重庆理工大学学报2025，Vol.39Issue(5)：243-250,8.DOI:10.3969/j.issn.1674-8425(z).2025.03.030
计及风温时序相关性的电热耦合概率潮流计算
Electric-thermal coupled probabilistic power flow calculation considering temporal correlation of wind and temperature
摘要
Abstract
Since the implementation of the development goals of"carbon peaking&carbon neutrality"and the construction of a new type of power system,the integration capacity of new clean energy sources such as wind and solar power has been growing rapidly.The power system gradually evolves into a new one characterized by a high proportion of new energy integration and a high proportion of power electronic devices feeding into the grid.This markedly increases the uncertainties in the operation and planning of the new power system,posing challenges to the economy,safety,and reliability requirements of system operation and planning. Probabilistic power flow(PPF)is a key to addressing the impact of uncertainties in power systems.The PPF method considers the effects of various stochastic uncertainties on the power flow characteristics of the system.The PPF of the power network depends not only on the probabilistic characteristics of power generation output and load but also on the changes in the parameters of the power system components caused by external environmental factors.Although PPF algorithms have become relatively mature,several obvious deficiencies still exist.First,existing studies normally fail to consider the variation in line impedance when performing probabilistic power flow calculations,and instead use fixed line parameters corresponding to a temperature of 20℃.This approach overlooks the influence of complex environmental factors(such as wind speed,temperature,and sunlight)on conductor temperature,even though environmental factors exert significant impacts on the electrical parameters of transmission lines.Second,current research gives no further account of the correlation between environmental factors and their effects on the results of PPF calculations.Research has shown there are significant long-range correlations between wind speed and temperature.Although many well-performing methods exist for handling correlations,such as the copula function,orthogonal transformation method,and Rosenblatt transformation,these methods are often too simplistic when dealing with stochastic processes that exhibit long-term dependencies(such as autoregressive structures in time series data)or memory effects.Consequently,they fail to adequately quantify and consider these dependencies,and have not effectively incorporated the intrinsic correlation between wind speed and temperature into PPF calculations. To address the limitations of conventional PPF and effectively account for the impact of wind temperature on line impedance,we propose an electro-thermal coupled PPF calculation model that incorporates the temporal correlation between wind speed and temperature.It corrects the power flow results by thoroughly considering the long-range correlation between wind speed and air temperature.First,a heat balance equation is introduced to establish the relationship between environmental factors and line temperature.Next,a rank-long short-term memory(RLSTM)neural network is built to account for the correlation between wind speed and temperature.Based on impedance correction factors calculated under different scenarios,the system parameters are updated in real-time,and probabilistic power flow calculations are performed employing a Monte Carlo simulation based on Latin hypercube sampling.Finally,the model is tested and verified on the IEEE-30 node system.Our results show after considering wind-temperature correlation,the error in the PPF results is markedly reduced.The average absolute errors in active power loss,the flow power,and the voltage amplitude decrease by 61.27%,41.25%and 28.90%respectively.The root mean square errors are down by 59.04%,20.98%and 31.18%respectively.These findings demonstrate incorporating the temporal correlation of wind and temperature into the RLSTM neural network model effectively improves the PPF results,making them more aligned with the actual operation of the power grid and enhancing the model's accuracy and reliability.关键词
时序相关性/概率潮流/热平衡方程/蒙特卡洛模拟/长短期记忆神经网络Key words
temporal correlation/PPF/heat balance equation/Monte Carlo simulation/long short-term memory neural networks分类
信息技术与安全科学引用本文复制引用
陈将宏,唐云滨,苗媛媛,谌论佳..计及风温时序相关性的电热耦合概率潮流计算[J].重庆理工大学学报,2025,39(5):243-250,8.基金项目
强电磁工程与新技术国家重点实验室开放基金项目(2022KF005) （2022KF005）
