Designing future energy systems with high penetrations of variable renewable energy and third-party owned devices requires information with high spatial and temporal granularity. Existing public datasets focus on specific resource classes (ex. bulk generators, residential solar, or electric vehicles), and cannot inform holistic planning or policy decisions. Further, with the high penetration of distributed energy resources (DERs) located in the distribution grid, datasets and models which focus only on the bulk system will no longer be sufficient. To meet this modelling need, this paper presents a project-driven dataset of DERs for the contiguous U.S., generated using only publicly available data. We integrate the resources into a high-resolution test system of the U.S. grid. Our integrated U.S. grid model and DER dataset enables planners, operators, and policy makers to pose questions and conduct data-driven analysis of rapid decarbonization pathways for the electricity system. We pose a set of research questions in our Research Project Database.
{"title":"Towards closing the data gap: A project-driven distributed energy resource dataset for the U.S. Grid","authors":"R. Haider, Yixing Xu, Weiwei Yang","doi":"10.1145/3599733.3600250","DOIUrl":"https://doi.org/10.1145/3599733.3600250","url":null,"abstract":"Designing future energy systems with high penetrations of variable renewable energy and third-party owned devices requires information with high spatial and temporal granularity. Existing public datasets focus on specific resource classes (ex. bulk generators, residential solar, or electric vehicles), and cannot inform holistic planning or policy decisions. Further, with the high penetration of distributed energy resources (DERs) located in the distribution grid, datasets and models which focus only on the bulk system will no longer be sufficient. To meet this modelling need, this paper presents a project-driven dataset of DERs for the contiguous U.S., generated using only publicly available data. We integrate the resources into a high-resolution test system of the U.S. grid. Our integrated U.S. grid model and DER dataset enables planners, operators, and policy makers to pose questions and conduct data-driven analysis of rapid decarbonization pathways for the electricity system. We pose a set of research questions in our Research Project Database.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116474950","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}
Himanshu Sharma, Joshua D. Suetterlein, Sumathi Lakshmiranganatha, T. Flynn, D. Vrabie, Christine M. Sweeney, V. Ramakrishniah
With recent advances in deep learning and large-scale computing, learning-based controls have become increasingly attractive for complex physical systems. Consequently, developing generalized learning-based control software that takes into account the next generation of computing architectures is paramount. Specifically, for the case of complex control, we present the Easily eXtendable Architecture for Reinforcement Learning (EXARL), which aims to support various scientific applications seeking to leverage reinforcement learning (RL) on exascale computing architectures. We demonstrate the efficacy and performance of the EXARL library for the scientific use case of designing a complex control policy to stabilize a power system after experiencing a fault. We use a parallel augmented random search method developed within EXARL and present its preliminary validation and performance stabilization of a fault for the IEEE 39-bus system.
{"title":"EXARL-PARS: Parallel Augmented Random Search Using Reinforcement Learning at Scale for Applications in Power Systems","authors":"Himanshu Sharma, Joshua D. Suetterlein, Sumathi Lakshmiranganatha, T. Flynn, D. Vrabie, Christine M. Sweeney, V. Ramakrishniah","doi":"10.1145/3599733.3600261","DOIUrl":"https://doi.org/10.1145/3599733.3600261","url":null,"abstract":"With recent advances in deep learning and large-scale computing, learning-based controls have become increasingly attractive for complex physical systems. Consequently, developing generalized learning-based control software that takes into account the next generation of computing architectures is paramount. Specifically, for the case of complex control, we present the Easily eXtendable Architecture for Reinforcement Learning (EXARL), which aims to support various scientific applications seeking to leverage reinforcement learning (RL) on exascale computing architectures. We demonstrate the efficacy and performance of the EXARL library for the scientific use case of designing a complex control policy to stabilize a power system after experiencing a fault. We use a parallel augmented random search method developed within EXARL and present its preliminary validation and performance stabilization of a fault for the IEEE 39-bus system.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120867514","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}
S. A. R. Naqvi, V. Chandan, S. Bhattacharya, Na Luo, K. Kar, C. Sivaraman, Nikitha Radhakrishnan
In this paper, we use publicly available data of a highly instrumented office building to estimate how zonal temperature and carbon dioxide (CO2) concentration are related to some key operational and environmental measurements. Subsequently, we have developed, simulated, and evaluated an optimization framework for minimizing the energy consumption of the central heating, ventilation and air conditioning (HVAC) unit while meeting zonal temperature and indoor air quality (IAQ) standards. Finally, we have evaluated the achievable energy savings for our proposed approach as compared to a baseline approach and reported significant savings potential.
{"title":"Data-Driven Co-optimization of Energy Efficiency and Indoor Environmental Quality in Commercial Buildings","authors":"S. A. R. Naqvi, V. Chandan, S. Bhattacharya, Na Luo, K. Kar, C. Sivaraman, Nikitha Radhakrishnan","doi":"10.1145/3599733.3600262","DOIUrl":"https://doi.org/10.1145/3599733.3600262","url":null,"abstract":"In this paper, we use publicly available data of a highly instrumented office building to estimate how zonal temperature and carbon dioxide (CO2) concentration are related to some key operational and environmental measurements. Subsequently, we have developed, simulated, and evaluated an optimization framework for minimizing the energy consumption of the central heating, ventilation and air conditioning (HVAC) unit while meeting zonal temperature and indoor air quality (IAQ) standards. Finally, we have evaluated the achievable energy savings for our proposed approach as compared to a baseline approach and reported significant savings potential.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130022587","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}
With the global push to decarbonize the building sector and growing interest in occupant-centric building controls, numerous simulation and field studies have been conducted to explore the trade-off between energy efficiency and occupant comfort. These studies largely disregard individual differences in thermal comfort and assume each zone has a fixed occupancy schedule. In office buildings, there is often some leeway in how occupants are grouped and assigned to different building spaces (e.g., offices and meeting rooms). In this paper we investigate the extent of the impact of the space allocation strategy on the energy-comfort trade-off in office buildings, and whether it depends on specific building characteristics. Our simulation shows that varying the space allocation strategy in a medium office building can lead to over 3.5%/15.1% change in annual/monthly energy consumption, and over 15% change in average thermal comfort when using the personal comfort model. This finding calls for the joint optimization of HVAC operation and space allocation, possibly at different timescales.
{"title":"Investigating the Impact of Space Allocation Strategy on Energy-Comfort Trade-off in Office Buildings","authors":"Tianyu Zhang, Omid Ardakanian","doi":"10.1145/3599733.3600263","DOIUrl":"https://doi.org/10.1145/3599733.3600263","url":null,"abstract":"With the global push to decarbonize the building sector and growing interest in occupant-centric building controls, numerous simulation and field studies have been conducted to explore the trade-off between energy efficiency and occupant comfort. These studies largely disregard individual differences in thermal comfort and assume each zone has a fixed occupancy schedule. In office buildings, there is often some leeway in how occupants are grouped and assigned to different building spaces (e.g., offices and meeting rooms). In this paper we investigate the extent of the impact of the space allocation strategy on the energy-comfort trade-off in office buildings, and whether it depends on specific building characteristics. Our simulation shows that varying the space allocation strategy in a medium office building can lead to over 3.5%/15.1% change in annual/monthly energy consumption, and over 15% change in average thermal comfort when using the personal comfort model. This finding calls for the joint optimization of HVAC operation and space allocation, possibly at different timescales.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115729312","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}
Dorina Werling, Maximilian Beichter, Benedikt Heidrich, Kaleb Phipps, R. Mikut, V. Hagenmeyer
Transforming the energy system to decentralised, renewable energy sources requires measures to balance their fluctuating nature and stabilise the energy system. One such measure is a dispatchable feeder, which combines inflexible prosumption with a flexible energy storage system. The energy storage system’s management is formulated as a stochastic optimisation problem that requires energy time series forecasts as input. These forecasts can significantly influence the performance of the dispatchable feeder: the forecasts have a so-called forecast value for the dispatchable feeder, which is not directly reflected by error-based forecast quality metrics. Therefore, we analyse how the considered forecast value for the dispatchable feeder is related to the considered forecast quality and influenced by forecasts with different characteristics. Furthermore, we examine the impact of problem-specific parameters such as the data and the battery capacity. To this means, we create forecasts with different characteristics using neural networks with varying loss functions and perform the analysis using a data set with 300 buildings. The results of our analysis show that the relation between the considered forecast quality and forecast value for the dispatchable feeder is non-monotonic. Furthermore, we show that the forecast characteristics influence the forecast value differently depending on the data and the battery capacity.
{"title":"The Impact of Forecast Characteristics on the Forecast Value for the Dispatchable Feeder","authors":"Dorina Werling, Maximilian Beichter, Benedikt Heidrich, Kaleb Phipps, R. Mikut, V. Hagenmeyer","doi":"10.1145/3599733.3600251","DOIUrl":"https://doi.org/10.1145/3599733.3600251","url":null,"abstract":"Transforming the energy system to decentralised, renewable energy sources requires measures to balance their fluctuating nature and stabilise the energy system. One such measure is a dispatchable feeder, which combines inflexible prosumption with a flexible energy storage system. The energy storage system’s management is formulated as a stochastic optimisation problem that requires energy time series forecasts as input. These forecasts can significantly influence the performance of the dispatchable feeder: the forecasts have a so-called forecast value for the dispatchable feeder, which is not directly reflected by error-based forecast quality metrics. Therefore, we analyse how the considered forecast value for the dispatchable feeder is related to the considered forecast quality and influenced by forecasts with different characteristics. Furthermore, we examine the impact of problem-specific parameters such as the data and the battery capacity. To this means, we create forecasts with different characteristics using neural networks with varying loss functions and perform the analysis using a data set with 300 buildings. The results of our analysis show that the relation between the considered forecast quality and forecast value for the dispatchable feeder is non-monotonic. Furthermore, we show that the forecast characteristics influence the forecast value differently depending on the data and the battery capacity.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114737961","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}
Non-Intrusive Load Monitoring (NILM), which provides sufficient load information from the energy consumption of the entire building, has become crucial in improving the operation of energy systems. Although it can decompose overall energy consumption into individual electrical sub-loads, it struggles to identify such thermal-driven sub-loads as occupants. This paper explores and proposes a Non-Intrusive Thermal Load Monitoring (NITLM) with recursive models and input data selection to accurately disaggregate the overall thermal load into sub-loads, focusing on occupant thermal load. In experiments, we generated a thermal load dataset derived from a whole building energy simulation and compared the accuracy of the monitoring results with the generated reference data. Our experimental results show that our designed model reduces MAE by up to 77.0% more than the existing NITLM approach.
{"title":"Exploring of Recursive Model-based Non-Intrusive Thermal Load Monitoring for Building Cooling Load","authors":"Kazuki Okazawa, Naoya Kaneko, Dafang Zhao, Hiroki Nishikawa, Ittetsu Taniguchi, Takao Onoye","doi":"10.1145/3599733.3600259","DOIUrl":"https://doi.org/10.1145/3599733.3600259","url":null,"abstract":"Non-Intrusive Load Monitoring (NILM), which provides sufficient load information from the energy consumption of the entire building, has become crucial in improving the operation of energy systems. Although it can decompose overall energy consumption into individual electrical sub-loads, it struggles to identify such thermal-driven sub-loads as occupants. This paper explores and proposes a Non-Intrusive Thermal Load Monitoring (NITLM) with recursive models and input data selection to accurately disaggregate the overall thermal load into sub-loads, focusing on occupant thermal load. In experiments, we generated a thermal load dataset derived from a whole building energy simulation and compared the accuracy of the monitoring results with the generated reference data. Our experimental results show that our designed model reduces MAE by up to 77.0% more than the existing NITLM approach.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132421679","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}
Alessandro Salatiello, Ye Wang, G. Wichern, T. Koike-Akino, Yoshihiro Ohta, Yosuke Kaneko, C. Laughman, A. Chakrabarty
The generation of time-series profiles of building operation requires expensive and time-consuming data consolidation and modeling efforts that rely on extensive domain knowledge and need frequent revisions due to evolving energy systems, user behavior, and environmental conditions. Generative deep learning may be used to provide an automatic, scalable, data-source-agnostic, and efficient method to synthesize these artificial time-series profiles by learning the distribution of the original data. While a range of generative neural networks have been proposed, generative adversarial networks (GANs) and variational autoencoders (VAEs) are most popular models; GANs typically require considerable customization to stabilize the training procedure, while VAEs are often reported to generate lower-quality samples compared to GANs. In this paper, we propose a network architecture and training procedure that combines the strengths of VAEs and GANs by incorporating Regularized Adversarial Fine-Tuning (RAFT). We imbue the architecture with conditional inputs to reflect ambient/outdoor conditions and operating conditions, and demonstrate its effectiveness by using operational data collected over 585 days from SUSTIE: Mitsubishi Electric’s net-zero energy building. Comparing against classical GAN, VAE, Wasserstein-GAN, and VAE-GAN, our proposed conditional RAFT-VAE-GAN outperforms its competitors in terms of mean accuracy, training stability, and several metrics that ascertain how close the synthetic distribution is to the measured data distribution.
生成建筑操作的时间序列概要需要昂贵且耗时的数据整合和建模工作，这些工作依赖于广泛的领域知识，并且由于不断发展的能源系统、用户行为和环境条件，需要经常进行修订。生成式深度学习可以提供一种自动的、可扩展的、与数据源无关的、有效的方法，通过学习原始数据的分布来合成这些人工时间序列轮廓。虽然已经提出了一系列的生成神经网络，但生成对抗网络(gan)和变分自编码器(VAEs)是最流行的模型;GANs通常需要大量的定制来稳定训练过程，而与GANs相比，VAEs经常报告生成质量较低的样本。在本文中，我们提出了一种网络架构和训练过程，通过结合正则化对抗性微调(RAFT)，结合了vae和gan的优势。我们为建筑注入了条件输入，以反映环境/室外条件和运行条件，并通过使用从三菱电机的净零能耗建筑SUSTIE收集的585天的运行数据来证明其有效性。与经典GAN、VAE、Wasserstein-GAN和vee -GAN相比，我们提出的条件raft - vee -GAN在平均准确率、训练稳定性和确定合成分布与测量数据分布的接近程度的几个指标方面优于其竞争对手。
{"title":"Synthesizing Building Operation Data with Generative Models: VAEs, GANs, or Something In Between?","authors":"Alessandro Salatiello, Ye Wang, G. Wichern, T. Koike-Akino, Yoshihiro Ohta, Yosuke Kaneko, C. Laughman, A. Chakrabarty","doi":"10.1145/3599733.3600260","DOIUrl":"https://doi.org/10.1145/3599733.3600260","url":null,"abstract":"The generation of time-series profiles of building operation requires expensive and time-consuming data consolidation and modeling efforts that rely on extensive domain knowledge and need frequent revisions due to evolving energy systems, user behavior, and environmental conditions. Generative deep learning may be used to provide an automatic, scalable, data-source-agnostic, and efficient method to synthesize these artificial time-series profiles by learning the distribution of the original data. While a range of generative neural networks have been proposed, generative adversarial networks (GANs) and variational autoencoders (VAEs) are most popular models; GANs typically require considerable customization to stabilize the training procedure, while VAEs are often reported to generate lower-quality samples compared to GANs. In this paper, we propose a network architecture and training procedure that combines the strengths of VAEs and GANs by incorporating Regularized Adversarial Fine-Tuning (RAFT). We imbue the architecture with conditional inputs to reflect ambient/outdoor conditions and operating conditions, and demonstrate its effectiveness by using operational data collected over 585 days from SUSTIE: Mitsubishi Electric’s net-zero energy building. Comparing against classical GAN, VAE, Wasserstein-GAN, and VAE-GAN, our proposed conditional RAFT-VAE-GAN outperforms its competitors in terms of mean accuracy, training stability, and several metrics that ascertain how close the synthetic distribution is to the measured data distribution.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"123 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132518336","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}
Anupama Sitaraman, Noman Bashir, David E. Irwin, P. Shenoy
Recent studies analyze the carbon footprint of residential heating and propose transitioning to electric heat pumps as an important step towards decarbonization. Electric heat pumps are more energy-efficient than gas furnaces and use electric grid power. However, electric grids in most parts of the world are primarily powered by carbon-intensive fossil fuels and may never be completely carbon-free, and widespread usage of heat pumps may trigger expensive upgrades in the electric grid. A low-cost, deep decarbonization of residential heating can be achieved by using co-located solar photovoltaic (PV) systems alongside heat pump retrofits. In this poster, we investigate the problem of sizing solar panels and storage to completely offset the added demand and investigate the tradeoff between cost and carbon emission reduction benefits. Our analysis suggests that co-located solar PV systems can reduce carbon emissions by at least 57.7%.
{"title":"Leveraging Solar PV and Storage for Deep Decarbonization of Residential Heating Systems","authors":"Anupama Sitaraman, Noman Bashir, David E. Irwin, P. Shenoy","doi":"10.1145/3599733.3606302","DOIUrl":"https://doi.org/10.1145/3599733.3606302","url":null,"abstract":"Recent studies analyze the carbon footprint of residential heating and propose transitioning to electric heat pumps as an important step towards decarbonization. Electric heat pumps are more energy-efficient than gas furnaces and use electric grid power. However, electric grids in most parts of the world are primarily powered by carbon-intensive fossil fuels and may never be completely carbon-free, and widespread usage of heat pumps may trigger expensive upgrades in the electric grid. A low-cost, deep decarbonization of residential heating can be achieved by using co-located solar photovoltaic (PV) systems alongside heat pump retrofits. In this poster, we investigate the problem of sizing solar panels and storage to completely offset the added demand and investigate the tradeoff between cost and carbon emission reduction benefits. Our analysis suggests that co-located solar PV systems can reduce carbon emissions by at least 57.7%.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133043296","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}
VPPs (Virtual Power Plants) play an important role in balancing supply and demand. In order to make VPP revenue, it is necessary to forecast market prices and bidding energy for supply and demand adjustment markets, called FCAS (Frequency Control Ancillary Service) markets. However, price forecasting for FCAS markets is still challenging because they have multiple different response times and one price, directly and indirectly, influences each other. There is no study on electricity price forecasting in FCAS markets, and a novel forecasting model considering not only its price but also the other prices of the different response times is necessary. This work presents a market price forecasting model for a FCAS market by exploring the forecasting models derived from a wholesale market, and then it takes into account the markets with different response times as well as the target one from AEMO (Australian Energy Market Operator). Through the experiments, our forecasting model achieves 7.8$/MWh of RMSE on the electricity price in AEMO’s 6-Second-Raise market. The proposed forecasting model reduces RMSE by 80% compared to the forecast price published by AEMO.
虚拟电厂在平衡电力供需方面发挥着重要作用。为了获得VPP收益，有必要对供需调节市场(FCAS (Frequency Control auxiliary Service，频率控制辅助服务)市场进行市场价格预测和能源投标。然而，FCAS市场的价格预测仍然具有挑战性，因为它们有多个不同的响应时间，并且一个价格直接或间接地相互影响。目前还没有对FCAS市场的电价预测进行研究，有必要建立一种既考虑本电价又考虑不同响应时间下其他电价的预测模型。本文通过对批发市场的预测模型进行探索，提出了一个FCAS市场的市场价格预测模型，并考虑了不同响应时间的市场以及AEMO(澳大利亚能源市场运营商)的目标市场。通过实验，我们的预测模型对AEMO 6秒提价市场的电价RMSE达到了7.8美元/兆瓦时。与AEMO公布的预测价格相比，该预测模型的均方根误差降低了80%。
{"title":"Exploring Models of Electricity Price Forecasting: Case Study on A FCAS Market","authors":"Kenshiro Kato, Koki Iwabuchi, Daichi Watari, Dafang Zhao, Hiroki Nishikawa, Ittetsu Taniguchi, Takao Onoye","doi":"10.1145/3599733.3600258","DOIUrl":"https://doi.org/10.1145/3599733.3600258","url":null,"abstract":"VPPs (Virtual Power Plants) play an important role in balancing supply and demand. In order to make VPP revenue, it is necessary to forecast market prices and bidding energy for supply and demand adjustment markets, called FCAS (Frequency Control Ancillary Service) markets. However, price forecasting for FCAS markets is still challenging because they have multiple different response times and one price, directly and indirectly, influences each other. There is no study on electricity price forecasting in FCAS markets, and a novel forecasting model considering not only its price but also the other prices of the different response times is necessary. This work presents a market price forecasting model for a FCAS market by exploring the forecasting models derived from a wholesale market, and then it takes into account the markets with different response times as well as the target one from AEMO (Australian Energy Market Operator). Through the experiments, our forecasting model achieves 7.8$/MWh of RMSE on the electricity price in AEMO’s 6-Second-Raise market. The proposed forecasting model reduces RMSE by 80% compared to the forecast price published by AEMO.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121243175","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}
Rickard Brännvall, Tina Stark, Jonas Gustafsson, Mats Eriksson, J. Summers
This article investigates the problem of where to place the computation workload in an edge-cloud network topology considering the trade-off between the location-specific cost of computation and data communication. For this purpose, a Monte Carlo simulation model is defined that accounts for different workload types, their distribution across time and location, as well as correlation structure. Results confirm and quantify the intuition that optimization can be achieved by distributing a part of cloud computation to make efficient use of resources in an edge data center network, with operational energy savings of 4–6% and up to 50% reduction in its claim for cloud capacity.
{"title":"Cost Optimization for the Edge-Cloud Continuum by Energy-Aware Workload Placement","authors":"Rickard Brännvall, Tina Stark, Jonas Gustafsson, Mats Eriksson, J. Summers","doi":"10.1145/3599733.3600253","DOIUrl":"https://doi.org/10.1145/3599733.3600253","url":null,"abstract":"This article investigates the problem of where to place the computation workload in an edge-cloud network topology considering the trade-off between the location-specific cost of computation and data communication. For this purpose, a Monte Carlo simulation model is defined that accounts for different workload types, their distribution across time and location, as well as correlation structure. Results confirm and quantify the intuition that optimization can be achieved by distributing a part of cloud computation to make efficient use of resources in an edge data center network, with operational energy savings of 4–6% and up to 50% reduction in its claim for cloud capacity.","PeriodicalId":114998,"journal":{"name":"Companion Proceedings of the 14th ACM International Conference on Future Energy Systems","volume":"383 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115479552","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：