Pub Date : 2024-09-27DOI: 10.1016/j.apenergy.2024.124246
Aiming at the problem of load fluctuation at the power end of large ports, we propose a hybrid neural network joint model based on Mode Decomposition (MD) and Change Point Detection (CPD) to accomplish the load forecasting. In this study, a two-stage joint prediction model is constructed. First, the number of Intrinsic Mode Functions (IMFs) in the Variational Mode Decomposition (VMD) process was dynamically adjusted by introducing an improved Signal Energy (SE) evaluation metric. Subsequently, a Bidirectional Gated Recurrent Unit (Bi-GRU) network is employed to predict these IMFs, and the potential effect of the breakpoints on the prediction outcomes is investigated using the Iterative Cumulative Sum of Squares (ICSS) method. Finally, the eigenmode functions are summed and reconstructed, and then combined with the breakpoint data as inputs for the second stage prediction. To ensure the efficiency of the second stage prediction, the Mogrifier Long-and Short-Term Memory (Mogrifier-LSTM) network structure is improved. In the two-stage model, the adaptive tuning of hyperparameters is implemented by a Hunter-Prey Optimization (HPO) algorithm based on a redesigned chaotic mapping strategy. During the simulation, various neural network topologies were employed to confirm the effectiveness of the model in port power load forecasting.
{"title":"A power load forecasting method in port based on VMD-ICSS-hybrid neural network","authors":"","doi":"10.1016/j.apenergy.2024.124246","DOIUrl":"10.1016/j.apenergy.2024.124246","url":null,"abstract":"<div><div>Aiming at the problem of load fluctuation at the power end of large ports, we propose a hybrid neural network joint model based on Mode Decomposition (MD) and Change Point Detection (CPD) to accomplish the load forecasting. In this study, a two-stage joint prediction model is constructed. First, the number of Intrinsic Mode Functions (IMFs) in the Variational Mode Decomposition (VMD) process was dynamically adjusted by introducing an improved Signal Energy (SE) evaluation metric. Subsequently, a Bidirectional Gated Recurrent Unit (Bi-GRU) network is employed to predict these IMFs, and the potential effect of the breakpoints on the prediction outcomes is investigated using the Iterative Cumulative Sum of Squares (ICSS) method. Finally, the eigenmode functions are summed and reconstructed, and then combined with the breakpoint data as inputs for the second stage prediction. To ensure the efficiency of the second stage prediction, the Mogrifier Long-and Short-Term Memory (Mogrifier-LSTM) network structure is improved. In the two-stage model, the adaptive tuning of hyperparameters is implemented by a Hunter-Prey Optimization (HPO) algorithm based on a redesigned chaotic mapping strategy. During the simulation, various neural network topologies were employed to confirm the effectiveness of the model in port power load forecasting.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142328203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-27DOI: 10.1016/j.apenergy.2024.124593
Life Cycle Assessment (LCA) is widely used to externally compare environmental indicators across different systems. Although uncertainty analysis is required by standards, it is often neglected, which threatens the reliability of the comparisons. The authors highlights how different assumptions and uncertainty sources can shape LCA outcomes. A case study on public bus fleet electrification was conducted, involving 20 bus models with various modeling assumptions. The impact of following factors on LCA uncertainty was analyzed: LCI database, LCIA method, modeling approach, energy carrier consumption and lifetime. The most significant discrepancies, comparing with baseline models of diesel and electric bus, occurred when different LCIA methods were applied, with results varying by up to 649.0%. The use of alternate LCI caused changes of up to 99.4%. The maximum discrepancies due to modeling approach, energy carrier consumption, and lifetime were 33.0%, 35.7%, and 20.9%, respectively. The paper recommends that comprehensive LCA studies should include multiple indicators, and clearly explained uncertainty sources, assumptions and limitations. Modeling approaches, databases, and LCIA methods should align with the analysis goals. Standardization of LCA methodologies by EPD program operators are suggested to reduce variability. When comparing studies with different assumptions, recalculating results to harmonize assumptions is advised. Transparency and understanding of model uncertainties are essential for drawing reliable conclusions. The study demonstrated that comparing deterministic LCA results undermines reliability. As LCA gains importance in environmental and sustainability communications, increasing awareness of LCA uncertainty and applying the novel findings of this paper is essential for informed decision-making.
{"title":"The importance of uncertainty sources in LCA for the reliability of environmental comparisons: A case study on public bus fleet electrification","authors":"","doi":"10.1016/j.apenergy.2024.124593","DOIUrl":"10.1016/j.apenergy.2024.124593","url":null,"abstract":"<div><div>Life Cycle Assessment (LCA) is widely used to externally compare environmental indicators across different systems. Although uncertainty analysis is required by standards, it is often neglected, which threatens the reliability of the comparisons. The authors highlights how different assumptions and uncertainty sources can shape LCA outcomes. A case study on public bus fleet electrification was conducted, involving 20 bus models with various modeling assumptions. The impact of following factors on LCA uncertainty was analyzed: LCI database, LCIA method, modeling approach, energy carrier consumption and lifetime. The most significant discrepancies, comparing with baseline models of diesel and electric bus, occurred when different LCIA methods were applied, with results varying by up to 649.0%. The use of alternate LCI caused changes of up to 99.4%. The maximum discrepancies due to modeling approach, energy carrier consumption, and lifetime were 33.0%, 35.7%, and 20.9%, respectively. The paper recommends that comprehensive LCA studies should include multiple indicators, and clearly explained uncertainty sources, assumptions and limitations. Modeling approaches, databases, and LCIA methods should align with the analysis goals. Standardization of LCA methodologies by EPD program operators are suggested to reduce variability. When comparing studies with different assumptions, recalculating results to harmonize assumptions is advised. Transparency and understanding of model uncertainties are essential for drawing reliable conclusions. The study demonstrated that comparing deterministic LCA results undermines reliability. As LCA gains importance in environmental and sustainability communications, increasing awareness of LCA uncertainty and applying the novel findings of this paper is essential for informed decision-making.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142328204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-27DOI: 10.1016/j.apenergy.2024.124570
The thermal safety of battery systems is a common and key technical problem restricting industrial development. Welding is one of the most important electrical connection methods for lithium-ion battery groups, and the quality of welding directly determines the thermal safety of battery modules. In this research, the inconsistencies and thermal safety of cylindrical lithium-ion battery modules are studied based on cold welding technology. Secondly, the electrochemical characteristics and thermal runaway characteristics of the battery were experimentally studied. Finally, the battery (Table-Based) module launched by the SIMULINK tool of MATLAB software in 2018 was used to build a battery pack model simulating the discharge process to simulate and analyze the battery electrical characteristics. The relevant data show that the temperature difference between the batteries is less than 4 °C and the maximum battery temperature is less than 60 °C when the cold welded module is discharged at a current ratio(C) of 3 high rate, which has good temperature equalization and thermal safety. The output power is higher and the discharge energy increase by 3 % ~ 5 % when the cold-welded module is discharged at different rates. The results for heat abuse conditions show that the overall temperature rise of the cold-welded module is lower, the maximum temperature of the single battery is reduced by 10.7 %, and the maximum temperature rise rate is reduced by 41.2 %. The simulation results show that the current difference between the cells in the hot welding module is large, and there is an obvious overdischarge phenomenon in the late discharge period. The maximum SOC difference between the single battery of the cold-welded module is less than 0.02 when discharging at 3C. The requirements for SOC estimation are met. The above research results confirm that the relevant research will provide new ideas and theoretical value for the research of the consistency improvement of power battery packs, and solve the problem of the electrical/thermal balance difference of the existing resistance thermal welding process from another dimension based on the cold welding strategy.
{"title":"Effect of cold welding on the inconsistencies and thermal safety of battery modules based on a constructed discharge model","authors":"","doi":"10.1016/j.apenergy.2024.124570","DOIUrl":"10.1016/j.apenergy.2024.124570","url":null,"abstract":"<div><div>The thermal safety of battery systems is a common and key technical problem restricting industrial development. Welding is one of the most important electrical connection methods for lithium-ion battery groups, and the quality of welding directly determines the thermal safety of battery modules. In this research, the inconsistencies and thermal safety of cylindrical lithium-ion battery modules are studied based on cold welding technology. Secondly, the electrochemical characteristics and thermal runaway characteristics of the battery were experimentally studied. Finally, the battery (Table-Based) module launched by the SIMULINK tool of MATLAB software in 2018 was used to build a battery pack model simulating the discharge process to simulate and analyze the battery electrical characteristics. The relevant data show that the temperature difference between the batteries is less than 4 °C and the maximum battery temperature is less than 60 °C when the cold welded module is discharged at a current ratio(C) of 3 high rate, which has good temperature equalization and thermal safety. The output power is higher and the discharge energy increase by 3 % ~ 5 % when the cold-welded module is discharged at different rates. The results for heat abuse conditions show that the overall temperature rise of the cold-welded module is lower, the maximum temperature of the single battery is reduced by 10.7 %, and the maximum temperature rise rate is reduced by 41.2 %. The simulation results show that the current difference between the cells in the hot welding module is large, and there is an obvious overdischarge phenomenon in the late discharge period. The maximum SOC difference between the single battery of the cold-welded module is less than 0.02 when discharging at 3C. The requirements for SOC estimation are met. The above research results confirm that the relevant research will provide new ideas and theoretical value for the research of the consistency improvement of power battery packs, and solve the problem of the electrical/thermal balance difference of the existing resistance thermal welding process from another dimension based on the cold welding strategy.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142326714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-26DOI: 10.1016/j.apenergy.2024.124513
Energy management strategies (EMSs) for fuel cell vehicles aim at high fuel efficiency but must also consider the lifetimes of the fuel cell system (FCS) and the battery. Regarding both objectives, fuel cell stack shutdowns play a decisive role in real-world driving situations with low or negative power demand. However, each stack start/stop event is associated with degradation, which is why it is important to keep the number of starts/stops low. This work proposes a predictive EMS with optimal stack start/stop control that takes advantage of a route-based prediction of the entire driving mission to minimize both the fuel consumption and the number of start/stop events. Before departure, the prediction of the entire driving mission is processed in a single offline optimization with dynamic programming. This optimization yields maps providing the real-time EMS with optimal control information that continuously adapts depending on the position along the driving mission and the battery state of charge. Considering this predictive information, the real-time EMS optimizes start/stop actions and the stack power such that the cost-to-go, i.e., the fuel consumption for the trip remainder including start/stop penalties, is implicitly minimized in each instant. In this way, the EMS continuously adapts to the actual conditions, making it robust against unpredicted disturbances, e.g., due to traffic. The superior performance of the proposed strategy compared to state-of-the-art start/stop methods is demonstrated in numerical studies based on real-world driving missions for different vehicle classes with single and multi-stack FCSs.
{"title":"Predictive energy management strategy with optimal stack start/stop control for fuel cell vehicles","authors":"","doi":"10.1016/j.apenergy.2024.124513","DOIUrl":"10.1016/j.apenergy.2024.124513","url":null,"abstract":"<div><div>Energy management strategies (EMSs) for fuel cell vehicles aim at high fuel efficiency but must also consider the lifetimes of the fuel cell system (FCS) and the battery. Regarding both objectives, fuel cell stack shutdowns play a decisive role in real-world driving situations with low or negative power demand. However, each stack start/stop event is associated with degradation, which is why it is important to keep the number of starts/stops low. This work proposes a predictive EMS with optimal stack start/stop control that takes advantage of a route-based prediction of the entire driving mission to minimize both the fuel consumption and the number of start/stop events. Before departure, the prediction of the entire driving mission is processed in a single offline optimization with dynamic programming. This optimization yields maps providing the real-time EMS with optimal control information that continuously adapts depending on the position along the driving mission and the battery state of charge. Considering this predictive information, the real-time EMS optimizes start/stop actions and the stack power such that the cost-to-go, i.e., the fuel consumption for the trip remainder including start/stop penalties, is implicitly minimized in each instant. In this way, the EMS continuously adapts to the actual conditions, making it robust against unpredicted disturbances, e.g., due to traffic. The superior performance of the proposed strategy compared to state-of-the-art start/stop methods is demonstrated in numerical studies based on real-world driving missions for different vehicle classes with single and multi-stack FCSs.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-26DOI: 10.1016/j.apenergy.2024.124322
Large-scale energy storage is highlighted as key for decarbonisation, yet there lacks consensus on the optimal types of storage required. Seasonal Thermal Energy Storage (STES) is an established feature of effective energy transitions in some countries, such as Denmark and the Netherlands, but it remains a marginal technology in the UK. This paper contributes to understanding how STES may develop in the UK, and the mechanisms and challenges for widescale STES deployment, through case studies of five local heat projects.
Based on an analysis of both the emergence and absence of STES in local energy projects, we identify key factors enabling and inhibiting STES deployment, intersecting technological, economic, organisational and governance. We conclude that the limited extent of STES in the UK reflects the inconsistent alignment of these factors in local level heat projects, and disconnects between national energy policy, local energy planning, and project-level contingencies. Our findings suggest that without resolving these tensions, the UK heat transition will continue to be haphazard rather than strategic. While the situatedness of low carbon heat supply, storage and demand suggests moves towards more local energy governance, this needs to be accompanied by multi-level alignment and capacity building.
{"title":"Explaining the emergence and absence of Seasonal Thermal Energy Storage in the UK: Evidence from local case studies","authors":"","doi":"10.1016/j.apenergy.2024.124322","DOIUrl":"10.1016/j.apenergy.2024.124322","url":null,"abstract":"<div><div>Large-scale energy storage is highlighted as key for decarbonisation, yet there lacks consensus on the optimal types of storage required. Seasonal Thermal Energy Storage (STES) is an established feature of effective energy transitions in some countries, such as Denmark and the Netherlands, but it remains a marginal technology in the UK. This paper contributes to understanding how STES may develop in the UK, and the mechanisms and challenges for widescale STES deployment, through case studies of five local heat projects.</div><div>Based on an analysis of both the emergence and absence of STES in local energy projects, we identify key factors enabling and inhibiting STES deployment, intersecting technological, economic, organisational and governance. We conclude that the limited extent of STES in the UK reflects the inconsistent alignment of these factors in local level heat projects, and disconnects between national energy policy, local energy planning, and project-level contingencies. Our findings suggest that without resolving these tensions, the UK heat transition will continue to be haphazard rather than strategic. While the situatedness of low carbon heat supply, storage and demand suggests moves towards more local energy governance, this needs to be accompanied by multi-level alignment and capacity building.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-26DOI: 10.1016/j.apenergy.2024.124571
The setting of Intergovernmental Panel on Climate Change (IPCC) raises concerns on curbing the excessive carbon emissions, and European Union (EU) has been proactively attempting “green” initiatives striving to achieve carbon neutrality. The current Carbon Border Adjustment Mechanism (CBAM) only works on short-term carbon reduction without much “greenness” investment incentives so far, and few studies focus on the long-standing inner-industrial carbon reduction coordination. To address these issues, this paper creatively presents a dynamic multi-objective carbon responsibility allocation model (CRAM) for EU carbon market, allows the carbon trade with a taxation based on carbon offsets accounting, and considers the final targets of sustainable “greenness” investment incentives. To find the optimal results of the designed dynamic CRAM, an improved KT-NSGA-II algorithm is proposed to detect the Pareto frontiers of each successive periods. Data from EU Cement and Aluminium industries are then selected for empirical analysis to compare the proposed CRAM to conventional CBAM model. The findings demonstrate that the superiority of CRAM in carbon emissions reduction, economic benefits improvement and green invests encouragement. With the adjustment of the inner-industrial carbon allowance trade, total carbon emissions decreased by 28.03% and the “greenness” investment initiative increased by 39.24% in contrast to the CBAM. The sensitivity analysis of the model also provides suggestions on the settings of carbon quota and tax with different industrial production process, and proposes policy recommendations for CRAM implementation.
{"title":"A dynamic multi-objective optimization model for inner-industry carbon responsibility allocation based on carbon tax and carbon offsets accounting","authors":"","doi":"10.1016/j.apenergy.2024.124571","DOIUrl":"10.1016/j.apenergy.2024.124571","url":null,"abstract":"<div><div>The setting of Intergovernmental Panel on Climate Change (IPCC) raises concerns on curbing the excessive carbon emissions, and European Union (EU) has been proactively attempting “green” initiatives striving to achieve carbon neutrality. The current Carbon Border Adjustment Mechanism (CBAM) only works on short-term carbon reduction without much “greenness” investment incentives so far, and few studies focus on the long-standing inner-industrial carbon reduction coordination. To address these issues, this paper creatively presents a dynamic multi-objective carbon responsibility allocation model (CRAM) for EU carbon market, allows the carbon trade with a taxation based on carbon offsets accounting, and considers the final targets of sustainable “greenness” investment incentives. To find the optimal results of the designed dynamic CRAM, an improved KT-NSGA-II algorithm is proposed to detect the Pareto frontiers of each successive periods. Data from EU Cement and Aluminium industries are then selected for empirical analysis to compare the proposed CRAM to conventional CBAM model. The findings demonstrate that the superiority of CRAM in carbon emissions reduction, economic benefits improvement and green invests encouragement. With the adjustment of the inner-industrial carbon allowance trade, total carbon emissions decreased by 28.03% and the “greenness” investment initiative increased by 39.24% in contrast to the CBAM. The sensitivity analysis of the model also provides suggestions on the settings of carbon quota and tax with different industrial production process, and proposes policy recommendations for CRAM implementation.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-26DOI: 10.1016/j.apenergy.2024.124508
This research proposes a high-fidelity based numerical tank designed to analyze the modified hydrodynamics that develops in waves–current fields, aimed at generating power matrices for wave energy converters (WEC). This tank is developed within the open source DualSPHysics Lagrangian framework using the Smoothed Particle Hydrodynamics (SPH) method, validated with physical data, and applied to simulate a point-absorber WEC. Our proposed numerical facility implements open boundary conditions, employing third-order consistent wave theory for direct generation, with flow field constrained by a Doppler correlation function. Reference data is collected from dedicated physical tests for monochromatic waves; the wave–current numerical basin demonstrates very high accuracy in terms of wave transformation and velocity field. In the second segment of this paper, a current-aware power transfer function is computed for the taut-moored point-absorber Uppsala University WEC (UUWEC). Parametrically defined regular waves with uniform currents are utilized to map an operational sea state featuring currents of different directions and intensities. In terms of power capture capabilities, the modified dynamics observed in presence of currents translates in a dependence of the WEC’s power matrix not only on wave parameters, but also on current layouts. The UUWEC’s power output has revealed that regardless of current directionality, annual output consistently decreases, with a registered power drop as high as 10% when an expected current field is introduced.
{"title":"Development of an SPH-based numerical wave–current tank and application to wave energy converters","authors":"","doi":"10.1016/j.apenergy.2024.124508","DOIUrl":"10.1016/j.apenergy.2024.124508","url":null,"abstract":"<div><div>This research proposes a high-fidelity based numerical tank designed to analyze the modified hydrodynamics that develops in waves–current fields, aimed at generating power matrices for wave energy converters (WEC). This tank is developed within the open source DualSPHysics Lagrangian framework using the Smoothed Particle Hydrodynamics (SPH) method, validated with physical data, and applied to simulate a point-absorber WEC. Our proposed numerical facility implements open boundary conditions, employing third-order consistent wave theory for direct generation, with flow field constrained by a Doppler correlation function. Reference data is collected from dedicated physical tests for monochromatic waves; the wave–current numerical basin demonstrates very high accuracy in terms of wave transformation and velocity field. In the second segment of this paper, a current-aware power transfer function is computed for the taut-moored point-absorber Uppsala University WEC (UUWEC). Parametrically defined regular waves with uniform currents are utilized to map an operational sea state featuring currents of different directions and intensities. In terms of power capture capabilities, the modified dynamics observed in presence of currents translates in a dependence of the WEC’s power matrix not only on wave parameters, but also on current layouts. The UUWEC’s power output has revealed that regardless of current directionality, annual output consistently decreases, with a registered power drop as high as 10% when an expected current field is introduced.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-26DOI: 10.1016/j.apenergy.2024.124580
Accurate prediction of building electricity load is essential for grid management and building optimization operations. This paper proposes a novel approach based on spatiotemporal correlations and electricity consumption behavior information. The K-Medoids algorithm and the Derivative Dynamic Time Warping (DDTW) distance are employed to explore the correlation between electricity consumption behaviors among different partitions and floors. Different partitions and floors are clustered and grouped, followed by modifying the adjacency matrix with electricity consumption behaviors. The hybrid model and K-Medoids-LSTM model are proposed separately for clusterable nodes and non-clustered nodes. For clusterable nodes, spatial-temporal features are extracted, trained, and predicted with the hybrid model based on graph neural networks (GNNs) and LSTM models. A K-Medoids-LSTM model based on the K-Medoids algorithm is proposed to predict the electricity load of the non-clustered nodes. To explore the model's practicality, we predicted the building electrical load under different dataset sizes. The model achieves an R2 above 0.89, and the MAE, MSE, and RMSE of the GCN-LSTM and GAT-LSTM models all remain below 0.1, indicating strong predictive capabilities. The results demonstrate that, without relying on other external features, the proposed method can accurately predict the building electricity load for different partitions and floors simultaneously.
{"title":"Building electricity load forecasting based on spatiotemporal correlation and electricity consumption behavior information","authors":"","doi":"10.1016/j.apenergy.2024.124580","DOIUrl":"10.1016/j.apenergy.2024.124580","url":null,"abstract":"<div><div>Accurate prediction of building electricity load is essential for grid management and building optimization operations. This paper proposes a novel approach based on spatiotemporal correlations and electricity consumption behavior information. The K-Medoids algorithm and the Derivative Dynamic Time Warping (DDTW) distance are employed to explore the correlation between electricity consumption behaviors among different partitions and floors. Different partitions and floors are clustered and grouped, followed by modifying the adjacency matrix with electricity consumption behaviors. The hybrid model and K-Medoids-LSTM model are proposed separately for clusterable nodes and non-clustered nodes. For clusterable nodes, spatial-temporal features are extracted, trained, and predicted with the hybrid model based on graph neural networks (GNNs) and LSTM models. A K-Medoids-LSTM model based on the K-Medoids algorithm is proposed to predict the electricity load of the non-clustered nodes. To explore the model's practicality, we predicted the building electrical load under different dataset sizes. The model achieves an R<sup>2</sup> above 0.89, and the MAE, MSE, and RMSE of the GCN-LSTM and GAT-LSTM models all remain below 0.1, indicating strong predictive capabilities. The results demonstrate that, without relying on other external features, the proposed method can accurately predict the building electricity load for different partitions and floors simultaneously.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323047","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-26DOI: 10.1016/j.apenergy.2024.124540
The integrated electricity-heat microgrid (IEHM) is characterized by low inertia and a high share of renewable generation. Although sector coupling in the IEHM enhances energy efficiency, the integration of electricity and heat systems restricts the primary frequency regulation (PFR) ability of coupling equipment, further threatening frequency security. This paper presents a frequency-secured planning method for virtual inertia suppliers and combined heat and power (CHP) units in the IEHM. First, we examine the PFR in IEHMs from both device and system perspectives. We develop steady-state models for sector-coupled equipment, specifically CHP units and large-scale heat pumps, for PFR and sizing purposes. To describe the system’s frequency response under emergency conditions, we explicitly derive frequency constraints that account for varying system inertia of heterogeneous resources. Furthermore, we conduct a comprehensive analysis of the impact of PFR on heating systems within the IEHM. Second, we propose a frequency-constrained planning model for IEHMs based on the aforementioned modeling framework. This model balances the power supply, heating supply, and PFR reserves deployment of IEHM by properly sizing the regulation resources. It also leverages distributionally robust (DR) chance constraints to address uncertain wind power generation. To improve the tractability of this model, we introduce a well-tailored reformulation approach that handles the nonconvexity of system inertia and DR chance constraints. Case studies conducted on two test systems demonstrate the effectiveness of the proposed method in securing frequency stability while improving the economic performance of IEHM. This method ensures both dynamic and static frequency security across 100% of time steps by optimally deploying system inertia and PFR reserves. Moreover, by coordinating heterogeneous PFR resources with time-varying system inertia, the proposed approach yields superior economic performance, achieving over a 9% reduction in total capacity compared to average benchmarks.
{"title":"A frequency-secured planning method for integrated electricity-heat microgrids with virtual inertia suppliers","authors":"","doi":"10.1016/j.apenergy.2024.124540","DOIUrl":"10.1016/j.apenergy.2024.124540","url":null,"abstract":"<div><div>The integrated electricity-heat microgrid (IEHM) is characterized by low inertia and a high share of renewable generation. Although sector coupling in the IEHM enhances energy efficiency, the integration of electricity and heat systems restricts the primary frequency regulation (PFR) ability of coupling equipment, further threatening frequency security. This paper presents a frequency-secured planning method for virtual inertia suppliers and combined heat and power (CHP) units in the IEHM. First, we examine the PFR in IEHMs from both device and system perspectives. We develop steady-state models for sector-coupled equipment, specifically CHP units and large-scale heat pumps, for PFR and sizing purposes. To describe the system’s frequency response under emergency conditions, we explicitly derive frequency constraints that account for varying system inertia of heterogeneous resources. Furthermore, we conduct a comprehensive analysis of the impact of PFR on heating systems within the IEHM. Second, we propose a frequency-constrained planning model for IEHMs based on the aforementioned modeling framework. This model balances the power supply, heating supply, and PFR reserves deployment of IEHM by properly sizing the regulation resources. It also leverages distributionally robust (DR) chance constraints to address uncertain wind power generation. To improve the tractability of this model, we introduce a well-tailored reformulation approach that handles the nonconvexity of system inertia and DR chance constraints. Case studies conducted on two test systems demonstrate the effectiveness of the proposed method in securing frequency stability while improving the economic performance of IEHM. This method ensures both dynamic and static frequency security across 100% of time steps by optimally deploying system inertia and PFR reserves. Moreover, by coordinating heterogeneous PFR resources with time-varying system inertia, the proposed approach yields superior economic performance, achieving over a 9% reduction in total capacity compared to average benchmarks.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-26DOI: 10.1016/j.apenergy.2024.124492
The spatiotemporal prediction of wind power is of great significance for the grid connected operation of multiple wind farms in the wind power system. However, due to the complex temporal and spatial dependencies among multiple wind farms, developing advanced models to make accurate wind power predictions under their mutual influence is equally challenging. Furthermore, most of existing models are not ideal for long-term prediction of multivariate and non-stationary wind farm power datasets. To solve these problems, this paper proposes a novel Transformer-based model named Non-stationary GNNCrossformer for non-stationary multivariate Spatio-Temporal forecasting, utilizing Nonstationary-Two-Stage-Attention for both non-stationary cross-time dependency and cross-dimension dependency, as well as using the new graph convolutional neural network with Chebyshev interpolation for extracting temporally conditioned topological information from multiple wind farms efficiently. To tackle the dilemma between series predictability and model capability, we also propose Series Stationarization to complement Nonstationary-Two-Stage-Attention. While series stationarization makes sequence representation more generalized, the Nonstationary-Two-Stage-Attention can be devised to recover the intrinsic non-stationary information into temporal dependencies by approximating distinguishable attentions learned from raw series. Besides, the new graph convolutional neural network with Chebyshev interpolation can converge faster, be more robust, and have stronger generalization ability than the traditional one with Chebyshev approximation. In our experiment, two real-world wind power datasets were used to validate the proposed model. Numerical experiments have demonstrated the effectiveness and robustness of the proposed method compared to state-of-the-art spatiotemporal models.
{"title":"Non-stationary GNNCrossformer: Transformer with graph information for non-stationary multivariate Spatio-Temporal wind power data forecasting","authors":"","doi":"10.1016/j.apenergy.2024.124492","DOIUrl":"10.1016/j.apenergy.2024.124492","url":null,"abstract":"<div><div>The spatiotemporal prediction of wind power is of great significance for the grid connected operation of multiple wind farms in the wind power system. However, due to the complex temporal and spatial dependencies among multiple wind farms, developing advanced models to make accurate wind power predictions under their mutual influence is equally challenging. Furthermore, most of existing models are not ideal for long-term prediction of multivariate and non-stationary wind farm power datasets. To solve these problems, this paper proposes a novel Transformer-based model named Non-stationary GNNCrossformer for non-stationary multivariate Spatio-Temporal forecasting, utilizing Nonstationary-Two-Stage-Attention for both non-stationary cross-time dependency and cross-dimension dependency, as well as using the new graph convolutional neural network with Chebyshev interpolation for extracting temporally conditioned topological information from multiple wind farms efficiently. To tackle the dilemma between series predictability and model capability, we also propose Series Stationarization to complement Nonstationary-Two-Stage-Attention. While series stationarization makes sequence representation more generalized, the Nonstationary-Two-Stage-Attention can be devised to recover the intrinsic non-stationary information into temporal dependencies by approximating distinguishable attentions learned from raw series. Besides, the new graph convolutional neural network with Chebyshev interpolation can converge faster, be more robust, and have stronger generalization ability than the traditional one with Chebyshev approximation. In our experiment, two real-world wind power datasets were used to validate the proposed model. Numerical experiments have demonstrated the effectiveness and robustness of the proposed method compared to state-of-the-art spatiotemporal models.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":null,"pages":null},"PeriodicalIF":10.1,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}