检索结果-南通市图书馆

Load Shedding Control Strategy in Power Grid Emergency State Based on Deep Reinforcement Learning

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CSEE Journal of Power and Energy Systems 2022年第4期8卷 1175-1182页

作者： Jian Li Sheng Chen Xinying Wang tianjiao pu China Electric Power Research Institute Haidian DistrictBeijing 100192China

In viewing the power grid for large-scale new energy integration and power electrification of power grid equipment,the impact of power system faults is increased,and the ability of anti-disturbance is decreased,which makes the power system fault clearance more *** this paper,a load shedding control strategy based on artificial intelligence is proposed,this action strategy of load shedding,which is selected by deep reinforcement learning,can support autonomous voltage ***,the power system operation data is used as the basic data to construct the network training dataset,and then a novel reward function for voltage is *** value function,which conforms to the power grid operation characteristics,will act as the reward value for deep reinforcement learning,and the Deep Deterministic Policy Gradient algorithm(DDPG)algorithm,with the continuous action strategy,will be ***,the deep reinforcement learning network is continuously trained,and the load shedding strategy concerning the grid voltage control problem will be obtained in the power system emergency control situation,and this strategy action is input into the Pypower module for simulation verification,thereby realizing the joint drive of data and *** to the numerical simulation analysis,it shows that this method can effectively determine the accurate action selection of load shedding,and improve the stable operational ability of the power system.

关键词： Artificial intelligence deep reinforcement learning load shedding control power system emergency state

Short-term Forecasting of Individual Residential Load Based on Deep Learning and K-means Clustering

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CSEE Journal of Power and Energy Systems 2021年第2期7卷 261-269页

作者： Fujia Han tianjiao pu Maozhen Li Gareth Taylor Artificial Intelligence Application Department China Electric Power Research InstituteBeijing 100192China Department of Electronic and Computer Engineering Brunei University LondonUxbridgeUKUB83PH Brunei Institute of Power Systems Brunei University LondonUxbridgeUKUB83PH

In order to currently motivate a wide range of various interactions between power network operators and electricity customers,residential load forecasting plays an increasingly important role in demand side response(DSR).Due to high volatility and uncertainty of residential load,it is significantly challenging to forecast it ***,this paper presents a short-term individual residential load forecasting method based on a combination of deep learning and k-means clustering,which is capable of effectively extracting the similarity of residential load and performing residential load forecasting accurately at the individual *** first makes full use of k-means clustering to extract similarity among residential load and then employs deep learning to extract complicated patterns of residential *** presented method is tested and validated on a real-life Irish residential load dataset,and the experimental results suggest that it can achieve a much higher prediction accuracy,in comparison with a published benchmark method.

关键词： Deep learning demand side response(DSR) interactions k-means clustering residential load forecasting similarity

Distributionally robust optimization model of active distribution network considering uncertainties of source and load

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Modern Power Systems and Clean Energy 2019年第6期7卷 1585-1595页

作者： Lei DONG Jia LI tianjiao pu Naishi CHEN Power System Research Institute North China Electric Power UniversityBeijing 102206China China Electric Power Research Institute Beijing 100192China

To ensure the safety and reliability of the distribution network and adapt to the uncertain development of renewable energy sources and loads,a two-stage distributionally robust optimization model is proposed for the active distribution network(ADN)optimization problem considering the uncertainties of the source and load in this *** establishing an ambiguity set to capture the uncertainties of the photovoltaic(PV)power,wind power and load,the piecewise-linear function and auxiliary parameters are introduced to help characterize the probability distribution of uncertain *** optimization goal of the model is to minimize the total expected cost under the worst-case distribution in the ambiguity *** first-stage expected cost is obtained based on the predicted value of the uncertainty *** second-stage expected cost is based on the actual value of the uncertainty variable to solve the first-stage *** generalized linear decision rule approximates the two-stage optimization model,and the affine function is introduced to provide a closer approximation to the second-stage optimization ***,the improved IEEE 33-node and IEEE 118-node systems are simulated and analyzed with deterministic methods,stochastic programming,and robust optimization methods to verify the feasibility and superiority of the proposed model and algorithm.

关键词： Active distribution network(ADN) Source-load uncertainty Two-stage distributionally robust optimization Ambiguity set Generalized linear decision rule

Spatio-temporal Convolutional Network Based Power Forecasting of Multiple Wind Farms

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Journal of Modern Power Systems and Clean Energy 2022年第2期10卷 388-398页

作者： Xiaochong Dong Yingyun Sun Ye Li Xinying Wang tianjiao pu the School of Electrical and Electronic Engineering North China Electric Power UniversityBeijing 102206China he China Electric Power Research Institute Beijing 100192China

The rapidly increasing wind power penetration presents new challenges to the operation of power *** the accuracy of wind power forecasting is a possible solution under this *** the power forecasting of mul-tiple wind farms,determining the spatio-temporal correlation of multiple wind farms is critical for improving the forecasting *** paper proposes a spatio-temporal convolutional network(STCN)that utilizes a directed graph convolutional structure.A temporal convolutional network is also adopted to characterize the temporal features of wind *** data from 15 wind farms in Australia are used in the case *** forecasting results show that the proposed model has higher accuracy than the existing *** on the structure of the STCN,asymmetric spatial correlation at different temporal scales can be observed,which shows the effectiveness of the proposed model.

关键词： Deep learning spatio-temporal correlation wind power forecasting graph conventional network(GCN).

Successive MISOCP Algorithm for Islanded Distribution Networks with Soft Open Points

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

CSEE Journal of Power and Energy Systems 2023年第1期9卷 209-220页

作者： Tao Zhang Yunfei Mu Hongjie Jia Xinying Wang tianjiao pu Key Laboratory of Smart Grid of Ministry of Education Key Laboratory of Smart Energy&Information Technology of Tianjin MunicipalityTianjin UniversityTianjin 300072China e China Electric Power Research Institute Beijing 100192China

An optimal operation scheme is of great significance in islanded distribution networks to restore critical loads and has recently attracted considerable *** this paper,an optimal power flow(OPF)model for islanded distribution networks equipped with soft open points(SOPs)is *** in the grid-connected mode,the adequacy of local power generation in distribution networks is critical for islanded *** proposed approach utilizes the power output of local distributed generations(DGs)and the benefits of reactive power compensation provided by SOPs to allow maximum *** exploit the available resources,an optimal secondary droop control strategy is introduced for the islanded distribution networks,thereby minimizing load *** formulated OPF problem is essentially a mixed-integer nonlinear programming(MINLP)*** guarantee the computation efficiency and accuracy.A successive mixed-integer second-order cone programming(SMISOCP)algorithm is proposed for handling the nonlinear islanded power flow *** case studies,incorporating a modified IEEE 33-bus system and IEEE 123-bus system,are performed to test the effectiveness of the proposed approach.

关键词： Distribution network islanded operation second-order cone programming soft open points successive approximation

THz Wave Detection of Gap Defects Based on a Convolutional Neural Network Improved by a Residual Shrinkage Network

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CSEE Journal of Power and Energy Systems 2023年第3期9卷 1078-1089页

作者： Zhonghao Zhang Guozheng Peng Yuanpeng Tan tianjiao pu Liming Wang China Electric Power Research Institute Tsinghua Shenzhen International Gradual School Shenzhen 518055China

Internal air gap is a serious type of defect in the insulation equipment,which threatens the safe operation of the power *** order to diagnose the position and thickness of the internal air gap,this paper proposes a terahertz wave detection method based on wavelet analysis and a CNN(convolution neural network)*** to the time-frequency characteristics of the wavelet cluster,the calculation method of air gap depth is *** determine the thickness of the internal air gap,the performances of several classification methods,such as waveform feature analysis,Bayes,MLP(Multi-layer Perceptron),SVM(Support Vector Machine)and CNN are *** results show that the CNN modified by a residual shrinkage network and SVM(CNN-RSN-SVM)has the best *** adjusting the parameters,the classification accuracy of the CNN-RSN-SVM model can be elevated to 98.91%.Furthermore,the 3D imaging method of air gap defect based on wavelet analysis and CNNRSN-SVM classification model is formed.

关键词： Convolutional neural network defect detection insulation equipment terahertz wave

Review of Learning-assisted Power System Optimization

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CSEE Journal of Power and Energy Systems 2021年第2期7卷 221-231页

作者： Guangchun Ruan Haiwang Zhong Guanglun Zhang Yiliu He Xuan Wang tianjiao pu Department of Electrical Engineering Tsinghua UniversityBeijing 100084China China Electric Power Research Institute Beijing 100192China

With dramatic breakthroughs in recent years,machine learning is showing great potential to upgrade the toolbox for power system *** the strength and limitation of machine learning approaches is crucial to decide when and how to deploy them to boost the optimization *** paper pays special attention to the coordination between machine learning approaches and optimization models,and carefully evaluates how such data-driven analysis may improve the rule-based *** typical references are selected and categorized into four groups:the boundary parameter improvement,the optimization option selection,the surrogate model,and the hybrid *** taxonomy provides a novel perspective to elaborate the latest research progress and *** further compare the design patterns of different categories,and discuss several key challenges and opportunities as *** integration between machine learning approaches and optimization models is expected to become the most promising technical trend.

关键词： Artificial intelligence data-driven deep learning machine learning neural network smart grid

An Edge Visual Incremental Perception Framework Based on Deep Semi-supervised Learning for Monitoring Power Transmission Lines

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CSEE Journal of Power and Energy Systems 2023年第2期9卷 759-768页

作者： Jun Zhang Jiye Wang Rui Song Guozheng Peng tianjiao pu Shuhua Zhang College of Electrical and Information Engineering Hunan UniversityChangsha 410082China China Electric Power Research Institute Beijing 100192China

In recent years,several efforts have been made to develop power transmission line abnormal target detection models based on edge ***,updates to these models rely on participation of the cloud,which means that network resource shortages can lead to update failures,followed by unsatisfactory recognition and detection performance in practical *** address this problem,this article proposes an edge visual incremental perception framework,based on deep semisupervised learning,for monitoring power transmission *** generation of the initial model using a small amount of labeled data,models trained using this framework can update themselves based on unlabeled data.A teacher-student joint training strategy,a data augmentation strategy,and a model updating strategy are also designed and adopted to improve the performance of the models trained with this *** proposed framework is then examined with various transmission line datasets with 1%,2%,5%,and 10%labeled *** performance enhancement is thus confirmed against traditional supervised learning *** the 10%labeled data training set,the recognition accuracy of the model is improved to exceed 80%,meeting the practical needs of power system operation,and thus clearly validating the effectiveness of the framework.

关键词： Edge intelligence semi-supervised learning transmission lines monitoring visual incremental perception

Hierarchical Game Approach for Optimization of Regional Integrated Energy System Clusters Considering Bounded Rationality

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CSEE Journal of Power and Energy Systems 2024年第1期10卷 302-313页

作者： Lei Dong Mengting Li Junjie Hu Sheng Chen Tao Zhang Xinying Wang tianjiao pu school of Electrical and Electronic Engineering North China Electric Power UniversityBeijing 102206China Department of Artificial Intelligence Application Research China Electric Power Research InstituteBeijing 100192China Key Laboratory of Smart Grid of Ministry of Education Key Laboratory of Smart Energy&Information Technology of Tianjin MunicipalityTianjin UniversityTianjin 300072China

Regional integrated energy system(RIES)cluster,i.e.,multi-source integration and multi-region coordination,is an effective approach for increasing energy utilization *** hierarchical architecture and limited information sharing of RIES cluster make it difficult for traditional game theory to accurately describe their game ***,a hierarchical game approach considering bounded rationality is proposed in this paper to balance the interests of optimizing RIES cluster under privacy protection.A Stackelberg game with the cluster operator(CO)as the leader and multiple RIES as followers is developed to simultaneously optimize leader benefit and RIES utilization ***,a slight altruistic function is introduced to simulate the game behavior of each RIES agent on whether to cooperate or *** introducing an evolutionary game based on bounded rationality in the lower layer,the flaw of the assumption that participants are completely rational can be ***,for autonomous optimal dispatching,each RIES is treated as a prosumer,fexibly switching its market participation role to achieve cluster coordination *** studies on a RIES cluster verify effectiveness of the proposed approach.

关键词： Bounded rationality coordinated optimization evolutionary game regionaIilntegrated energysystem stackelberg game