This paper introduces a method for modeling the entire aggregated electric vehicle(EV)charging process and analyzing its dispatchable capabilities.The methodology involves developing a model for aggregated EV charging...This paper introduces a method for modeling the entire aggregated electric vehicle(EV)charging process and analyzing its dispatchable capabilities.The methodology involves developing a model for aggregated EV charging at the charging station level,estimating its physical dispatchable capability,determining its economic dispatchable capability under economic incentives,modeling its participation in the grid,and investigating the effects of different scenarios and EV penetration on the aggregated load dispatch and dispatchable capability.The results indicate that using economic dispatchable capability reduces charging prices by 9.7%compared to physical dispatchable capability and 9.3%compared to disorderly charging.Additionally,the peak-to-valley difference is reduced by 64.6%when applying economic dispatchable capability with 20%EV penetration and residential base load,compared to disorderly charging.展开更多
Time series foundation models provide a universal solution for generating forecasts to support optimization problems in energy systems.Those foundation models are typically trained in a prediction-focused manner to ma...Time series foundation models provide a universal solution for generating forecasts to support optimization problems in energy systems.Those foundation models are typically trained in a prediction-focused manner to maximize forecast quality.In contrast,decision-focused learning directly improves the resulting value of the forecast in downstream optimization rather than merely maximizing forecasting quality.The practical integration of forecast values into forecasting models is challenging,particularly when addressing complex applications with diverse instances,such as buildings.This becomes even more complicated when instances possess specific characteristics that require instance-specific,tailored predictions to increase the forecast value.To tackle this challenge,we use decision-focused fine-tuning within time series foundation models to offer a scalable and efficient solution for decision-focused learning applied to the dispatchable feeder optimization problem.To obtain more robust predictions for scarce building data,we use Moirai as a state-of-the-art foundation model,which offers robust and generalized results with few-shot parameter-efficient fine-tuning.Comparing the decision-focused fine-tuned Moirai with a state-of-the-art classical prediction-focused fine-tuning Moirai,we observe an improvement of 9.45%in Average Daily Total Costs.展开更多
This paper describes the research on a largescale dispatchable grid-connected photovoltaic(PV)system for supplying power to the grid for dispatch instead of supplying the electricity to a local load.In order to maximi...This paper describes the research on a largescale dispatchable grid-connected photovoltaic(PV)system for supplying power to the grid for dispatch instead of supplying the electricity to a local load.In order to maximise the value of the solar energy,a hybrid electricity storage consisting of batteries and supercapacitors is used with the PV system.This paper proposes a control strategy focusing on theDCpower at theDClink rather than at the grid-connected inverter.Two typical sets of real data,collected from existing sites,are used to demonstrate the practicality of the system.Finally,the simulation results are used to demonstrate the good performance and feasibility of the proposed system together with the proposed control strategy.展开更多
A large amount of renewable energy generation(REG)has been integrated into power systems,challenging the operational security of power networks.In a real-time dispatch,system operators need to estimate the ability of ...A large amount of renewable energy generation(REG)has been integrated into power systems,challenging the operational security of power networks.In a real-time dispatch,system operators need to estimate the ability of the power network to accommodate REG with a limited reserve capacity.The real-time dispatchable region(RTDR)is defined as the largest range of a power injection that the power network can accommodate in a certain dispatch interval for a given dispatch base point.State-of-the-art research on the RTDR adopts a DC power flow model regardless of the voltage profiles and reactive power,which can overlook potentially insecure operational states of the system.To address this issue,this paper proposes an AC power flow based RTDR model simultaneously considering the reactive power and voltage profiles constraints.The nonlinear constraints in our model are approximated using a linear power flow model together with a polytope approximation technique for quadratic constraints.An adaptive constraint generation algorithm is used to calculate the RTDR.Simulation results using the IEEE 5-bus and 30-bus systems illustrate the advantages of the proposed model.展开更多
Uncertainty in distributed renewable generation threatens the security of power distribution systems.The concept of dispatchable region is developed to assess the ability of power systems to accommodate renewable gene...Uncertainty in distributed renewable generation threatens the security of power distribution systems.The concept of dispatchable region is developed to assess the ability of power systems to accommodate renewable generation at a given operating point.Although DC and linearized AC power flow equations are typically used to model dispatchable regions for transmission systems,these equations are rarely suitable for distribution networks.To achieve a suitable trade-off between accuracy and efficiency,this paper proposes a dispatchable region formulation for distribution networks using tight convex relaxation.Secondorder cone relaxation is adopted to reformulate AC power flow equations,which are then approximated by a polyhedron to improve tractability.Further,an efficient adaptive constraint generation algorithm is employed to construct the proposed dispatchable region.Case studies on distribution systems of various scales validate the computational efficiency and accuracy of the proposed method.展开更多
The advent of microgrids in modern energy systems heralds a promising era of resilience,sustainability,and efficiency.Within the realm of grid-tied microgrids,the selection of an optimal optimization algorithm is crit...The advent of microgrids in modern energy systems heralds a promising era of resilience,sustainability,and efficiency.Within the realm of grid-tied microgrids,the selection of an optimal optimization algorithm is critical for effective energy management,particularly in economic dispatching.This study compares the performance of Particle Swarm Optimization(PSO)and Genetic Algorithms(GA)in microgrid energy management systems,implemented using MATLAB tools.Through a comprehensive review of the literature and sim-ulations conducted in MATLAB,the study analyzes performance metrics,convergence speed,and the overall efficacy of GA and PSO,with a focus on economic dispatching tasks.Notably,a significant distinction emerges between the cost curves generated by the two algo-rithms for microgrid operation,with the PSO algorithm consistently resulting in lower costs due to its effective economic dispatching capabilities.Specifically,the utilization of the PSO approach could potentially lead to substantial savings on the power bill,amounting to approximately$15.30 in this evaluation.Thefindings provide insights into the strengths and limitations of each algorithm within the complex dynamics of grid-tied microgrids,thereby assisting stakeholders and researchers in arriving at informed decisions.This study contributes to the discourse on sustainable energy management by offering actionable guidance for the advancement of grid-tied micro-grid technologies through MATLAB-implemented optimization algorithms.展开更多
The penetration rate of new wind and photovoltaic energy in the power system has increased significantly,and the dramatic fluctuation of the net load of the grid has led to a severe lack of flexibility in the regional...The penetration rate of new wind and photovoltaic energy in the power system has increased significantly,and the dramatic fluctuation of the net load of the grid has led to a severe lack of flexibility in the regional grid.This paper proposes a hierarchical optimal dispatch strategy for a high proportion of new energy power systems that considers the balanced response of grid flexibility.Firstly,various flexibility resource regulation capabilities on the source-load side are analyzed,and then flexibility demand and flexibility response are matched,and flexibility demand response assessment is proposed;then,a hierarchical optimal dispatch model of the grid taking flexibility adjustment capability into account is established,and the upper model optimizes the net load curve with the objectives of minimizing the fluctuation of the net load,maximizing the benefits of energy storage and controllable loads,and optimizing the flexibility adjustment capability.The upper layer model optimizes the net load curve by minimizing net load fluctuation,maximizing energy storage and controllable load revenue,and optimizing flexibility adjustment capability.In contrast,the lower layer model optimizes the power allocation of thermal power units and regulates the lost load of wind and solar power generation by minimizing the total system operating cost.The results show that the proposed strategy improves the flexibility of the grid by 15.2%,gives full play to the regulation capability of each flexibility resource,and reduces the fluctuation of the net load by 15.6%to achieve optimal coordination between different types of flexibility resources.展开更多
With the severe challenges brought by global climate change,exploring and developing clean and renewable energy systems to upgrade the energy structure has become an inevitable trend in related research.The comprehens...With the severe challenges brought by global climate change,exploring and developing clean and renewable energy systems to upgrade the energy structure has become an inevitable trend in related research.The comprehensive park systems integrated with photovoltaic,energy storage,direct current,and flexible loads(PEDF)is able to play an important role in promoting energy transformation and achieving sustainable development.In order to fully understand the advantages of PEDF parks in energy conservation and carbon reduction,this paper summarizes existing studies and prospects future research directions on the low-carbon operation of the PEDF park.This paper first introduces carbon emission monitoring and evaluation methods,and then analyzes bi-level optimal dispatch strategies for flexible loads.Meanwhile,the paper provides a prospective analysis of the innovations that can be brought by advanced technologies to the PEDF park.Finally,this paper puts forward the challenges faced by current research and provides prospects for future research directions.This paper emphasizes that related research should focus on collaborating key technologies of PEDF systems and integrating advanced innovations to address challenges and fully leverage the advantages of PEDF technology in energy saving and carbon reduction.This paper aims to provide systematic theoretical guidance and supplements for the research and practice of the PEDF technology.展开更多
With the intensification of the energy crisis and the worsening greenhouse effect,the development of sustainable integrated energy systems(IES)has become a crucial direction for energy transition.In this context,this ...With the intensification of the energy crisis and the worsening greenhouse effect,the development of sustainable integrated energy systems(IES)has become a crucial direction for energy transition.In this context,this paper proposes a low-carbon economic dispatch strategy under the green hydrogen certificate trading(GHCT)and the ladder-type carbon emission trading(CET)mechanism,enabling the coordinated utilization of green and blue hydrogen.Specifically,a proton exchange membrane electrolyzer(PEME)model that accounts for dynamic efficiency characteristics,and a steam methane reforming(SMR)model incorporating waste heat recovery,are developed.Based on these models,a hydrogen production–storage–utilization framework is established to enable the coordinated deployment of green and blue hydrogen.Furthermore,the gas turbine(GT)unit are retrofitted using oxygenenriched combustion carbon capture(OCC)technology,wherein the oxygen produced by PEME is employed to create an oxygen-enriched combustion environment.This approach reduces energy waste and facilitates low-carbon power generation.In addition,the GHCT mechanism is integrated into the system alongside the ladder-type CET mechanism,and their complementary effects are investigated.A comprehensive optimization model is then formulated to simultaneously achieve carbon reduction and economic efficiency across the system.Case study results show that the proposed strategy reduces wind curtailment by 7.77%,carbon emissions by 65.98%,and total cost by 12.57%.This study offers theoretical reference for the low-carbon,economic,and efficient operation of future energy systems.展开更多
The economic dispatch problem(EDP) of microgrids operating in both grid-connected and isolated modes within an energy internet framework is addressed in this paper. The multi-agent leader-following consensus algorithm...The economic dispatch problem(EDP) of microgrids operating in both grid-connected and isolated modes within an energy internet framework is addressed in this paper. The multi-agent leader-following consensus algorithm is employed to address the EDP of microgrids in grid-connected mode, while the push-pull algorithm with a fixed step size is introduced for the isolated mode. The proposed algorithm of isolated mode is proven to converge to the optimum when the interaction digraph of microgrids is strongly connected. A unified algorithmic framework is proposed to handle the two modes of operation of microgrids simultaneously, enabling our algorithm to achieve optimal power allocation and maintain the balance between power supply and demand in any mode and any mode switching. Due to the push-pull structure of the algorithm and the use of fixed step size,the proposed algorithm can better handle the case of unbalanced graphs, and the convergence speed is improved. It is documented that when the transmission topology is strongly connected and there is bi-directional communication between the energy router and its neighbors, the proposed algorithm in composite mode achieves economic dispatch even with arbitrary mode switching.Finally, we demonstrate the effectiveness and superiority of our algorithm through numerical simulations.展开更多
Considering the special features of dynamic environment economic dispatch of power systems with high dimensionality,strong coupling,nonlinearity,and non-convexity,a GA-DE multi-objective optimization algorithm based o...Considering the special features of dynamic environment economic dispatch of power systems with high dimensionality,strong coupling,nonlinearity,and non-convexity,a GA-DE multi-objective optimization algorithm based on dual-population pseudo-parallel genetic algorithm-differential evolution is proposed in this paper.The algorithm is based on external elite archive and Pareto dominance,and it adopts the cooperative co-evolution mechanism of differential evolution and genetic algorithm.Average entropy and cubic chaoticmapping initialization strategies are proposed to increase population diversity.In the proposed method,we analyze the distribution of neighboring solutions and apply a new Pareto solution set pruning approach.Unlike traditional models,this work takes the transmission losses as an optimization target and overcomes complex model constraints through a dynamic relaxation constraint approach.To solve the uncertainty caused by integrating wind and photovoltaic energy in power system scheduling,a multi-objective dynamic environment economical dispatch model is set up that takes the system spinning reserve and network highest losses into account.In this paper,the DE algorithm is improved to form the DGAGE algorithm for the objective optimization of the overall power system,The DE algorithm part of DGAGE is combined with the JAYA algorithm to form the system scheduling HDJ algorithm for multiple energy sources connected to the grid.The effectiveness of the proposed method is demonstrated using CEC2022 and CEC2005 test functions,showing robust optimization performance.Validation on a classical 10-unit system confirms the feasibility of the proposed algorithm in addressing power system scheduling issues.This approach provides a novel solution for dynamic power dispatch systems.展开更多
The integration of deep learning into smart grid operations addresses critical challenges in dynamic load forecasting and optimal dispatch amid increasing renewable energy penetration.This study proposes a hybrid LSTM...The integration of deep learning into smart grid operations addresses critical challenges in dynamic load forecasting and optimal dispatch amid increasing renewable energy penetration.This study proposes a hybrid LSTM-Transformer architecture for multi-scale temporal-spatial load prediction,achieving 28%RMSE reduction on real-world datasets(CAISO,PJM),coupled with a deep reinforcement learning framework for multi-objective dispatch optimization that lowers operational costs by 12.4%while ensuring stability constraints.The synergy between adaptive forecasting models and scenario-based stochastic optimization demonstrates superior performance in handling renewable intermittency and demand volatility,validated through grid-scale case studies.Methodological innovations in federated feature extraction and carbon-aware scheduling further enhance scalability for distributed energy systems.These advancements provide actionable insights for grid operators transitioning to low-carbon paradigms,emphasizing computational efficiency and interoperability with legacy infrastructure.展开更多
In recent years,railway construction in China has developed vigorously.With continuous improvements in the highspeed railway network,the focus is gradually shifting from large-scale construction to large-scale operati...In recent years,railway construction in China has developed vigorously.With continuous improvements in the highspeed railway network,the focus is gradually shifting from large-scale construction to large-scale operations.However,several challenges have emerged within the high-speed railway dispatching and command system,including the heavy workload faced by dispatchers,the difficulty of quantifying subjective expertise,and the need for effective training of professionals.Amid the growing application of artificial intelligence technologies in railway systems,this study leverages Large Language Model(LLM)technology.LLMs bring enhanced intelligence,predictive capabilities,robust memory,and adaptability to diverse real-world scenarios.This study proposes a human-computer interactive intelligent scheduling auxiliary training system built on LLM technology.The system offers capabilities including natural dialogue,knowledge reasoning,and human feedback learning.With broad applicability,the system is suitable for vocational education,guided inquiry,knowledge-based Q&A,and other training scenarios.Validation results demonstrate its effectiveness in auxiliary training,providing substantial support for educators,students,and dispatching personnel in colleges and professional settings.展开更多
With the increasing penetration of renewable energy resources in power systems,conventional timescale separated load frequency control(LFC)and economic dispatch may degrade frequency performance and reduce economic ef...With the increasing penetration of renewable energy resources in power systems,conventional timescale separated load frequency control(LFC)and economic dispatch may degrade frequency performance and reduce economic efficiency.This paper proposes a novel data-driven adaptive distributed optimal disturbance rejection control(DODRC)method for real-time economic LFC problem in nonlinear power systems.Firstly,a basic DODRC method is proposed by integrating the active disturbance rejection control method and the partial primal–dual algorithm.Then,to deal with the tie-line power flow constraints,the logarithmic barrier function is employed to reconstruct the Lagrange function to obtain the constrained DODRC method.By analyzing the sensitivity of the uncertain parameters of power systems,a data-driven adaptive DODRC method is finally proposed with a neural network.The effectiveness of the proposed method is demonstrated by experimental results using real-time equipment.展开更多
In order to address the synergistic optimization of energy efficiency improvement in the waste incineration power plant(WIPP)and renewable energy accommodation,an electricity-hydrogen-waste multi-energy system integra...In order to address the synergistic optimization of energy efficiency improvement in the waste incineration power plant(WIPP)and renewable energy accommodation,an electricity-hydrogen-waste multi-energy system integrated with phase change material(PCM)thermal storage is proposed.First,a thermal energy management framework is constructed,combining PCM thermal storage with the alkaline electrolyzer(AE)waste heat recovery and the heat pump(HP),while establishing a PCM-driven waste drying system to enhance the efficiency of waste incineration power generation.Next,a flue gas treatment method based on purification-separation-storage coordination is adopted,achieving spatiotemporal decoupling between waste incineration and flue gas treatment.Subsequently,a two-stage optimal dispatching strategy for the multi-energy system is developed:the first stage establishes a dayahead economic dispatch model with the objective of minimizing net system costs,while the second stage introduces model predictive control(MPC)to realize intraday rolling optimization.Finally,The optimal dispatching strategies under different scenarios are obtained using the Gurobi solver,followed by a comparative analysis of the optimized operational outcomes.Simulation results demonstrate that the proposed system optimizes the output and operational states of each unit,simultaneously reducing carbon trading costs while increasing electricity sales revenue.The proposed scheduling strategy demonstrates effective grid peak-shaving functionality,thereby simultaneously improving the system’s economic performance and operational flexibility while providing an innovative technical pathway for municipal solid waste(MSW)resource utilization and low-carbon transformation of energy systems.展开更多
The exponential growth in the scale of power systems has led to a significant increase in the complexity of dispatch problem resolution,particularly within multi-area interconnected power grids.This complexity necessi...The exponential growth in the scale of power systems has led to a significant increase in the complexity of dispatch problem resolution,particularly within multi-area interconnected power grids.This complexity necessitates the employment of distributed solution methodologies,which are not only essential but also highly desirable.In the realm of computational modelling,the multi-area economic dispatch problem(MAED)can be formulated as a linearly constrained separable convex optimization problem.The proximal point algorithm(PPA)is particularly adept at addressing such mathematical constructs effectively.This study introduces parallel(PPPA)and serial(SPPA)variants of the PPA as distributed algorithms,specifically designed for the computational modelling of the MAED.The PPA introduces a quadratic term into the objective function,which,while potentially complicating the iterative updates of the algorithm,serves to dampen oscillations near the optimal solution,thereby enhancing the convergence characteristics.Furthermore,the convergence efficiency of the PPA is significantly influenced by the parameter c.To address this parameter sensitivity,this research draws on trend theory from stock market analysis to propose trend theory-driven distributed PPPA and SPPA,thereby enhancing the robustness of the computational models.The computational models proposed in this study are anticipated to exhibit superior performance in terms of convergence behaviour,stability,and robustness with respect to parameter selection,potentially outperforming existing methods such as the alternating direction method of multipliers(ADMM)and Auxiliary Problem Principle(APP)in the computational simulation of power system dispatch problems.The simulation results demonstrate that the trend theory-based PPPA,SPPA,ADMM and APP exhibit significant robustness to the initial value of parameter c,and show superior convergence characteristics compared to the residual balancing ADMM.展开更多
As future ship system,hybrid energy ship system has a wide range of application prospects for solving the serious energy crisis.However,current optimization scheduling works lack the consideration of sea conditions an...As future ship system,hybrid energy ship system has a wide range of application prospects for solving the serious energy crisis.However,current optimization scheduling works lack the consideration of sea conditions and navigational circumstances.There-fore,this paper aims at establishing a two-stage optimization framework for hybrid energy ship power system.The proposed framework considers multiple optimizations of route,speed planning,and energy management under the constraints of sea conditions during navigation.First,a complex hybrid ship power model consisting of diesel generation system,propulsion system,energy storage system,photovoltaic power generation system,and electric boiler system is established,where sea state information and ship resistance model are considered.With objective optimization functions of cost and greenhouse gas(GHG)emissions,a two-stage optimization framework consisting of route planning,speed scheduling,and energy management is constructed.Wherein the improved A-star algorithm and grey wolf optimization algorithm are introduced to obtain the optimal solutions for route,speed,and energy optimization scheduling.Finally,simulation cases are employed to verify that the proposed two-stage optimization scheduling model can reduce load energy consumption,operating costs,and carbon emissions by 17.8%,17.39%,and 13.04%,respectively,compared with the non-optimal control group.展开更多
Taking Huanghua Port Railway Station of the Shuozhou-Huanghua Railway as a demonstration case,an overall solution for the 5G-based intelligent shunting system at heavy haul railway stations was developed to address th...Taking Huanghua Port Railway Station of the Shuozhou-Huanghua Railway as a demonstration case,an overall solution for the 5G-based intelligent shunting system at heavy haul railway stations was developed to address the operational complexities,inadequacies of outdated equipment,and low efficiency experienced by shunting operators.The system utilizes a 5G communication platform to facilitate automated and intelligent shunting operations at heavy haul railway stations.Advanced technological equipment for intelligent shunting in heavy haul railways was developed,encompassing a big data center,intelligent dispatching and control systems,automated and remote operation of locomotives,intelligent cloud-based video surveillance,intelligent dual-powered electric locomotive,and a customized 5G private network.Technical measures are implemented to reduce operators'labor intensity,decrease the number of on-site personnel,ensure effective safety protection for operators,improve utilization of arrival and departure tracks at heavy haul railway stations,and promote the development of“smart,intelligent,interconnected,and sensing”heavy haul railway stations.展开更多
A chance-constrained energy dispatch model based on the distributed stochastic model predictive control(DSMPC)approach for an islanded multi-microgrid system is proposed.An ambiguity set considering the inherent uncer...A chance-constrained energy dispatch model based on the distributed stochastic model predictive control(DSMPC)approach for an islanded multi-microgrid system is proposed.An ambiguity set considering the inherent uncertainties of renewable energy sources(RESs)is constructed without requiring the full distribution knowledge of the uncertainties.The power balance chance constraint is reformulated within the framework of the distributionally robust optimization(DRO)approach.With the exchange of information and energy flow,each microgrid can achieve its local supply-demand balance.Furthermore,the closed-loop stability and recursive feasibility of the proposed algorithm are proved.The comparative results with other DSMPC methods show that a trade-off between robustness and economy can be achieved.展开更多
This paper presents a novel approach to economic dispatch in smart grids equipped with diverse energy devices.This method integrates features including photovoltaic(PV)systems,energy storage coupling,varied energy rol...This paper presents a novel approach to economic dispatch in smart grids equipped with diverse energy devices.This method integrates features including photovoltaic(PV)systems,energy storage coupling,varied energy roles,and energy supply and demand dynamics.The systemmodel is developed by considering energy devices as versatile units capable of fulfilling various functionalities and playing multiple roles simultaneously.To strike a balance between optimality and feasibility,renewable energy resources are modeled with considerations for forecasting errors,Gaussian distribution,and penalty factors.Furthermore,this study introduces a distributed event-triggered surplus algorithm designed to address the economic dispatch problem by minimizing production costs.Rooted in surplus theory and finite time projection,the algorithm effectively rectifies network imbalances caused by directed graphs and addresses local inequality constraints.The algorithm greatly reduces the communication burden through event triggering mechanism.Finally,both theoretical proofs and numerical simulations verify the convergence and event-triggered nature of the algorithm.展开更多
基金State Grid Henan Power Company Science and Technology Project‘Key Technology and Demonstration Application of Multi-Domain Electric Vehicle Aggregated Charging Load Dispatch’(5217L0240003).
文摘This paper introduces a method for modeling the entire aggregated electric vehicle(EV)charging process and analyzing its dispatchable capabilities.The methodology involves developing a model for aggregated EV charging at the charging station level,estimating its physical dispatchable capability,determining its economic dispatchable capability under economic incentives,modeling its participation in the grid,and investigating the effects of different scenarios and EV penetration on the aggregated load dispatch and dispatchable capability.The results indicate that using economic dispatchable capability reduces charging prices by 9.7%compared to physical dispatchable capability and 9.3%compared to disorderly charging.Additionally,the peak-to-valley difference is reduced by 64.6%when applying economic dispatchable capability with 20%EV penetration and residential base load,compared to disorderly charging.
基金funded by the Helmholtz Association’s Initiative and Networking Fund through Helmholtz AI,the Helmholtz Association under the Program“Energy System Design”the German Research Foundation(DFG)as part of the Research Training Group 2153“En-ergy Status Data:Informatics Methods for its Collection,Analysis and Exploitation”+1 种基金supported by the Helmholtz Association Initiative and Networking Fund on the HAICORE@KIT partitionsupport by the KIT-Publication Fund of the Karlsruhe Institute of Technology.
文摘Time series foundation models provide a universal solution for generating forecasts to support optimization problems in energy systems.Those foundation models are typically trained in a prediction-focused manner to maximize forecast quality.In contrast,decision-focused learning directly improves the resulting value of the forecast in downstream optimization rather than merely maximizing forecasting quality.The practical integration of forecast values into forecasting models is challenging,particularly when addressing complex applications with diverse instances,such as buildings.This becomes even more complicated when instances possess specific characteristics that require instance-specific,tailored predictions to increase the forecast value.To tackle this challenge,we use decision-focused fine-tuning within time series foundation models to offer a scalable and efficient solution for decision-focused learning applied to the dispatchable feeder optimization problem.To obtain more robust predictions for scarce building data,we use Moirai as a state-of-the-art foundation model,which offers robust and generalized results with few-shot parameter-efficient fine-tuning.Comparing the decision-focused fine-tuned Moirai with a state-of-the-art classical prediction-focused fine-tuning Moirai,we observe an improvement of 9.45%in Average Daily Total Costs.
基金This work was supported by National High Technology Research and Development Program,“863 key technologies and development on large-scale grid-connected PV plants”(No.2011AA05A301).
文摘This paper describes the research on a largescale dispatchable grid-connected photovoltaic(PV)system for supplying power to the grid for dispatch instead of supplying the electricity to a local load.In order to maximise the value of the solar energy,a hybrid electricity storage consisting of batteries and supercapacitors is used with the PV system.This paper proposes a control strategy focusing on theDCpower at theDClink rather than at the grid-connected inverter.Two typical sets of real data,collected from existing sites,are used to demonstrate the practicality of the system.Finally,the simulation results are used to demonstrate the good performance and feasibility of the proposed system together with the proposed control strategy.
基金This work was supported in part by the Science and Technology Program of Guangzhou under Grant 201904010215the State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources under Grant LAPS19011the Fundamental Research Funds for the Central Universities.
文摘A large amount of renewable energy generation(REG)has been integrated into power systems,challenging the operational security of power networks.In a real-time dispatch,system operators need to estimate the ability of the power network to accommodate REG with a limited reserve capacity.The real-time dispatchable region(RTDR)is defined as the largest range of a power injection that the power network can accommodate in a certain dispatch interval for a given dispatch base point.State-of-the-art research on the RTDR adopts a DC power flow model regardless of the voltage profiles and reactive power,which can overlook potentially insecure operational states of the system.To address this issue,this paper proposes an AC power flow based RTDR model simultaneously considering the reactive power and voltage profiles constraints.The nonlinear constraints in our model are approximated using a linear power flow model together with a polytope approximation technique for quadratic constraints.An adaptive constraint generation algorithm is used to calculate the RTDR.Simulation results using the IEEE 5-bus and 30-bus systems illustrate the advantages of the proposed model.
基金the National Natural Science Foundation of China(Grant No.52177086)the Fundamental Research Funds for the Central Universities(Grant No.2023ZYGXZR063)。
文摘Uncertainty in distributed renewable generation threatens the security of power distribution systems.The concept of dispatchable region is developed to assess the ability of power systems to accommodate renewable generation at a given operating point.Although DC and linearized AC power flow equations are typically used to model dispatchable regions for transmission systems,these equations are rarely suitable for distribution networks.To achieve a suitable trade-off between accuracy and efficiency,this paper proposes a dispatchable region formulation for distribution networks using tight convex relaxation.Secondorder cone relaxation is adopted to reformulate AC power flow equations,which are then approximated by a polyhedron to improve tractability.Further,an efficient adaptive constraint generation algorithm is employed to construct the proposed dispatchable region.Case studies on distribution systems of various scales validate the computational efficiency and accuracy of the proposed method.
文摘The advent of microgrids in modern energy systems heralds a promising era of resilience,sustainability,and efficiency.Within the realm of grid-tied microgrids,the selection of an optimal optimization algorithm is critical for effective energy management,particularly in economic dispatching.This study compares the performance of Particle Swarm Optimization(PSO)and Genetic Algorithms(GA)in microgrid energy management systems,implemented using MATLAB tools.Through a comprehensive review of the literature and sim-ulations conducted in MATLAB,the study analyzes performance metrics,convergence speed,and the overall efficacy of GA and PSO,with a focus on economic dispatching tasks.Notably,a significant distinction emerges between the cost curves generated by the two algo-rithms for microgrid operation,with the PSO algorithm consistently resulting in lower costs due to its effective economic dispatching capabilities.Specifically,the utilization of the PSO approach could potentially lead to substantial savings on the power bill,amounting to approximately$15.30 in this evaluation.Thefindings provide insights into the strengths and limitations of each algorithm within the complex dynamics of grid-tied microgrids,thereby assisting stakeholders and researchers in arriving at informed decisions.This study contributes to the discourse on sustainable energy management by offering actionable guidance for the advancement of grid-tied micro-grid technologies through MATLAB-implemented optimization algorithms.
文摘The penetration rate of new wind and photovoltaic energy in the power system has increased significantly,and the dramatic fluctuation of the net load of the grid has led to a severe lack of flexibility in the regional grid.This paper proposes a hierarchical optimal dispatch strategy for a high proportion of new energy power systems that considers the balanced response of grid flexibility.Firstly,various flexibility resource regulation capabilities on the source-load side are analyzed,and then flexibility demand and flexibility response are matched,and flexibility demand response assessment is proposed;then,a hierarchical optimal dispatch model of the grid taking flexibility adjustment capability into account is established,and the upper model optimizes the net load curve with the objectives of minimizing the fluctuation of the net load,maximizing the benefits of energy storage and controllable loads,and optimizing the flexibility adjustment capability.The upper layer model optimizes the net load curve by minimizing net load fluctuation,maximizing energy storage and controllable load revenue,and optimizing flexibility adjustment capability.In contrast,the lower layer model optimizes the power allocation of thermal power units and regulates the lost load of wind and solar power generation by minimizing the total system operating cost.The results show that the proposed strategy improves the flexibility of the grid by 15.2%,gives full play to the regulation capability of each flexibility resource,and reduces the fluctuation of the net load by 15.6%to achieve optimal coordination between different types of flexibility resources.
基金This work was supported by National Key R&D Program of China for International S&T Cooperation Projects(Grant No.2019YFE0118700)which was provided by the Ministry of Science and Technology of the People’s Republic of China(https://www.most.gov.cn/(accessed on 1 January 2025))+2 种基金the grant was received by Yun Zhao.This work was supported by Science and Technology Project of CSG Electric Power Research Institute(Grant No.SEPRIK23B052)which was provided by CSG Electric Power Research Institute(http://www.sepri.csg.cn/(accessed on 1 January 2025))the grant was received by Ziwen Cai.
文摘With the severe challenges brought by global climate change,exploring and developing clean and renewable energy systems to upgrade the energy structure has become an inevitable trend in related research.The comprehensive park systems integrated with photovoltaic,energy storage,direct current,and flexible loads(PEDF)is able to play an important role in promoting energy transformation and achieving sustainable development.In order to fully understand the advantages of PEDF parks in energy conservation and carbon reduction,this paper summarizes existing studies and prospects future research directions on the low-carbon operation of the PEDF park.This paper first introduces carbon emission monitoring and evaluation methods,and then analyzes bi-level optimal dispatch strategies for flexible loads.Meanwhile,the paper provides a prospective analysis of the innovations that can be brought by advanced technologies to the PEDF park.Finally,this paper puts forward the challenges faced by current research and provides prospects for future research directions.This paper emphasizes that related research should focus on collaborating key technologies of PEDF systems and integrating advanced innovations to address challenges and fully leverage the advantages of PEDF technology in energy saving and carbon reduction.This paper aims to provide systematic theoretical guidance and supplements for the research and practice of the PEDF technology.
基金supported by National Natural Science Foundation of China(52477101)Natural Science Foundation of Jiangsu Province(BK20210932).
文摘With the intensification of the energy crisis and the worsening greenhouse effect,the development of sustainable integrated energy systems(IES)has become a crucial direction for energy transition.In this context,this paper proposes a low-carbon economic dispatch strategy under the green hydrogen certificate trading(GHCT)and the ladder-type carbon emission trading(CET)mechanism,enabling the coordinated utilization of green and blue hydrogen.Specifically,a proton exchange membrane electrolyzer(PEME)model that accounts for dynamic efficiency characteristics,and a steam methane reforming(SMR)model incorporating waste heat recovery,are developed.Based on these models,a hydrogen production–storage–utilization framework is established to enable the coordinated deployment of green and blue hydrogen.Furthermore,the gas turbine(GT)unit are retrofitted using oxygenenriched combustion carbon capture(OCC)technology,wherein the oxygen produced by PEME is employed to create an oxygen-enriched combustion environment.This approach reduces energy waste and facilitates low-carbon power generation.In addition,the GHCT mechanism is integrated into the system alongside the ladder-type CET mechanism,and their complementary effects are investigated.A comprehensive optimization model is then formulated to simultaneously achieve carbon reduction and economic efficiency across the system.Case study results show that the proposed strategy reduces wind curtailment by 7.77%,carbon emissions by 65.98%,and total cost by 12.57%.This study offers theoretical reference for the low-carbon,economic,and efficient operation of future energy systems.
基金supported by the National Natural Science Foundation of China(62103203)
文摘The economic dispatch problem(EDP) of microgrids operating in both grid-connected and isolated modes within an energy internet framework is addressed in this paper. The multi-agent leader-following consensus algorithm is employed to address the EDP of microgrids in grid-connected mode, while the push-pull algorithm with a fixed step size is introduced for the isolated mode. The proposed algorithm of isolated mode is proven to converge to the optimum when the interaction digraph of microgrids is strongly connected. A unified algorithmic framework is proposed to handle the two modes of operation of microgrids simultaneously, enabling our algorithm to achieve optimal power allocation and maintain the balance between power supply and demand in any mode and any mode switching. Due to the push-pull structure of the algorithm and the use of fixed step size,the proposed algorithm can better handle the case of unbalanced graphs, and the convergence speed is improved. It is documented that when the transmission topology is strongly connected and there is bi-directional communication between the energy router and its neighbors, the proposed algorithm in composite mode achieves economic dispatch even with arbitrary mode switching.Finally, we demonstrate the effectiveness and superiority of our algorithm through numerical simulations.
基金funded by the Major Humanities and Social Sciences Research Projects in Zhejiang Higher Education Institutions,grant number 2023QN131National Innovation Training Program Project in China,grant number 202410451009.
文摘Considering the special features of dynamic environment economic dispatch of power systems with high dimensionality,strong coupling,nonlinearity,and non-convexity,a GA-DE multi-objective optimization algorithm based on dual-population pseudo-parallel genetic algorithm-differential evolution is proposed in this paper.The algorithm is based on external elite archive and Pareto dominance,and it adopts the cooperative co-evolution mechanism of differential evolution and genetic algorithm.Average entropy and cubic chaoticmapping initialization strategies are proposed to increase population diversity.In the proposed method,we analyze the distribution of neighboring solutions and apply a new Pareto solution set pruning approach.Unlike traditional models,this work takes the transmission losses as an optimization target and overcomes complex model constraints through a dynamic relaxation constraint approach.To solve the uncertainty caused by integrating wind and photovoltaic energy in power system scheduling,a multi-objective dynamic environment economical dispatch model is set up that takes the system spinning reserve and network highest losses into account.In this paper,the DE algorithm is improved to form the DGAGE algorithm for the objective optimization of the overall power system,The DE algorithm part of DGAGE is combined with the JAYA algorithm to form the system scheduling HDJ algorithm for multiple energy sources connected to the grid.The effectiveness of the proposed method is demonstrated using CEC2022 and CEC2005 test functions,showing robust optimization performance.Validation on a classical 10-unit system confirms the feasibility of the proposed algorithm in addressing power system scheduling issues.This approach provides a novel solution for dynamic power dispatch systems.
文摘The integration of deep learning into smart grid operations addresses critical challenges in dynamic load forecasting and optimal dispatch amid increasing renewable energy penetration.This study proposes a hybrid LSTM-Transformer architecture for multi-scale temporal-spatial load prediction,achieving 28%RMSE reduction on real-world datasets(CAISO,PJM),coupled with a deep reinforcement learning framework for multi-objective dispatch optimization that lowers operational costs by 12.4%while ensuring stability constraints.The synergy between adaptive forecasting models and scenario-based stochastic optimization demonstrates superior performance in handling renewable intermittency and demand volatility,validated through grid-scale case studies.Methodological innovations in federated feature extraction and carbon-aware scheduling further enhance scalability for distributed energy systems.These advancements provide actionable insights for grid operators transitioning to low-carbon paradigms,emphasizing computational efficiency and interoperability with legacy infrastructure.
基金the Talent Fund of Beijing Jiaotong University(Grant No.2024XKRC055).
文摘In recent years,railway construction in China has developed vigorously.With continuous improvements in the highspeed railway network,the focus is gradually shifting from large-scale construction to large-scale operations.However,several challenges have emerged within the high-speed railway dispatching and command system,including the heavy workload faced by dispatchers,the difficulty of quantifying subjective expertise,and the need for effective training of professionals.Amid the growing application of artificial intelligence technologies in railway systems,this study leverages Large Language Model(LLM)technology.LLMs bring enhanced intelligence,predictive capabilities,robust memory,and adaptability to diverse real-world scenarios.This study proposes a human-computer interactive intelligent scheduling auxiliary training system built on LLM technology.The system offers capabilities including natural dialogue,knowledge reasoning,and human feedback learning.With broad applicability,the system is suitable for vocational education,guided inquiry,knowledge-based Q&A,and other training scenarios.Validation results demonstrate its effectiveness in auxiliary training,providing substantial support for educators,students,and dispatching personnel in colleges and professional settings.
基金supported in part by the State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources under Grant LAPS24009in part by the Guangdong Basic and Applied Basic Research Foundation under Grant 2021A1515110016in part by the National Natural Science Foundation of China under Grant 52206009.
文摘With the increasing penetration of renewable energy resources in power systems,conventional timescale separated load frequency control(LFC)and economic dispatch may degrade frequency performance and reduce economic efficiency.This paper proposes a novel data-driven adaptive distributed optimal disturbance rejection control(DODRC)method for real-time economic LFC problem in nonlinear power systems.Firstly,a basic DODRC method is proposed by integrating the active disturbance rejection control method and the partial primal–dual algorithm.Then,to deal with the tie-line power flow constraints,the logarithmic barrier function is employed to reconstruct the Lagrange function to obtain the constrained DODRC method.By analyzing the sensitivity of the uncertain parameters of power systems,a data-driven adaptive DODRC method is finally proposed with a neural network.The effectiveness of the proposed method is demonstrated by experimental results using real-time equipment.
文摘In order to address the synergistic optimization of energy efficiency improvement in the waste incineration power plant(WIPP)and renewable energy accommodation,an electricity-hydrogen-waste multi-energy system integrated with phase change material(PCM)thermal storage is proposed.First,a thermal energy management framework is constructed,combining PCM thermal storage with the alkaline electrolyzer(AE)waste heat recovery and the heat pump(HP),while establishing a PCM-driven waste drying system to enhance the efficiency of waste incineration power generation.Next,a flue gas treatment method based on purification-separation-storage coordination is adopted,achieving spatiotemporal decoupling between waste incineration and flue gas treatment.Subsequently,a two-stage optimal dispatching strategy for the multi-energy system is developed:the first stage establishes a dayahead economic dispatch model with the objective of minimizing net system costs,while the second stage introduces model predictive control(MPC)to realize intraday rolling optimization.Finally,The optimal dispatching strategies under different scenarios are obtained using the Gurobi solver,followed by a comparative analysis of the optimized operational outcomes.Simulation results demonstrate that the proposed system optimizes the output and operational states of each unit,simultaneously reducing carbon trading costs while increasing electricity sales revenue.The proposed scheduling strategy demonstrates effective grid peak-shaving functionality,thereby simultaneously improving the system’s economic performance and operational flexibility while providing an innovative technical pathway for municipal solid waste(MSW)resource utilization and low-carbon transformation of energy systems.
基金funded by Guangxi Science and Technology Base and Talent Special Project,grant number GuiKeAD20159077Foundation of Guilin University of Technology,grant number GLUTQD2018001.
文摘The exponential growth in the scale of power systems has led to a significant increase in the complexity of dispatch problem resolution,particularly within multi-area interconnected power grids.This complexity necessitates the employment of distributed solution methodologies,which are not only essential but also highly desirable.In the realm of computational modelling,the multi-area economic dispatch problem(MAED)can be formulated as a linearly constrained separable convex optimization problem.The proximal point algorithm(PPA)is particularly adept at addressing such mathematical constructs effectively.This study introduces parallel(PPPA)and serial(SPPA)variants of the PPA as distributed algorithms,specifically designed for the computational modelling of the MAED.The PPA introduces a quadratic term into the objective function,which,while potentially complicating the iterative updates of the algorithm,serves to dampen oscillations near the optimal solution,thereby enhancing the convergence characteristics.Furthermore,the convergence efficiency of the PPA is significantly influenced by the parameter c.To address this parameter sensitivity,this research draws on trend theory from stock market analysis to propose trend theory-driven distributed PPPA and SPPA,thereby enhancing the robustness of the computational models.The computational models proposed in this study are anticipated to exhibit superior performance in terms of convergence behaviour,stability,and robustness with respect to parameter selection,potentially outperforming existing methods such as the alternating direction method of multipliers(ADMM)and Auxiliary Problem Principle(APP)in the computational simulation of power system dispatch problems.The simulation results demonstrate that the trend theory-based PPPA,SPPA,ADMM and APP exhibit significant robustness to the initial value of parameter c,and show superior convergence characteristics compared to the residual balancing ADMM.
基金supported by the National Natural Science Foundation of China under Grant 62473328by the Open Research Fund of Jiangsu Collaborative Innovation Center for Smart Distribution Network,Nanjing Institute of Technology under No.XTCX202203.
文摘As future ship system,hybrid energy ship system has a wide range of application prospects for solving the serious energy crisis.However,current optimization scheduling works lack the consideration of sea conditions and navigational circumstances.There-fore,this paper aims at establishing a two-stage optimization framework for hybrid energy ship power system.The proposed framework considers multiple optimizations of route,speed planning,and energy management under the constraints of sea conditions during navigation.First,a complex hybrid ship power model consisting of diesel generation system,propulsion system,energy storage system,photovoltaic power generation system,and electric boiler system is established,where sea state information and ship resistance model are considered.With objective optimization functions of cost and greenhouse gas(GHG)emissions,a two-stage optimization framework consisting of route planning,speed scheduling,and energy management is constructed.Wherein the improved A-star algorithm and grey wolf optimization algorithm are introduced to obtain the optimal solutions for route,speed,and energy optimization scheduling.Finally,simulation cases are employed to verify that the proposed two-stage optimization scheduling model can reduce load energy consumption,operating costs,and carbon emissions by 17.8%,17.39%,and 13.04%,respectively,compared with the non-optimal control group.
文摘Taking Huanghua Port Railway Station of the Shuozhou-Huanghua Railway as a demonstration case,an overall solution for the 5G-based intelligent shunting system at heavy haul railway stations was developed to address the operational complexities,inadequacies of outdated equipment,and low efficiency experienced by shunting operators.The system utilizes a 5G communication platform to facilitate automated and intelligent shunting operations at heavy haul railway stations.Advanced technological equipment for intelligent shunting in heavy haul railways was developed,encompassing a big data center,intelligent dispatching and control systems,automated and remote operation of locomotives,intelligent cloud-based video surveillance,intelligent dual-powered electric locomotive,and a customized 5G private network.Technical measures are implemented to reduce operators'labor intensity,decrease the number of on-site personnel,ensure effective safety protection for operators,improve utilization of arrival and departure tracks at heavy haul railway stations,and promote the development of“smart,intelligent,interconnected,and sensing”heavy haul railway stations.
基金Supported by the National Natural Science Foundation of China(No.U24B20156)the National Defense Basic Scientific Research Program of China(No.JCKY2021204B051)the National Laboratory of Space Intelligent Control of China(Nos.HTKJ2023KL502005 and HTKJ2024KL502007)。
文摘A chance-constrained energy dispatch model based on the distributed stochastic model predictive control(DSMPC)approach for an islanded multi-microgrid system is proposed.An ambiguity set considering the inherent uncertainties of renewable energy sources(RESs)is constructed without requiring the full distribution knowledge of the uncertainties.The power balance chance constraint is reformulated within the framework of the distributionally robust optimization(DRO)approach.With the exchange of information and energy flow,each microgrid can achieve its local supply-demand balance.Furthermore,the closed-loop stability and recursive feasibility of the proposed algorithm are proved.The comparative results with other DSMPC methods show that a trade-off between robustness and economy can be achieved.
基金The Science and Technology Project of the State Grid Corporation of China(Research and Demonstration of Loss Reduction Technology Based on Reactive Power Potential Exploration and Excitation of Distributed Photovoltaic-Energy Storage Converters:5400-202333241A-1-1-ZN).
文摘This paper presents a novel approach to economic dispatch in smart grids equipped with diverse energy devices.This method integrates features including photovoltaic(PV)systems,energy storage coupling,varied energy roles,and energy supply and demand dynamics.The systemmodel is developed by considering energy devices as versatile units capable of fulfilling various functionalities and playing multiple roles simultaneously.To strike a balance between optimality and feasibility,renewable energy resources are modeled with considerations for forecasting errors,Gaussian distribution,and penalty factors.Furthermore,this study introduces a distributed event-triggered surplus algorithm designed to address the economic dispatch problem by minimizing production costs.Rooted in surplus theory and finite time projection,the algorithm effectively rectifies network imbalances caused by directed graphs and addresses local inequality constraints.The algorithm greatly reduces the communication burden through event triggering mechanism.Finally,both theoretical proofs and numerical simulations verify the convergence and event-triggered nature of the algorithm.
