Pub Date : 2021-08-11DOI: 10.1109/SEGE52446.2021.9534987
Zeqing Wu, Weishen Chu
With the development of the Internet of things (IoT), energy consumption of smart buildings has been widely concerned. The prediction of building energy consumption is of great significance for energy conservation and environmental protection as well as the construction of smart city. With the development of artificial intelligence, machine learning technology has been introduced to energy consumption prediction. In this study, multiple learning algorithms including Support Vector Regression (SVR), Artificial Neural Network (ANN), Random Forest (RF) are developed to perform energy consumption prediction. The most appropriate machine learning algorithm for energy consumption prediction has been investigated and found to be the random forest algorithm. Based on the developed machine learning models, studies on the sampling strategy for energy consumption prediction have been conducted. It is found that the variance of data has a significant effect on the prediction accuracy, and a better prediction result can be achieved by increasing the sampling density over the data with high variance. This result can be used to optimize the machine learning algorithm for building energy consumption prediction and improve the computational efficiency.
{"title":"Sampling Strategy Analysis of Machine Learning Models for Energy Consumption Prediction","authors":"Zeqing Wu, Weishen Chu","doi":"10.1109/SEGE52446.2021.9534987","DOIUrl":"https://doi.org/10.1109/SEGE52446.2021.9534987","url":null,"abstract":"With the development of the Internet of things (IoT), energy consumption of smart buildings has been widely concerned. The prediction of building energy consumption is of great significance for energy conservation and environmental protection as well as the construction of smart city. With the development of artificial intelligence, machine learning technology has been introduced to energy consumption prediction. In this study, multiple learning algorithms including Support Vector Regression (SVR), Artificial Neural Network (ANN), Random Forest (RF) are developed to perform energy consumption prediction. The most appropriate machine learning algorithm for energy consumption prediction has been investigated and found to be the random forest algorithm. Based on the developed machine learning models, studies on the sampling strategy for energy consumption prediction have been conducted. It is found that the variance of data has a significant effect on the prediction accuracy, and a better prediction result can be achieved by increasing the sampling density over the data with high variance. This result can be used to optimize the machine learning algorithm for building energy consumption prediction and improve the computational efficiency.","PeriodicalId":438266,"journal":{"name":"2021 IEEE 9th International Conference on Smart Energy Grid Engineering (SEGE)","volume":"89 S1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114101342","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 : 2021-08-11DOI: 10.1109/SEGE52446.2021.9534958
Quanjing Zhang, Didi Liu, Hongbin Chen, Junxiu Liu, Cong Hu
Energy storage can save end user costs in local energy markets that have time-varying pricing. However, energy storage device incur fixed acquisition costs which depend on their capacity. End user is faced with sophisticated energy scheduling tradeoffs in the local energy markets to account for these costs. In this paper, we consider a typical energy usage scenario where the end user draws energy from multiple types of energy supplies: the local power provider, the external power grid, and the user’s own energy storage device. Our objective is to minimize the user’s total costs (the total of purchased energy and storage) while meeting their energy demand in each time slot. Furthermore, the end user’s energy demand, the local power supplier’s prices, and the external power grid prices all vary over time. To deal with this variability, we formulated the energy scheduling problem as a stochastic optimization. We propose a dynamic algorithm based on Lyapunov optimization, and it is theoretically proved that the proposed algorithm can make the optimization target infinitely close to optimum. Finally, the effectiveness of the proposed algorithm is verified by simulation comparison. The algorithm provides a tool for end user energy scheduling where the user is equipped with energy storage device.
{"title":"Dynamic Energy Scheduling Algorithm for an End-user with Energy Storage Device to Save Total Costs","authors":"Quanjing Zhang, Didi Liu, Hongbin Chen, Junxiu Liu, Cong Hu","doi":"10.1109/SEGE52446.2021.9534958","DOIUrl":"https://doi.org/10.1109/SEGE52446.2021.9534958","url":null,"abstract":"Energy storage can save end user costs in local energy markets that have time-varying pricing. However, energy storage device incur fixed acquisition costs which depend on their capacity. End user is faced with sophisticated energy scheduling tradeoffs in the local energy markets to account for these costs. In this paper, we consider a typical energy usage scenario where the end user draws energy from multiple types of energy supplies: the local power provider, the external power grid, and the user’s own energy storage device. Our objective is to minimize the user’s total costs (the total of purchased energy and storage) while meeting their energy demand in each time slot. Furthermore, the end user’s energy demand, the local power supplier’s prices, and the external power grid prices all vary over time. To deal with this variability, we formulated the energy scheduling problem as a stochastic optimization. We propose a dynamic algorithm based on Lyapunov optimization, and it is theoretically proved that the proposed algorithm can make the optimization target infinitely close to optimum. Finally, the effectiveness of the proposed algorithm is verified by simulation comparison. The algorithm provides a tool for end user energy scheduling where the user is equipped with energy storage device.","PeriodicalId":438266,"journal":{"name":"2021 IEEE 9th International Conference on Smart Energy Grid Engineering (SEGE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125813924","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 : 2021-08-11DOI: 10.1109/SEGE52446.2021.9534932
Sobhan Dorahaki, R. Dashti, H. Shaker
Nowadays, the smart MicroGrid (MG) is known as a challenging and interesting concept to effectively solve the problems and issues of the power system. In this paper, a novel probabilistic risk base optimization model has been proposed to manage the operation cost and risk cost of the smart MG. The electrical and thermal Demand Response (DR) has been considered in the proposed structure. The Probability Distribution Function (PDF) has been used to model the uncertainty of the model. Also, the K-means and Mixed Integer Linear Programming (MILP) scenario reduction methods have been used to decrease the number of scenarios. Furthermore, the objective function of the proposed optimization is modeled as MILP. The CPLEX solver in the GAMS environment is used to solve the problem. Results show that the electrical and thermal DR causes a decrease in the risk cost of the smart MG.
{"title":"A Novel Probabilistic Risk-Based Energy Management Model in the Smart MicroGrids","authors":"Sobhan Dorahaki, R. Dashti, H. Shaker","doi":"10.1109/SEGE52446.2021.9534932","DOIUrl":"https://doi.org/10.1109/SEGE52446.2021.9534932","url":null,"abstract":"Nowadays, the smart MicroGrid (MG) is known as a challenging and interesting concept to effectively solve the problems and issues of the power system. In this paper, a novel probabilistic risk base optimization model has been proposed to manage the operation cost and risk cost of the smart MG. The electrical and thermal Demand Response (DR) has been considered in the proposed structure. The Probability Distribution Function (PDF) has been used to model the uncertainty of the model. Also, the K-means and Mixed Integer Linear Programming (MILP) scenario reduction methods have been used to decrease the number of scenarios. Furthermore, the objective function of the proposed optimization is modeled as MILP. The CPLEX solver in the GAMS environment is used to solve the problem. Results show that the electrical and thermal DR causes a decrease in the risk cost of the smart MG.","PeriodicalId":438266,"journal":{"name":"2021 IEEE 9th International Conference on Smart Energy Grid Engineering (SEGE)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130554863","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 : 2021-08-11DOI: 10.1109/SEGE52446.2021.9535109
Mostafa Rezaeimozafar, R. Monaghan, E. Barrett, M. Duffy
The increasing deployment of photovoltaic systems and behind-the-meter batteries into power distribution systems has increased interest in optimal system operating conditions. Electricity tariff, as an indirect factor, plays a pivotal role in controlling the customers’ behavior, especially in the presence of batteries. The residential sector, as one of the largest consumers, requires accurate analysis of the impacts of tariffs on its load profile for short-term and long-term planning. In this paper, a household equipped with a photovoltaic array and battery is modeled and the effects of flat-rate, stepped rate, time-of-use, and demand charge pricing structures on the battery charge/discharge model are analyzed. Furthermore, the effects of COVID-influenced consumption patterns and the increase in feed-in tariff for photovoltaic energy on battery scheduling are investigated. The battery scheduling problem is formulated as a non-linear optimization function, to minimize electricity costs for customers, and is solved using a Genetic algorithm.
{"title":"Optimal Scheduling for Behind-the-Meter Batteries under Different Tariff Structures","authors":"Mostafa Rezaeimozafar, R. Monaghan, E. Barrett, M. Duffy","doi":"10.1109/SEGE52446.2021.9535109","DOIUrl":"https://doi.org/10.1109/SEGE52446.2021.9535109","url":null,"abstract":"The increasing deployment of photovoltaic systems and behind-the-meter batteries into power distribution systems has increased interest in optimal system operating conditions. Electricity tariff, as an indirect factor, plays a pivotal role in controlling the customers’ behavior, especially in the presence of batteries. The residential sector, as one of the largest consumers, requires accurate analysis of the impacts of tariffs on its load profile for short-term and long-term planning. In this paper, a household equipped with a photovoltaic array and battery is modeled and the effects of flat-rate, stepped rate, time-of-use, and demand charge pricing structures on the battery charge/discharge model are analyzed. Furthermore, the effects of COVID-influenced consumption patterns and the increase in feed-in tariff for photovoltaic energy on battery scheduling are investigated. The battery scheduling problem is formulated as a non-linear optimization function, to minimize electricity costs for customers, and is solved using a Genetic algorithm.","PeriodicalId":438266,"journal":{"name":"2021 IEEE 9th International Conference on Smart Energy Grid Engineering (SEGE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121725749","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}
In the last decade, Artificial Intelligence (AI) have been applied overwhelmingly in various research domains in the context of smart grid. It has been one of the main streams of advanced technological approaches that the research community offered for developing smart grids. However, the broad scope of the subject matter has launched complexity for scholars to identify effective research approaches. In this paper, we present a literature review about utilizing AI in the key elements of smart grids including grid-connected vehicles, data-driven components, and the power system network. This will result in highlighting technical challenges of the integration of electric vehicles to the grid and the power network operation as well. Moreover, we discuss the four key research areas in the context of AI and its applications in intelligent power grids. The proposed research fields aid PhD candidates to consider these areas as the promising domains for investigation.
{"title":"Applications of Artificial Intelligence in Smart Grids: Present and Future Research Domains","authors":"Farhad Khosrojerdi, Stephane Gagnon, Raul Valverde","doi":"10.1109/SEGE52446.2021.9534914","DOIUrl":"https://doi.org/10.1109/SEGE52446.2021.9534914","url":null,"abstract":"In the last decade, Artificial Intelligence (AI) have been applied overwhelmingly in various research domains in the context of smart grid. It has been one of the main streams of advanced technological approaches that the research community offered for developing smart grids. However, the broad scope of the subject matter has launched complexity for scholars to identify effective research approaches. In this paper, we present a literature review about utilizing AI in the key elements of smart grids including grid-connected vehicles, data-driven components, and the power system network. This will result in highlighting technical challenges of the integration of electric vehicles to the grid and the power network operation as well. Moreover, we discuss the four key research areas in the context of AI and its applications in intelligent power grids. The proposed research fields aid PhD candidates to consider these areas as the promising domains for investigation.","PeriodicalId":438266,"journal":{"name":"2021 IEEE 9th International Conference on Smart Energy Grid Engineering (SEGE)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114292665","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}
These years, the connection of distributed generation (DG) to active distribution systems or microgrid has been widely used. The DGs integration and the novel optimization algorithms, makes the realization of optimal power flow (OPF) at the distribution level feasible. OPF not only reduces system power losses, but also decreases the DGs generation costs; simultaneously, the new control strategy improves the voltage profiles, which is a significant factor of power qualities. In this paper, a multi-objective function is converted into a single objective problem that defines a nonlinear power flow for optimization. Cuckoo Search (CS) algorithm is put into used.
{"title":"Optimization of Unbalanced Active Distribution Systems Using Cuckoo Search Algorithm","authors":"Tianjian Wang, Ying Wang, Yangcheng Hou, Fei Gu, Shaoshuai Hou, Wei Jin","doi":"10.1109/SEGE52446.2021.9534957","DOIUrl":"https://doi.org/10.1109/SEGE52446.2021.9534957","url":null,"abstract":"These years, the connection of distributed generation (DG) to active distribution systems or microgrid has been widely used. The DGs integration and the novel optimization algorithms, makes the realization of optimal power flow (OPF) at the distribution level feasible. OPF not only reduces system power losses, but also decreases the DGs generation costs; simultaneously, the new control strategy improves the voltage profiles, which is a significant factor of power qualities. In this paper, a multi-objective function is converted into a single objective problem that defines a nonlinear power flow for optimization. Cuckoo Search (CS) algorithm is put into used.","PeriodicalId":438266,"journal":{"name":"2021 IEEE 9th International Conference on Smart Energy Grid Engineering (SEGE)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126134651","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 : 2021-08-11DOI: 10.1109/SEGE52446.2021.9534982
R. Mohamed, Bilal Boudy, H. Gabbar
This paper addresses the optimization of the Fractional Order PID controller (FOPID) parameters used to control the frequency and power deviation of hybrid power system based renewable energy generation. This proposed system consists of renewable energy generation like wind and photovoltaic systems with conventional sources such as diesel generator and fuel cell along with Energy Storage Systems (Battery Energy Storage Systems (BESS) and Flywheel Energy Storage Systems (FESS)). The Krill Herd algorithm is used to determine the gains parameters of the Fractional Order PID controller. The scope of this paper is to eliminate the frequency and power deviation to provide stability of the proposed system. The obtained results show that the proposed controller enhances the system stability performance in comparison with the PID controller.
{"title":"Fractional PID Controller Tuning Using Krill Herd for Renewable Power Systems Control","authors":"R. Mohamed, Bilal Boudy, H. Gabbar","doi":"10.1109/SEGE52446.2021.9534982","DOIUrl":"https://doi.org/10.1109/SEGE52446.2021.9534982","url":null,"abstract":"This paper addresses the optimization of the Fractional Order PID controller (FOPID) parameters used to control the frequency and power deviation of hybrid power system based renewable energy generation. This proposed system consists of renewable energy generation like wind and photovoltaic systems with conventional sources such as diesel generator and fuel cell along with Energy Storage Systems (Battery Energy Storage Systems (BESS) and Flywheel Energy Storage Systems (FESS)). The Krill Herd algorithm is used to determine the gains parameters of the Fractional Order PID controller. The scope of this paper is to eliminate the frequency and power deviation to provide stability of the proposed system. The obtained results show that the proposed controller enhances the system stability performance in comparison with the PID controller.","PeriodicalId":438266,"journal":{"name":"2021 IEEE 9th International Conference on Smart Energy Grid Engineering (SEGE)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124180862","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 : 2021-08-11DOI: 10.1109/SEGE52446.2021.9535006
Zeenat Hameed, S. Hashemi, H. Ipsen, C. Træholt
Battery energy storage systems (BESSs) are gaining potential recognition in modern power systems. They enable higher renewable shares in power networks by overcoming issues introduced by the intermittent nature of renewable resources. BESSs also provide various grid services such as frequency regulation, voltage support, energy management, and black start. Choosing an appropriate BESS location plays a key role in maximizing benefits from its services. This paper aims at investigating BESS placement for providing grid services at the point of installation. The previous studies extended in this direction have not considered the requirements of a real project under which BESS is being deployed and have mainly proposed solutions for standard IEEE bus systems. Also, the focus has not been on providing ancillary services using BESS, but mainly on loss minimization. This paper, on the other hand, presents a case study on the BESS placement problem by investigating various potential locations in Bornholm Island for fulfilling the objectives of a BESS-related industrial project, namely BOSS. This is achieved by considering factors like stackability of BESS-services, integration of large-scale renewable resources, and viability of business models.
{"title":"Placement of Battery Energy Storage for Provision of Grid Services – A Bornholm Case Study","authors":"Zeenat Hameed, S. Hashemi, H. Ipsen, C. Træholt","doi":"10.1109/SEGE52446.2021.9535006","DOIUrl":"https://doi.org/10.1109/SEGE52446.2021.9535006","url":null,"abstract":"Battery energy storage systems (BESSs) are gaining potential recognition in modern power systems. They enable higher renewable shares in power networks by overcoming issues introduced by the intermittent nature of renewable resources. BESSs also provide various grid services such as frequency regulation, voltage support, energy management, and black start. Choosing an appropriate BESS location plays a key role in maximizing benefits from its services. This paper aims at investigating BESS placement for providing grid services at the point of installation. The previous studies extended in this direction have not considered the requirements of a real project under which BESS is being deployed and have mainly proposed solutions for standard IEEE bus systems. Also, the focus has not been on providing ancillary services using BESS, but mainly on loss minimization. This paper, on the other hand, presents a case study on the BESS placement problem by investigating various potential locations in Bornholm Island for fulfilling the objectives of a BESS-related industrial project, namely BOSS. This is achieved by considering factors like stackability of BESS-services, integration of large-scale renewable resources, and viability of business models.","PeriodicalId":438266,"journal":{"name":"2021 IEEE 9th International Conference on Smart Energy Grid Engineering (SEGE)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124999081","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 : 2021-08-11DOI: 10.1109/SEGE52446.2021.9535108
Sirkka Porada, Leonard Schulte, A. Moser
The integration of wind turbines into the European power system poses new challenges for grid operations. One reason for this is the volatile feed-in behavior of wind turbines. Due to various meteorological influencing factors, feed-in profiles of wind turbines show not solely fluctuations in a hourly range, but also significant gradients in the timeframe of seconds to a few minutes. These short-term fluctuations of the power feed-in can cause local problems in the power system. Most studies address the generation of synthetic feed-in profiles with of temporal resolution of 15 till 60 minutes. To assess the impact of fluctuations in shorter timeframe, this paper focus on this paper focus on the generation of feed-in profiles with a resolution of 10 seconds. For this purpose, a stochastic method is developed generating feed-in profiles for wind turbines based on a Markov Chain Monte Carlo simulation. The generated feed-in profiles suitably represent the influence of meteorological phenomena in the seconds as well as in the hourly range.
{"title":"A Stochastic Approach to Generate Short-Term Feed-in Profiles of Wind Power Plants","authors":"Sirkka Porada, Leonard Schulte, A. Moser","doi":"10.1109/SEGE52446.2021.9535108","DOIUrl":"https://doi.org/10.1109/SEGE52446.2021.9535108","url":null,"abstract":"The integration of wind turbines into the European power system poses new challenges for grid operations. One reason for this is the volatile feed-in behavior of wind turbines. Due to various meteorological influencing factors, feed-in profiles of wind turbines show not solely fluctuations in a hourly range, but also significant gradients in the timeframe of seconds to a few minutes. These short-term fluctuations of the power feed-in can cause local problems in the power system. Most studies address the generation of synthetic feed-in profiles with of temporal resolution of 15 till 60 minutes. To assess the impact of fluctuations in shorter timeframe, this paper focus on this paper focus on the generation of feed-in profiles with a resolution of 10 seconds. For this purpose, a stochastic method is developed generating feed-in profiles for wind turbines based on a Markov Chain Monte Carlo simulation. The generated feed-in profiles suitably represent the influence of meteorological phenomena in the seconds as well as in the hourly range.","PeriodicalId":438266,"journal":{"name":"2021 IEEE 9th International Conference on Smart Energy Grid Engineering (SEGE)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127168121","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：