 200 households, making it possibly suitable for aggregated load forecasting.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126859025","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 : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587605
T. Pedersen, Laurynas Siksnys, B. Neupane
The recent spread of distributed renewable energy sources and smart IoT devices offer exciting new possibilities for the use of energy flexibility, opening a new era of the so-called bottom-up or cellular energy systems. In order to harness the full potential of flexibility, flexibility has to be modeled and represented in a manner that can be efficiently managed, manipulated, and traded on a market. In this paper, we provide a comprehensive overview of the FlexOffer concept, which offers an effective way of modeling and managing energy demand and supply flexibilities from a wide range of flexible resources and their aggregates. First, we define the basic concept and present the different phases of the FlexOffer life-cycle. Then, we discuss more advanced internal FlexOffer constraints as well as algorithms for FlexOffer generation, aggregation, disaggregation, and pricing that can significantly reduce energy management and trading complexities and increase overall efficiency. Finally, we present a general decentralized system architecture for trading flexibility (FlexOffers) in existing and new markets. Our experimental results show that (1) FlexOffers can be extracted with up to 98% accuracy, (2) aggregation and disaggregation can scale to 1000K FlexOffers and more, and (3) flexibility can be traded in the NordPool flexi order market while providing up to 89.9% (of optimal) reduction in the energy cost.
{"title":"Modeling and Managing Energy Flexibility Using FlexOffers","authors":"T. Pedersen, Laurynas Siksnys, B. Neupane","doi":"10.1109/SmartGridComm.2018.8587605","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587605","url":null,"abstract":"The recent spread of distributed renewable energy sources and smart IoT devices offer exciting new possibilities for the use of energy flexibility, opening a new era of the so-called bottom-up or cellular energy systems. In order to harness the full potential of flexibility, flexibility has to be modeled and represented in a manner that can be efficiently managed, manipulated, and traded on a market. In this paper, we provide a comprehensive overview of the FlexOffer concept, which offers an effective way of modeling and managing energy demand and supply flexibilities from a wide range of flexible resources and their aggregates. First, we define the basic concept and present the different phases of the FlexOffer life-cycle. Then, we discuss more advanced internal FlexOffer constraints as well as algorithms for FlexOffer generation, aggregation, disaggregation, and pricing that can significantly reduce energy management and trading complexities and increase overall efficiency. Finally, we present a general decentralized system architecture for trading flexibility (FlexOffers) in existing and new markets. Our experimental results show that (1) FlexOffers can be extracted with up to 98% accuracy, (2) aggregation and disaggregation can scale to 1000K FlexOffers and more, and (3) flexibility can be traded in the NordPool flexi order market while providing up to 89.9% (of optimal) reduction in the energy cost.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"241 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115593780","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 : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587525
Sheng Wang, Yi Ding, Changzheng Shao
Demand response (DR) is a framework that allows flexible load (FL) to self-schedule, including being curtailed or shifted to maintain system balance between energy supply and demand. With the integration of multi-energy system (MES) and development of information and communication technologies (ICTs), multi-energy infrastructures have expanded the ways FL participates in DR program. FL can shift to another energy carrier without noticeable delay. However, the chronological behavior and economic assessment for such DR methods have not been comprehensively discussed yet. This paper proposed a generalized self-scheduling model for demand side in MES. Firstly, the chronological response potentials for multi-energy FLs are explored. Moreover, the appliance-level economic loss of both load curtailment and shifting are calculated based on customer damage function. The optimization of self-scheduling is formulated as a mixed integer programing problem and solved by genetic algorithm. A test case based on energy hub is formed to illustrate the proposed modeling technique.
{"title":"Generalized Modeling of Self-scheduling Demand Resource in Multi-Energy System","authors":"Sheng Wang, Yi Ding, Changzheng Shao","doi":"10.1109/SmartGridComm.2018.8587525","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587525","url":null,"abstract":"Demand response (DR) is a framework that allows flexible load (FL) to self-schedule, including being curtailed or shifted to maintain system balance between energy supply and demand. With the integration of multi-energy system (MES) and development of information and communication technologies (ICTs), multi-energy infrastructures have expanded the ways FL participates in DR program. FL can shift to another energy carrier without noticeable delay. However, the chronological behavior and economic assessment for such DR methods have not been comprehensively discussed yet. This paper proposed a generalized self-scheduling model for demand side in MES. Firstly, the chronological response potentials for multi-energy FLs are explored. Moreover, the appliance-level economic loss of both load curtailment and shifting are calculated based on customer damage function. The optimization of self-scheduling is formulated as a mixed integer programing problem and solved by genetic algorithm. A test case based on energy hub is formed to illustrate the proposed modeling technique.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122937035","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 : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587541
Mehdi Zeinali, J. Thompson
Internet based communications is a necessary solution for enabling smart grid services such as demand side management, automatic metering infrastructure, and virtual power plants. However, the speed and reliability of the Internet for providing smart grid services needs practical investigation. In this paper, we evaluate the robustness of the U.K. Internet network for demand response services, based on the latency, packet loss and jitter for smart grid communication. We use the ping tool to identify the minimum achievable latency within a national internet topology and the analysis has also been extended to consumer internet access. Further, the internet connectivity of consumer’s premises has been evaluated regarding suitability of these solutions for demand response applications.
{"title":"Practical Evaluation of UK Internet Network Characteristics For Demand-Side Response Applications","authors":"Mehdi Zeinali, J. Thompson","doi":"10.1109/SmartGridComm.2018.8587541","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587541","url":null,"abstract":"Internet based communications is a necessary solution for enabling smart grid services such as demand side management, automatic metering infrastructure, and virtual power plants. However, the speed and reliability of the Internet for providing smart grid services needs practical investigation. In this paper, we evaluate the robustness of the U.K. Internet network for demand response services, based on the latency, packet loss and jitter for smart grid communication. We use the ping tool to identify the minimum achievable latency within a national internet topology and the analysis has also been extended to consumer internet access. Further, the internet connectivity of consumer’s premises has been evaluated regarding suitability of these solutions for demand response applications.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129496702","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 : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587518
Michael J. Rausch, V. Krishna, Peng Gu, Rupak Chandra, B. Feddersen, Ahmed M. Fawaz, W. Sanders
Wireless IoT mesh networks are being widely deployed for use in applications such as operational technology networks in power grids, city-scale surveillance, and monitoring. The benefits of such networks, which may include mission critical communications, can be undermined by an adversary who launches denial-of-service (DoS) attacks on them. In this paper, we present a peer-to-peer approach to detecting and localizing such adversaries by leveraging the topology of the mesh network. In doing so, we make three main contributions. First, we present insights from a preliminary implementation on a standards-based IoT platform used in real smart meter deployments. Second, we propose an optimal choice of peers that can help detect a jammed node, while minimizing the risk that the peers themselves are jammed. Finally, we present a tool to help generate datasets of city-scale IoT mesh topologies for simulation studies.
{"title":"Peer-to-peer Detection of DoS Attacks on City-Scale IoT Mesh Networks","authors":"Michael J. Rausch, V. Krishna, Peng Gu, Rupak Chandra, B. Feddersen, Ahmed M. Fawaz, W. Sanders","doi":"10.1109/SmartGridComm.2018.8587518","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587518","url":null,"abstract":"Wireless IoT mesh networks are being widely deployed for use in applications such as operational technology networks in power grids, city-scale surveillance, and monitoring. The benefits of such networks, which may include mission critical communications, can be undermined by an adversary who launches denial-of-service (DoS) attacks on them. In this paper, we present a peer-to-peer approach to detecting and localizing such adversaries by leveraging the topology of the mesh network. In doing so, we make three main contributions. First, we present insights from a preliminary implementation on a standards-based IoT platform used in real smart meter deployments. Second, we propose an optimal choice of peers that can help detect a jammed node, while minimizing the risk that the peers themselves are jammed. Finally, we present a tool to help generate datasets of city-scale IoT mesh topologies for simulation studies.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121485575","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 : 2018-10-01DOI: 10.1109/SmartGridComm.2018.8587454
Seunghyoung Ryu, Hyungeun Choi, Hyoseop Lee, Hongseok Kim, V. Wong
In energy data analytics, load profile clustering is essential for various smart grid applications such as demand response, load forecasting, and tariff design. Most of the conventional clustering techniques are based on a representative time domain load profile within a certain period, and the daily and seasonal variations are not well captured. In this paper, we propose a deep learning based customer load profile clustering framework that jointly captures daily and seasonal variations. By leveraging convolutional autoencoder (CAE), the yearly load profile in the time domain is converted into a representative vector in the smaller dimensional encoded space. The clusters are then determined based on the vectors encoded by the CAE. We apply the proposed framework to 1,405 households' yearly load profiles and verify that the trained CAE can encode those load profiles into approximately 100 times smaller dimensional space. The encoded load profiles can be decoded by the CAE with a negligible loss between 1–3%. The clustered load images can visualize both daily and seasonal variations, and clustering in the encoded space speeds up the clustering process by almost three orders of magnitude.
{"title":"Residential Load Profile Clustering via Deep Convolutional Autoencoder","authors":"Seunghyoung Ryu, Hyungeun Choi, Hyoseop Lee, Hongseok Kim, V. Wong","doi":"10.1109/SmartGridComm.2018.8587454","DOIUrl":"https://doi.org/10.1109/SmartGridComm.2018.8587454","url":null,"abstract":"In energy data analytics, load profile clustering is essential for various smart grid applications such as demand response, load forecasting, and tariff design. Most of the conventional clustering techniques are based on a representative time domain load profile within a certain period, and the daily and seasonal variations are not well captured. In this paper, we propose a deep learning based customer load profile clustering framework that jointly captures daily and seasonal variations. By leveraging convolutional autoencoder (CAE), the yearly load profile in the time domain is converted into a representative vector in the smaller dimensional encoded space. The clusters are then determined based on the vectors encoded by the CAE. We apply the proposed framework to 1,405 households' yearly load profiles and verify that the trained CAE can encode those load profiles into approximately 100 times smaller dimensional space. The encoded load profiles can be decoded by the CAE with a negligible loss between 1–3%. The clustered load images can visualize both daily and seasonal variations, and clustering in the encoded space speeds up the clustering process by almost three orders of magnitude.","PeriodicalId":213523,"journal":{"name":"2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128134338","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：