In this paper,a model free volt/var control(VVC)algorithm is developed by using deep reinforcement learning(DRL).We transform the VVC problem of distribution networks into the network framework of PPO algorithm,in ord...In this paper,a model free volt/var control(VVC)algorithm is developed by using deep reinforcement learning(DRL).We transform the VVC problem of distribution networks into the network framework of PPO algorithm,in order to avoid directly solving a large-scale nonlinear optimization problem.We select photovoltaic inverters as agents to adjust system voltage in a distribution network,taking the reactive power output of inverters as action variables.An appropriate reward function is designed to guide the interaction between photovoltaic inverters and the distribution network environment.OPENDSS is used to output system node voltage and network loss.This method realizes the goal of optimal VVC in distribution network.The IEEE 13-bus three phase unbalanced distribution system is used to verify the effectiveness of the proposed algorithm.Simulation results demonstrate that the proposed method has excellent performance in voltage and reactive power regulation of a distribution network.展开更多
In volt/var control(VVC)for active distribution networks,it is essential to integrate traditional voltage regulation devices with modern smart photovoltaic inverters to prevent voltage violations.However,model-based m...In volt/var control(VVC)for active distribution networks,it is essential to integrate traditional voltage regulation devices with modern smart photovoltaic inverters to prevent voltage violations.However,model-based multi-device VVC methods rely on accurate system models for decision-making,which can be challenging due to the extensive modeling workload.To tackle the complexities of multi-device cooperation in VVC,this paper proposes a two-timescale VVC method based on reinforcement learning with hybrid action space,termed the hybrid action representation twin delayed deep deterministic policy gradient(HAR-TD3)method.This method simultaneously manages traditional discrete voltage regulation devices,which operate on a slower timescale,and smart continuous voltage regulation devices,which function on a faster timescale.To enable effective collaboration between the different action spaces of these devices,we propose a variational auto-encoder based hybrid action reconstruction network.This network captures the interdependencies of hybrid actions by embedding both discrete and continuous actions into the latent representation space and subsequently decoding them for action reconstruction.The proposed method is validated on IEEE 33-bus,69-bus,and 123-bus distribution networks.Numerical results indicate that the proposed method successfully coordinates discrete and continuous voltage regulation devices,achieving fewer voltage violations compared with stateof-the-art reinforcement learning methods.展开更多
For active distribution networks(ADNs)integrated with massive inverter-based energy resources,it is impractical to maintain the accurate model and deploy measurements at all nodes due to the large-scale of ADNs.Thus,c...For active distribution networks(ADNs)integrated with massive inverter-based energy resources,it is impractical to maintain the accurate model and deploy measurements at all nodes due to the large-scale of ADNs.Thus,current models of ADNs usually involve significant errors or even unknown occurances.Moreover,ADNs are usually partially observable since only a few measurements are available at pilot nodes or nodes with significant users.To provide a practical Volt/Var control(VVC)strategy for such networks,a data-driven VVC method is proposed in this paper.First,the system response policy,approximating the relationship between the control variables and states of monitoring nodes,is estimated by a recursive regression closed-form solution.Then,based on real-time measurements and the newly updated system response policy,a VVC strategy with convergence guarantee is realized.Since the recursive regression solution is embedded in the control stage,a data-driven closedloop VVC framework is established.The effectiveness of the proposed method is validated in an unbalanced distribution system considering nonlinear loads,where not only the rapid and self-adaptive voltage regulation is realized,but also systemwide optimization is achieved.展开更多
When urban distribution systems are gradually modernized,the overhead lines are replaced by underground cables,whose shunt admittances can not be ignored.Traditional power flow(PF)model withπequivalent circuit shows ...When urban distribution systems are gradually modernized,the overhead lines are replaced by underground cables,whose shunt admittances can not be ignored.Traditional power flow(PF)model withπequivalent circuit shows non-convexity and long computing time,and most recently proposed linear PF models assume zero shunt elements.All of them are not suitable for fast calculation and optimization problems of modern distribution systems with non-negligible line shunts.Therefore,this paper proposes a linearized branch flow model considering line shunt(LBFS).The strength of LBFS lies in maintaining the linear structure and the convex nature after appropriately modeling theπequivalent circuit for network equipment like transformers.Simulation results show that the calculation accuracy in nodal voltage and branch current magnitudes is improved by considering shunt admittances.We show the application scope of LBFS by controlling the network voltages through a two-stage stochastic Volt/VAr control(VVC)problem with the uncertain active power output from renewable energy sources(RESs).Since LBFS results in a linear VVC program,the global solution is guaranteed.Case study exhibits that VVC framework can optimally dispatch the discrete control devices,viz.substation transformers and shunt capacitors,and also optimize the decision rules for real-time reactive power control of RES.Moreover,the computing efficiency is significantly improved compared with that of traditional VVC methods.展开更多
Photovoltaic(PV)inverter-based volt/var control(VVC)is highly promising to tackle the emerging voltage regulation challenges brought by increasing PV penetration.However,PV inverter operational reliability has arisen ...Photovoltaic(PV)inverter-based volt/var control(VVC)is highly promising to tackle the emerging voltage regulation challenges brought by increasing PV penetration.However,PV inverter operational reliability has arisen as a critical concern for practical VVC implementation.This paper proposes a new PV inverter based VVC optimization model and a Pareto front analysis method for maintaining a satisfactory inverter lifetime.First,reliability of the vulnerable DC-link capacitor inside a PV inverter is analyzed,and long-term VVC impact on inverter operational reliability is identified.Second,a multi-objective PV inverter based VVC optimization model is proposed for minimizing both inverter apparent power output and network power loss with a weighting factor.Third,a Pareto front analysis method is developed to visualize the impact of the weighting factor on VVC performance and inverter reliability,thus determining the effective weighting factor to reduce network power loss with expected inverter lifetime.Effectiveness of the proposed VVC optimization model and Pareto front analysis method are verified in a case study.展开更多
The major challenge to increase the decentralized generation share in distribution grids is the maintenance of the voltage within the limits. The inductive power injection is widely used as a remedial measure. The mai...The major challenge to increase the decentralized generation share in distribution grids is the maintenance of the voltage within the limits. The inductive power injection is widely used as a remedial measure. The main aim of this paper is to study the effect of the reactive power injection (by what-ever means) on radial grid structures and their impact on the voltage of the higher voltage-level grids. Various studies have shown that, in addition to the major local effect on the voltage at the injection point, the injection of the reactive power on a feeder has a global effect, which cannot be neglected. The reactive power flow and the voltage on the higher voltage level grid are significantly affected. In addition, a random effect is introduced by the DGs which are connected through inverters (using wind or PVs). Although their operation is in accordance with the grid code, a volatile reactive power flow circulates on the grid. Finally, this study proposes the implementation of the “Volt/var secondary control” interaction chain in order to increase the distributed generation share at every distribution voltage level, be it medium or low voltage, and at the same time to guarantee a stable operation of the power grid. Features of Volt/var secondary control loops ensure a resilient behavior of the whole chain.展开更多
The high proportion of renewable energy integration and the dynamic changes in grid topology necessitate the enhancement of voltage/var control(VVC)to manage voltage fluctuations more rapidly.Traditional model-based c...The high proportion of renewable energy integration and the dynamic changes in grid topology necessitate the enhancement of voltage/var control(VVC)to manage voltage fluctuations more rapidly.Traditional model-based control algorithms are becoming increasingly incompetent for VVC due to their high model dependence and slow online computation speed.To alleviate these issues,this paper introduces a graph attention network(GAT)based deep reinforcement learning for VVC of topologically variable power system.Firstly,combining the physical information of the actual power grid,a physics-informed GAT is proposed and embedded into the proximal policy optimization(PPO)algorithm.The GAT-PPO algorithm can capture topological and spatial correlations among the node features to tackle topology changes.To address the slow training,the Relief F-S algorithm identifies critical state variables,significantly reducing the dimensionality of state space.Then,the training samples retained in the experience buffer are designed to mitigate the sparse reward issue.Finally,the validation on the modified IEEE 39-bus system and an actual power grid demonstrates superior performance of the proposed algorithm compared with state-of-the-art algorithms,including PPO algorithm and twin delayed deep deterministic policy gradient(TD3)algorithm.The proposed algorithm exhibits enhanced convergence during training,faster solution speed,and improved VVC performance,even in scenarios involving grid topology changes and increased renewable energy integration.Meanwhile,in the adopted cases,the network loss is reduced by 6.9%,10.8%,and 7.7%,respectively,demonstrating favorable economic outcomes.展开更多
An active disturbance rejection controller （ADRC） is developed for load frequency control （LFC） and voltage regulation respectively in a power system. For LFC, the ADRC is constructed on a three-area interconnecte...An active disturbance rejection controller （ADRC） is developed for load frequency control （LFC） and voltage regulation respectively in a power system. For LFC, the ADRC is constructed on a three-area interconnected power system. The control goal is to maintain the frequency at nominal value （60Hz in North America） and keep tie-line power flow at scheduled value. For voltage regulation, the ADRC is applied to a static var compensator （SVC） as a supplementary controller. It is utilized to maintain the voltages at nearby buses within the ANSI C84.1 limits （or ＋5% tolerance）. Particularly, an alternative ADRC with smaller controller gains than classic ADRC is originally designed on the SVC system. From power generation and transmission to its distribution, both voltage and frequency regulating systems are subject to large and small disturbances caused by sudden load changes, transmission faults, and equipment loss/malfunction etc. The simulation results and theoretical analyses demonstrate the effectiveness of the ADRCs in compensating the disturbances and achieving the control goals.展开更多
This paper presents a design method of a variable structure svc controller. Its control principle is easy to realize, and it is not related to the parameters of the power network and operation conditions. The result o...This paper presents a design method of a variable structure svc controller. Its control principle is easy to realize, and it is not related to the parameters of the power network and operation conditions. The result of computer simulation shows that the proposed controller can improve the system's damping performance effctively.展开更多
基金supported by the Science and Technology Project of State Grid Zhejiang Electric Power Co.,Ltd.under Grant B311JY21000A。
文摘In this paper,a model free volt/var control(VVC)algorithm is developed by using deep reinforcement learning(DRL).We transform the VVC problem of distribution networks into the network framework of PPO algorithm,in order to avoid directly solving a large-scale nonlinear optimization problem.We select photovoltaic inverters as agents to adjust system voltage in a distribution network,taking the reactive power output of inverters as action variables.An appropriate reward function is designed to guide the interaction between photovoltaic inverters and the distribution network environment.OPENDSS is used to output system node voltage and network loss.This method realizes the goal of optimal VVC in distribution network.The IEEE 13-bus three phase unbalanced distribution system is used to verify the effectiveness of the proposed algorithm.Simulation results demonstrate that the proposed method has excellent performance in voltage and reactive power regulation of a distribution network.
基金supported in part by the National Science and Technology Major Project(No.2022ZD0116900)the National Natural Science Foundation of China(No.52277118)the Natural Science Foundation of Tianjin(No.22JCZDJC00660).
文摘In volt/var control(VVC)for active distribution networks,it is essential to integrate traditional voltage regulation devices with modern smart photovoltaic inverters to prevent voltage violations.However,model-based multi-device VVC methods rely on accurate system models for decision-making,which can be challenging due to the extensive modeling workload.To tackle the complexities of multi-device cooperation in VVC,this paper proposes a two-timescale VVC method based on reinforcement learning with hybrid action space,termed the hybrid action representation twin delayed deep deterministic policy gradient(HAR-TD3)method.This method simultaneously manages traditional discrete voltage regulation devices,which operate on a slower timescale,and smart continuous voltage regulation devices,which function on a faster timescale.To enable effective collaboration between the different action spaces of these devices,we propose a variational auto-encoder based hybrid action reconstruction network.This network captures the interdependencies of hybrid actions by embedding both discrete and continuous actions into the latent representation space and subsequently decoding them for action reconstruction.The proposed method is validated on IEEE 33-bus,69-bus,and 123-bus distribution networks.Numerical results indicate that the proposed method successfully coordinates discrete and continuous voltage regulation devices,achieving fewer voltage violations compared with stateof-the-art reinforcement learning methods.
基金supported by the Research Project of China Southern Power Grid Corporation:The demonstration and application of the virtual power plant intelligent operation and management platform with source-grid coordination,No.GDKJXM20185069 (032000KK 52180069)。
文摘For active distribution networks(ADNs)integrated with massive inverter-based energy resources,it is impractical to maintain the accurate model and deploy measurements at all nodes due to the large-scale of ADNs.Thus,current models of ADNs usually involve significant errors or even unknown occurances.Moreover,ADNs are usually partially observable since only a few measurements are available at pilot nodes or nodes with significant users.To provide a practical Volt/Var control(VVC)strategy for such networks,a data-driven VVC method is proposed in this paper.First,the system response policy,approximating the relationship between the control variables and states of monitoring nodes,is estimated by a recursive regression closed-form solution.Then,based on real-time measurements and the newly updated system response policy,a VVC strategy with convergence guarantee is realized.Since the recursive regression solution is embedded in the control stage,a data-driven closedloop VVC framework is established.The effectiveness of the proposed method is validated in an unbalanced distribution system considering nonlinear loads,where not only the rapid and self-adaptive voltage regulation is realized,but also systemwide optimization is achieved.
基金supported in part by the National Natural Science Foundation of China(No.51977115)。
文摘When urban distribution systems are gradually modernized,the overhead lines are replaced by underground cables,whose shunt admittances can not be ignored.Traditional power flow(PF)model withπequivalent circuit shows non-convexity and long computing time,and most recently proposed linear PF models assume zero shunt elements.All of them are not suitable for fast calculation and optimization problems of modern distribution systems with non-negligible line shunts.Therefore,this paper proposes a linearized branch flow model considering line shunt(LBFS).The strength of LBFS lies in maintaining the linear structure and the convex nature after appropriately modeling theπequivalent circuit for network equipment like transformers.Simulation results show that the calculation accuracy in nodal voltage and branch current magnitudes is improved by considering shunt admittances.We show the application scope of LBFS by controlling the network voltages through a two-stage stochastic Volt/VAr control(VVC)problem with the uncertain active power output from renewable energy sources(RESs).Since LBFS results in a linear VVC program,the global solution is guaranteed.Case study exhibits that VVC framework can optimally dispatch the discrete control devices,viz.substation transformers and shunt capacitors,and also optimize the decision rules for real-time reactive power control of RES.Moreover,the computing efficiency is significantly improved compared with that of traditional VVC methods.
基金This work was supported in part by NTU Grant No.021542-00001in part by Australian Government Research Training Program Scholarship。
文摘Photovoltaic(PV)inverter-based volt/var control(VVC)is highly promising to tackle the emerging voltage regulation challenges brought by increasing PV penetration.However,PV inverter operational reliability has arisen as a critical concern for practical VVC implementation.This paper proposes a new PV inverter based VVC optimization model and a Pareto front analysis method for maintaining a satisfactory inverter lifetime.First,reliability of the vulnerable DC-link capacitor inside a PV inverter is analyzed,and long-term VVC impact on inverter operational reliability is identified.Second,a multi-objective PV inverter based VVC optimization model is proposed for minimizing both inverter apparent power output and network power loss with a weighting factor.Third,a Pareto front analysis method is developed to visualize the impact of the weighting factor on VVC performance and inverter reliability,thus determining the effective weighting factor to reduce network power loss with expected inverter lifetime.Effectiveness of the proposed VVC optimization model and Pareto front analysis method are verified in a case study.
文摘The major challenge to increase the decentralized generation share in distribution grids is the maintenance of the voltage within the limits. The inductive power injection is widely used as a remedial measure. The main aim of this paper is to study the effect of the reactive power injection (by what-ever means) on radial grid structures and their impact on the voltage of the higher voltage-level grids. Various studies have shown that, in addition to the major local effect on the voltage at the injection point, the injection of the reactive power on a feeder has a global effect, which cannot be neglected. The reactive power flow and the voltage on the higher voltage level grid are significantly affected. In addition, a random effect is introduced by the DGs which are connected through inverters (using wind or PVs). Although their operation is in accordance with the grid code, a volatile reactive power flow circulates on the grid. Finally, this study proposes the implementation of the “Volt/var secondary control” interaction chain in order to increase the distributed generation share at every distribution voltage level, be it medium or low voltage, and at the same time to guarantee a stable operation of the power grid. Features of Volt/var secondary control loops ensure a resilient behavior of the whole chain.
基金supported by China Southern Power Grid Co.,Ltd.Yunnan Electric Power Dispatching Control Center(Cyber-physical-based“Source-network-load-storage”Coordinated Dispatch and Control Technologies and Application System Development,sub-project YNKJXM20222463)。
文摘The high proportion of renewable energy integration and the dynamic changes in grid topology necessitate the enhancement of voltage/var control(VVC)to manage voltage fluctuations more rapidly.Traditional model-based control algorithms are becoming increasingly incompetent for VVC due to their high model dependence and slow online computation speed.To alleviate these issues,this paper introduces a graph attention network(GAT)based deep reinforcement learning for VVC of topologically variable power system.Firstly,combining the physical information of the actual power grid,a physics-informed GAT is proposed and embedded into the proximal policy optimization(PPO)algorithm.The GAT-PPO algorithm can capture topological and spatial correlations among the node features to tackle topology changes.To address the slow training,the Relief F-S algorithm identifies critical state variables,significantly reducing the dimensionality of state space.Then,the training samples retained in the experience buffer are designed to mitigate the sparse reward issue.Finally,the validation on the modified IEEE 39-bus system and an actual power grid demonstrates superior performance of the proposed algorithm compared with state-of-the-art algorithms,including PPO algorithm and twin delayed deep deterministic policy gradient(TD3)algorithm.The proposed algorithm exhibits enhanced convergence during training,faster solution speed,and improved VVC performance,even in scenarios involving grid topology changes and increased renewable energy integration.Meanwhile,in the adopted cases,the network loss is reduced by 6.9%,10.8%,and 7.7%,respectively,demonstrating favorable economic outcomes.
文摘An active disturbance rejection controller （ADRC） is developed for load frequency control （LFC） and voltage regulation respectively in a power system. For LFC, the ADRC is constructed on a three-area interconnected power system. The control goal is to maintain the frequency at nominal value （60Hz in North America） and keep tie-line power flow at scheduled value. For voltage regulation, the ADRC is applied to a static var compensator （SVC） as a supplementary controller. It is utilized to maintain the voltages at nearby buses within the ANSI C84.1 limits （or ＋5% tolerance）. Particularly, an alternative ADRC with smaller controller gains than classic ADRC is originally designed on the SVC system. From power generation and transmission to its distribution, both voltage and frequency regulating systems are subject to large and small disturbances caused by sudden load changes, transmission faults, and equipment loss/malfunction etc. The simulation results and theoretical analyses demonstrate the effectiveness of the ADRCs in compensating the disturbances and achieving the control goals.
文摘This paper presents a design method of a variable structure svc controller. Its control principle is easy to realize, and it is not related to the parameters of the power network and operation conditions. The result of computer simulation shows that the proposed controller can improve the system's damping performance effctively.
