查看更多>>摘要：With integration of large-scale renewable energy,new controllable devices,and required reinforcement of power grids,modern power systems have typical characteristics such as uncertainty,vulnerability and openness,which makes operation and control of power grids face severe security challenges.Application of artificial intelligence(AI)technologies represented by machine learning in power grid regulation is limited by reliability,interpretability and generalization ability of complex modeling.Mode of hybrid-augmented intelligence(HAI)based on human-machine collaboration(HMC)is a pivotal direction for future development of AI technology in this field.Based on characteristics of applications in power grid regulation,this paper discusses system architecture and key technologies of human-machine hybrid-augmented intelligence(HHI)system for large-scale power grid dispatching and control(PGDC).First,theory and application scenarios of HHI are introduced and analyzed;then physical and functional architectures of HHI system and human-machine collaborative regulation process are proposed.Key technologies are discussed to achieve a thorough integration of human/machine intelligence.Finally,state-of-the-art and future development of HHI in power grid regulation are summarized,aiming to efficiently improve the intelligent level of power grid regulation in a human-machine interactive and collaborative way.
查看更多>>摘要：In artificial intelligence(AI)based-complex power system management and control technology,one of the ur-gent tasks is to evaluate AI intelligence and invent a way of autonomous intelligence evolution.However,there is,cur-rently,nearly no standard technical framework for objective and quantitative intelligence evaluation.In this article,based on a parallel system framework,a method is established to objectively and quantitatively assess the intelligence level of an AI agent for active power corrective control of modern power systems,by resorting to human intelligence evaluation theories.On this basis,this article puts forward an AI self-evolution method based on intelligence assessment through embedding a quantitative intelligence assessment method into automated reinforcement learning(AutoRL)systems.A parallel system based quantitative assessment and self-evolution(PLASE)system for power grid corrective control AI is thereby constructed,taking Bayesian Optimization as the measure of AI evolution to fulfill autonomous evolution of AI under guidance of their intelligence assessment results.Experiment results exemplified in the power grid corrective control AI agent show the PLASE system can reliably and quantitatively assess the intelligence level of the power grid corrective control agent,and it could promote evolution of the power grid corrective control agent under guidance of intelligence assessment results,effectively,as well as intuitively improving its intelligence level through self-evolution.
查看更多>>摘要：The complexity and uncertainty in power systems cause great challenges to controlling power grids.As a popu-lar data-driven technique,deep reinforcement learning(DRL)attracts attention in the control of power grids.However,DRL has some inherent drawbacks in terms of data efficiency and explainability.This paper presents a novel hierarchical task planning(HTP)approach,bridging planning and DRL,to the task of power line flow regulation.First,we introduce a three-level task hierarchy to model the task and model the sequence of task units on each level as a task planning-Markov decision processes(TP-MDPs).Second,we model the task as a sequential decision-making problem and introduce a higher planner and a lower planner in HTP to handle different levels of task units.In addition,we introduce a two-layer knowledge graph that can update dynamically during the planning procedure to assist HTP.Experimental results conducted on the IEEE 118-bus and IEEE 300-bus systems demonstrate our HTP approach outperforms proximal policy optimization,a state-of-the-art deep reinforcement learning(DRL)approach,improving efficiency by 26.16％and 6.86％on both systems.
查看更多>>摘要：Increase in permeability of renewable energy sources(RESs)leads to the prominent problem of voltage stability in power system,so it is urgent to have a system strength eval-uation method with both accuracy and practicability to control its access scale within a reasonable range.Therefore,a hybrid intelligence enhancement method is proposed by combining the advantages of mechanism method and data driven method.First,calculation of critical short circuit ratio(CSCR)is set as the direc-tion of intelligent enhancement by taking the multiple renewable energy station short circuit ratio as the quantitative indicator.Then,the construction process of CSCR dataset is proposed,and a batch simulation program of samples is developed accordingly,which provides a data basis for subsequent research.Finally,a multi-task learning model based on progressive layered extraction is used to simultaneously predict CSCR of each RESs connection point,which significantly reduces evaluation error caused by weak links.Predictive performance and anti-noise performance of the proposed method are verified on the CEPRI-FS-102 bus system,which provides strong technical support for real-time monitoring of system strength.
查看更多>>摘要：Transition towards carbon-neutral power systems has necessitated optimization of power dispatch in active dis-tribution networks(ADNs)to facilitate integration of distributed renewable generation.Due to unavailability of network topology and line impedance in many distribution networks,physical model-based methods may not be applicable to their operations.To tackle this challenge,some studies have proposed constraint learning,which replicates physical models by training a neural network to evaluate feasibility of a decision(i.e.,whether a decision satisfies all critical constraints or not).To ensure accuracy of this trained neural network,training set should contain sufficient feasible and infeasible samples.However,since ADNs are mostly operated in a normal status,only very few historical samples are infeasible.Thus,the historical dataset is highly imbalanced,which poses a significant obstacle to neural network training.To address this issue,we propose an enhanced constraint learning method.First,it leverages constraint learning to train a neural network as surrogate of ADN's model.Then,it introduces Synthetic Minority Oversampling Technique to generate infeasible samples to mitigate imbalance of histori-cal dataset.By incorporating historical and synthetic samples into the training set,we can significantly improve accuracy of neural network.Furthermore,we establish a trust region to constrain and thereafter enhance reliability of the solution.Simulations confirm the benefits of the proposed method in achieving desirable optimality and feasibility while maintaining low computational complexity.
查看更多>>摘要：Generator tripping scheme(GTS)is the most com-monly used scheme to prevent power systems from losing safety and stability.Usually,GTS is composed of offline predetermi-nation and real-time scenario match.However,it is extremely time-consuming and labor-intensive for manual predetermination for a large-scale modern power system.To improve efficiency of predetermination,this paper proposes a framework of knowl-edge fusion-based deep reinforcement learning(KF-DRL)for intelligent predetermination of GTS.First,the Markov Decision Process(MDP)for GTS problem is formulated based on transient instability events.Then,linear action space is developed to reduce dimensionality of action space for multiple controllable generators.Especially,KF-DRL leverages domain knowledge about GTS to mask invalid actions during the decision-making process.This can enhance the efficiency and learning process.Moreover,the graph convolutional network(GCN)is introduced to the policy network for enhanced learning ability.Numerical simulation results obtained on New England power system demonstrate superiority of the proposed KF-DRL framework for GTS over the purely data-driven DRL method.
查看更多>>摘要：Traditional assessment indexes could not fully de-scribe offshore wind resources,for the meteorological properties of offshore are more complex than onshore.As a result,the uncertainty of offshore wind power projects would be increased and final economic benefits would be affected.Therefore,a study on offshore wind resource assessment is carried out,including three processes of"studying data sources,conducting multi-dimensional indexes system and proposing an offshore wind resource assessment method based on analytic hierarchy process(AHP)".First,measured wind data and two kinds of reanalysis data are used to analyze the characteristics and reliability of data sources.Second,indexes such as effective wind speed occurrence,affluent level occurrence,coefficient of variation,neutral state occurrence have been proposed to depict availability,richness,and stability of offshore wind resources,respectively.Combined with existing parameters(wind power density,dominant wind direction occurrence,water depth,distance to coast),a multi-dimensional indexes system has been built and on this basis,an offshore wind energy potential assessment method has been proposed.Furthermore,the proposed method is verified by the annual energy production of five offshore wind turbines and practical operating data of four offshore wind farms in China.This study also compares the ranking results of the AHP model to two multi-criteria decision making(MCDM)models including weighted aggregated sum product assessment(WASPAS)and multi-attribute ideal real comparative analysis(MAIRCA).Results show the proposed method gains well in practical engineering applications,where the economic score values have been considered based on the offshore reasonable utilization hours of the whole life cycle in China.
查看更多>>摘要：A transformer is an essential but expensive power delivery equipment for a distribution utility.In many distribution utilities worldwide,a sizable percentage of transformers are near the end of their designed life.At the same time,distribution utilities are adopting smart inverter-based distributed solar photovoltaic(SPV)systems to maximize renewable generation.The central objective of this paper is to propose a methodology to quantify the effect of smart inverter-based distributed SPV systems on the aging of distribution transformers.The proposed method is first tested on a modified IEEE-123 node distribution feeder.After that,the procedure is applied to a practical distribution system,i.e.,the Indian Institute of Technology(IIT)Roorkee campus,India.The transformer aging models,alongside advanced control functionalities of grid-tied smart inverter-based SPV systems,are implemented in MATLAB.The open-source simulation tool(OpenDSS)is used to model distribution net-works.To analyze effectiveness of various inverter functionalities,time-series simulations are performed using exponential load models,considering daily load curves from multiple seasons,load types,current harmonics,etc.Findings show replacing a traditional inverter with a smart inverter-based SPV system can enable local reactive power generation and may extend the life of a distribution transformer.Simulation results demonstrate,simply by incorporating smart inverter-based SPV systems,transformer aging is reduced by 15％to 22％in comparison to SPV systems operating with traditional inverters.
查看更多>>摘要：Market participants can only bid with lagged in-formation disclosure under the existing market mechanism,which can lead to information asymmetry and irrational market behavior,thus influencing market efficiency.To promote rational bidding behavior of market participants and improve market efficiency,a novel electricity market mechanism based on cloud-edge collaboration is proposed in this paper.Critical market information,called residual demand curve,is published to market participants in real-time on the cloud side,while participants on the edge side are allowed to adjust their bids according to the information disclosure prior to closure gate.The proposed mech-anism can encourage rational bids in an incentive-compatible way through the process of dynamic equilibrium while protecting participants'privacy.This paper further formulates the math-ematical model of market equilibrium to simulate the process of each market participant's strategic bidding behavior towards equilibrium.A case study based on the IEEE 30-bus system shows the proposed market mechanism can effectively guide bidding behavior of market participants,while condensing exchanged information and protecting privacy of participants.
查看更多>>摘要：This paper presents a robust interval economic dispatch(RIED)model for power systems with large-scale wind power integration.Differing from existing interval optimization(IO)approaches that merely rely on the upper and lower boundaries of random variables,the distribution information retained in the historical data is introduced to the IO method in this paper.Based on the available probability distribution function(PDF),wind power curtailment and load shedding are quantified as the operational risk and incorporated into the decision-making process.In this model,we need not rely on the forecasted value of wind power,which is randomly fluctuating and quite unpredictable.Furthermore,when the PDFs of wind power are taken into account,the resulting dispatch solution makes a good tradeoff between the generation cost and the operational risk.Finally,the RIED model yields an optimal dispatch solution for thermal units and the allowable intervals of wind power for the wind farms,which efficiently mitigates the uncertainty in wind power generation and provides more practical suggestions for system operators.Simulation studies are conducted on a modified IEEE-118 bus system and the results verify the effectiveness of the proposed RIED model.
维普
万方数据