In recent years,reinforcement learning(RL)has emerged as a solution for model-free dynamic programming problem that cannot be effectively solved by traditional optimization methods.It has gradually been applied in the...In recent years,reinforcement learning(RL)has emerged as a solution for model-free dynamic programming problem that cannot be effectively solved by traditional optimization methods.It has gradually been applied in the fields such as economic dispatch of power systems due to its strong selflearning and self-optimizing capabilities.However,existing economic scheduling methods based on RL ignore security risks that the agent may bring during exploration,which poses a risk of issuing instructions that threaten the safe operation of power system.Therefore,we propose an improved proximal policy optimization algorithm for sequential security-constrained optimal power flow(SCOPF)based on expert knowledge and safety layer to determine active power dispatch strategy,voltage optimization scheme of the units,and charging/discharging dispatch of energy storage systems.The expert experience is introduced to improve the ability to enforce constraints such as power balance in training process while guiding agent to effectively improve the utilization rate of renewable energy.Additionally,to avoid line overload,we add a safety layer at the end of the policy network by introducing transmission constraints to avoid dangerous actions and tackle sequential SCOPF problem.Simulation results on an improved IEEE 118-bus system verify the effectiveness of the proposed algorithm.展开更多
The E24 profile slope analyzed belongs to a series of excavated slopes of the Haizhou Opencast Coal Mine. It seems to be divided into Downslope Part and Upslope Part. Its profile comprises two noticeable coal seams, c...The E24 profile slope analyzed belongs to a series of excavated slopes of the Haizhou Opencast Coal Mine. It seems to be divided into Downslope Part and Upslope Part. Its profile comprises two noticeable coal seams, called the 8# and 9# weak layers, considered as the potential failure surfaces. In consideration of the actual configuration as in the perspective of any modification, assessing the stability of this slope with various profile forms under given conditions, and assessing the risk of instability and quantifying the influence of earthworks or other modifications to the stability of this slope, have constituted the primordial objectives carried out. From assumed potential failure surfaces, any specific profiles and specified slip surfaces are defined. A factor of safety (FoS) is computed for each specified slip surface; the smallest FoS found corresponds to the least favorable slip surface. The safety factor values obtained are compared to the suggested safety factor. Limit equilibrium methods of vertical slices implemented in Slope/W, computer program for slope stability analyses, have been adopted to perform the E24 slope stability analysis. The safety factor values computed with 9# weak layer are lower than for 8#; the factors of safety obtained with Sarma's method are the smallest; more, without groundwater (long term) overall values are greater than those determined under groundwater condition (short term). The lowest safety factor value is found for a profile depending on an adopted earthwork sequence. The E24 profile slope stability analysis shows the instability risk for the deepest weak layer, and also shows the short and long term stability of this slope for the envisaged earth movements. However it demonstrates the existence of instability risk for any earthwork firstly affecting the downslope part.展开更多
基金supported in part by National Natural Science Foundation of China(No.52077076)in part by the National Key R&D Plan(No.2021YFB2601502)。
文摘In recent years,reinforcement learning(RL)has emerged as a solution for model-free dynamic programming problem that cannot be effectively solved by traditional optimization methods.It has gradually been applied in the fields such as economic dispatch of power systems due to its strong selflearning and self-optimizing capabilities.However,existing economic scheduling methods based on RL ignore security risks that the agent may bring during exploration,which poses a risk of issuing instructions that threaten the safe operation of power system.Therefore,we propose an improved proximal policy optimization algorithm for sequential security-constrained optimal power flow(SCOPF)based on expert knowledge and safety layer to determine active power dispatch strategy,voltage optimization scheme of the units,and charging/discharging dispatch of energy storage systems.The expert experience is introduced to improve the ability to enforce constraints such as power balance in training process while guiding agent to effectively improve the utilization rate of renewable energy.Additionally,to avoid line overload,we add a safety layer at the end of the policy network by introducing transmission constraints to avoid dangerous actions and tackle sequential SCOPF problem.Simulation results on an improved IEEE 118-bus system verify the effectiveness of the proposed algorithm.
基金Project of NSFC (No. 40472136)Scientific Research Foundation for the Returned Overseas Chinese Scholar, State Education Ministry(No. 120413133)985 Project of Jilin University (No. 105213200500007)
文摘The E24 profile slope analyzed belongs to a series of excavated slopes of the Haizhou Opencast Coal Mine. It seems to be divided into Downslope Part and Upslope Part. Its profile comprises two noticeable coal seams, called the 8# and 9# weak layers, considered as the potential failure surfaces. In consideration of the actual configuration as in the perspective of any modification, assessing the stability of this slope with various profile forms under given conditions, and assessing the risk of instability and quantifying the influence of earthworks or other modifications to the stability of this slope, have constituted the primordial objectives carried out. From assumed potential failure surfaces, any specific profiles and specified slip surfaces are defined. A factor of safety (FoS) is computed for each specified slip surface; the smallest FoS found corresponds to the least favorable slip surface. The safety factor values obtained are compared to the suggested safety factor. Limit equilibrium methods of vertical slices implemented in Slope/W, computer program for slope stability analyses, have been adopted to perform the E24 slope stability analysis. The safety factor values computed with 9# weak layer are lower than for 8#; the factors of safety obtained with Sarma's method are the smallest; more, without groundwater (long term) overall values are greater than those determined under groundwater condition (short term). The lowest safety factor value is found for a profile depending on an adopted earthwork sequence. The E24 profile slope stability analysis shows the instability risk for the deepest weak layer, and also shows the short and long term stability of this slope for the envisaged earth movements. However it demonstrates the existence of instability risk for any earthwork firstly affecting the downslope part.
