The complexity and uncertainty in power systems cause great challenges to controlling power grids.As a popular data-driven technique,deep reinforcement learning(DRL)attracts attention in the control of power grids.How...The complexity and uncertainty in power systems cause great challenges to controlling power grids.As a popular 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 threelevel 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.展开更多
The mode-based damping torque analysis(M-DTA)method for studying the effect of an external controller on power system low-frequency oscillations is proposed in this paper.First,based on the interconnection model betwe...The mode-based damping torque analysis(M-DTA)method for studying the effect of an external controller on power system low-frequency oscillations is proposed in this paper.First,based on the interconnection model between the system and the controller in the frequency domain,the oscillation loop corresponding to the electromechanical oscillation mode is built,and then the mode-based damping torque of the controller can be calculated.Then,the application of the M-DTA method in the power system is illustrated.The derivation shows that in the single-machine infinite-bus power system,the M-DTA method is completely equivalent to the classical damping torque analysis(C-DTA)method.In the multi-machine power system,the mode-based damping torque directily reflects the effect of the controller on the oscillation mode,overcoming the shortcomings of the C-DTA method in which there is no direct correspondence between the damping torque and the oscillation mode.By deriving the relationship with the residue index,the M-DTA method shows higher accuracy than the residue method in applications,such as controller parameter adjustment.Finally,two example power systems are presented to demonstrate the application of the proposed M-DTA method.Index Terms-Electromechanical oscillation mode,FACTS,interconnection model in the frequency domain,mode-based damping torque analysis(M-DTA),power system low-frequency oscillation,PSS,residue method.展开更多
基金supported in part by the National Key R&D Program(2018AAA0101501)of Chinathe science and technology project of SGCC(State Grid Corporation of China).
文摘The complexity and uncertainty in power systems cause great challenges to controlling power grids.As a popular 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 threelevel 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.
基金supported in part by the National Natural Science Foundation of China under Grant No.U1766202,51907179 and 51977197.
文摘The mode-based damping torque analysis(M-DTA)method for studying the effect of an external controller on power system low-frequency oscillations is proposed in this paper.First,based on the interconnection model between the system and the controller in the frequency domain,the oscillation loop corresponding to the electromechanical oscillation mode is built,and then the mode-based damping torque of the controller can be calculated.Then,the application of the M-DTA method in the power system is illustrated.The derivation shows that in the single-machine infinite-bus power system,the M-DTA method is completely equivalent to the classical damping torque analysis(C-DTA)method.In the multi-machine power system,the mode-based damping torque directily reflects the effect of the controller on the oscillation mode,overcoming the shortcomings of the C-DTA method in which there is no direct correspondence between the damping torque and the oscillation mode.By deriving the relationship with the residue index,the M-DTA method shows higher accuracy than the residue method in applications,such as controller parameter adjustment.Finally,two example power systems are presented to demonstrate the application of the proposed M-DTA method.Index Terms-Electromechanical oscillation mode,FACTS,interconnection model in the frequency domain,mode-based damping torque analysis(M-DTA),power system low-frequency oscillation,PSS,residue method.