【作者】 唐惠玲；

【导师】 吴杰康；

【作者基本信息】 广东工业大学 ， 控制科学与工程， 2019， 博士

【摘要】 能源和环境危机使风能和太阳能等可再生能源日益受到人们的关注,电动汽车零排放的特性也使得它正逐步成为汽车行业发展的新趋势。可再生能源发电出力受天气、地理位置以及季节变换等因素的影响而具有间歇性和不确定性,用户用车行为的不确定性也使电动汽车充电行为具有随机性。这些不确定性严重影响到分布式电源配电网的安全性。储能系统以其“源-荷”特性使得其可在两者之间实时切换,可以作为抑制可再生能源间歇出力的装置。新能源配电网是可再生能源、电动汽车和储能装置并存的系统,如何解决“源-荷-车-储”并存环境下配电网的线损和电压协同管理,保证分布式电源配电网的安全,已成为电力行业亟待解决的问题。本研究基于模糊集和锥规划相关理论,从新能源配电网的线损和电压协同管理的角度出发,针对新能源环境下配电网运行特点,研究了分布式电源出力的不确定性、电动汽车充电不确定性和负荷功率不确定性对配电网潮流分布的影响,综合考虑各种不确定性因素,建立多种协同管理模型及求解算法,较为系统地解决新能源环境下配电网线损和电压协同管理问题:(1)针对配电网中降损减压是一对相互制约的矛盾行为,进行线损和电压协同优化管理研究。分析配电网有功损耗与节点电压之间的关系模型,以DGs的出力作为控制变量,构建以线损指数最小和电压偏移指数小化为目标函数的多目标协调优化模型,比较PSO和模拟退火粒子群算法(SAPSO)两种优化算法的优略,并在EVs的两种充电模式下把SAPSO应用于多目标协调优化模型优化,得到新能源配电网线损和电压协同管理的最佳方案。以IEEE-118配电网验证了优化模型和算法是有效的。(2)针对并入分布式电源和电动汽车配电网的线损和电压协同优化管理问题,引入随机二阶锥规划方法,将电动汽车随机充电负荷和传统负荷分开考虑,实现优化管理。考虑节能减排和电网运行安全因素,以配电网的总线路损耗最小为目标函数,考虑系统运行约束,按照随机二阶锥规划的思想,将目标函数分解为两部分,一是配电网传统负荷运行时引起的线路损耗,二是电动汽车随机负荷引起的线路损耗,构造以分布式电源输出功率为优化变量的随机二阶锥规划模型。基于蒙特卡罗模拟电动汽车的负荷,嵌入内点法来求解随机二阶锥规划的模型。以IEEE-69配电系统为例验证了所提方法的有效性。(3)针对新能源环境下配电网中不确定因素风速、光照强度和电动汽车充电负荷等的随机模糊特性,基于概率统计及模糊集的相关理论对新能源配电网的线损和电压协同管理进行研究。首先根据随机模糊相容性原理,将不确定因素风速、光照强度和电动汽车充电负荷等的概率密度函数转化为可能性分布函数,再结合风机的风速-功率关系、光伏发电机的光照强度-功率关系得到风电机组和光伏发电机组有功出力的可能性分布函数,使得风力发电功率、光电发电功率和电动汽车充电功率变成了同时具备了随机性和模糊性的随机模糊变量,并建立包含有随机模糊变量的线损和电压协同管理的随机模糊机会约束规划数学模型,采用改进NSGA-Ⅱ算法求解随机模糊优化模型。最后以改进的IEEE-33配电网模型验证了该方案的有效性。(4)针对新能源配电网中DGs出力和EVs充电的不确定性引发的风险,基于CVaR风险评估模型的线损和电压风险指标,以DGs出力作为控制变量,建立以线损和电压风险为约束条件的线损和电压偏移最小多目标协同优化模型,通过变量替换将原优化模型转化为二阶锥规划模型,用原对偶内点法求解。以IEEE-118配电系统测试了本方法的有效性。更多还原

【Abstract】 Energy and environmental crisis makes renewable resource such as solar and wind resource attract people’s attention increasingly.Also,the zero-emission characteristics of electric vehicles make it gradually become a new trend of the development of the automobile industry.The output of renewable energy is affected by several factors such as weather,geographical location and seasonal changes.It is intermittent and uncertain.The uncertainty of users’ vehicle behavior also makes the charging behavior of electric vehicles random.These uncertainties seriously impact the safety of distributed power distribution networks.The energy storage system is capable of switching between the two in real time with its’source-load’ characteristic and can be used as a device for suppressing the intermittent output of renewable energy.The new energy distribution network is a system in which renewable energy,electric vehicles and energy storage devices coexist.How to solve the line loss and voltage cooperative management of the distribution network under the"source-load-vehicle-storage" coexistence environment and ensure the safety of the distributed power distribution network has become an urgent problem in the power industry.Based on the theory of fuzzy sets and cone programming,this study investigates two things from the perspective of line loss and voltage cooperative management of new energy distribution networks.For one thing,the influence of the uncertainty of distributed power output,the uncertainty of electric vehicle charging and the uncertainty of load power on the distribution of power flow in distribution network are studied.For the other thing,this study comprehensively considers various uncertain factors,establishes a variety of collaborative management models and solving algorithms,and systematically solves the problem of line loss and voltage coordination management of distribution network in new energy environment:(1)The power loss and voltage deviation in the distribution network is a pair of mutually restrictive and contradictory behaviors,research on line loss and voltage coordination optimization management is carried on.The relationship model between active power loss and node voltage of distribution network is analyzed.The output of DGs is used as the control variable to construct a multi-objective coordinated optimization model with the minimum line loss index and voltage deviation index as the objective function.The advantages of two optimization algorithms,PSO and simulated annealing particle swarm optimization(SAPSO)are compared.In the two charging modes of EVs,SAPSO is applied to multi-objective coordination optimization model optimization,and the best solution for line loss and voltage cooperative management of new energy distribution network is obtained.It is valid to verify the optimization model and algorithm with the IEEE-118 distribution network.(2)Aiming at the problem of line loss and voltage coordination optimization management in the distributed power supply and electric vehicle distribution network,a stochastic second-order cone programming method is introduced to consider the electric vehicle charging load and the general user load separately to achieve optimal management.Considering energy saving and emission reduction and grid operation safety factors,the total line loss of the distribution network is set as the objective function.Considering the system operation constraints,the objective function is divided into two parts according to the idea of stochastic second-order cone programming.The first part is the line loss caused by the general load operation of the distribution network.The second part is the line loss caused by the charging load of the electric vehicle.The stochastic second-order cone programming model,which sets the distributed power output power as the optimization variable,is constructed.The Monte Carlo method is used to simmulate the load of an electric vehicle,and the interior point method is embedded to solve the model of the stochastic second-order cone programming.The effectiveness of the proposed method is verified by an IEEE-69 node power distribution system.(3)Aiming at the stochastic fuzzy characteristics of uncertainties such as wind speed,illumination intensity and electric vehicle charging load in the new energy environment,the line loss and voltage cooperation management of new energy distribution network is studied based on the theory of probability statistics and fuzzy sets.Firstly,according to the principle of random fuzzy compatibility,the probability density function of uncertainties such as wind speed,illumination intensity and electric vehicle charging load is transformed into a probability distribution function.Combined with the wind speed-power relationship of the wind turbine and the light intensity-power relationship of the photovoltaic generator,the probability distribution function of the active output of the wind turbine and the photovoltaic generator set is obtained.Thus the wind power,the photovoltaic power and the electric vehicle charging power become simultaneously random fuzzy variables with randomness and ambiguity.A mathematical model of stochastic fuzzy chance constrained programming with line loss and voltage coordination management with random fuzzy variables is established.The improved NSGA-II algorithm is used to solve the stochastic fuzzy optimization model.Finally,the effectiveness of the scheme is verified by the improved IEEE-33 node distribution network model.(4)Aiming at the risk caused by the uncertainty of DGs output and EVs charging in the new energy distribution network,the multi-objective cooperation optimization model for line loss and voltage deviation is established.The model is restrained by the CVaR line loss risk and the CVaR voltage risk.The DG output is set as the control variable in the model.Variable substitution is used to transform the original optimization model into a second-order cone programming model,and the original dual interior point method is used to solve the problem.The effectiveness of the method is tested with an IEEE-118 node power distribution system.更多还原