【作者】 王慧；

【导师】 赵书强；

【作者基本信息】 华北电力大学(北京) ， 电气工程， 2024， 博士

【摘要】 通过虚拟惯量控制,可使变流器在暂态过程中主动支撑电网电压,进而提高电能质量。然而,虚拟惯量控制会改变变流器的端口特性,进而影响系统的稳定性及动态特性。为实现系统的稳定运行与优化控制,本文首先建立含虚拟惯量的直流微电网模型,为后续研究奠定基础;然后研究直流微电网负荷扰动下的大信号稳定运行方法,避免因负荷突变而引起的暂态失稳;最后,从分散和集中控制两个层面分别研究虚拟惯量自适应和集中优化方法,以改善直流电压的稳定性和动态特性。本文的主要内容与创新工作如下:(1)建立了含虚拟惯量的直流微电网模型,为系统的稳定运行与优化控制研究奠定了基础。建立含并网变流器、储能系统、新能源发电系统和负荷的直流微电网传统控制模型;在此基础上,为了抑制直流电压的快速波动并减小功率扰动对上级电网的影响,将虚拟电容和虚拟直流电机(VDCM)控制分别应用于储能系统;最后对比分析了不同控制策略下直流微电网的动态特性。(2)针对负荷变化较大,即使是稳态工作点之间切换,也可能发生暂态失稳的问题,提出了基于吸引域估计考虑VDCM和柔性负荷调控的大信号稳定运行方法。对于未知的一般负荷扰动以及已知的柔性负荷投入,估算直流微电网的目标稳态点,并基于Takagi-Sugeno(TS)模糊模型估算吸引域。若存在失稳风险,则分析VDCM虚拟阻尼、虚拟惯量控制参数或柔性负荷投入方案对系统稳定性的影响,进而进行优化,让目标稳态点的估计吸引域(EDA)直接或者间接包含初始稳态点,从而消除失稳风险。研究结果表明,基于吸引域估计对VDCM和柔性负荷进行调控,是一种实用有效保障系统稳定运行的方法。(3)针对变流器惯量支撑能力难以充分发挥以及暂态响应时间过长的问题,提出了基于下垂曲线截距调整的虚拟惯量自适应控制方法。揭示下垂曲线截距调整形成虚拟惯量的机理。在此基础上,考虑下垂曲线的截距限值,并在反正切函数中嵌套幂函数,进而根据直流电压及其变化率实时调整下垂曲线的截距,为直流微电网提供自适应的惯量支撑。研究结果表明,该方法在负荷突变瞬间能充分利用变流器的惯量支撑能力,并在负荷突变后能按照动态性能要求减少暂态响应时间。(4)针对传统虚拟惯量集中控制难以兼顾直流电压稳定性、动态特性和安全性的问题,提出了基于模型预测控制的虚拟惯量集中优化方法。建立直流微电网的线性离散模型,然后设计模型预测控制器:以直流电压跟踪误差、虚拟惯量系数两者的加权平方和最小为目标,希望直流电压维持稳定并且惯性响应的强度不要太大;以直流电压及其变化率限制为约束,防止因动态电压超过安全阈值而导致切机切负荷甚至系统崩溃问题;在每个采样时刻,求解各控制单元的虚拟惯量系数并将其输出,从而实现虚拟惯量的集中优化控制。研究结果表明,该方法在提高直流电压稳定性、动态特性的同时兼顾了安全性。更多还原

【Abstract】 Through virtual inertia control,the inverter can actively support the grid voltage during transient processes,thereby improving power quality.However,virtual inertia control can alter the port characteristics of the inverter,thereby affecting the stability and dynamic characteristics of the system.To achieve stable operation and optimized control of the system,this paper first establishes a DC microgrid model with virtual inertia,laying the foundation for subsequent research;Then study the large signal stable operation method of DC microgrid under load disturbance to avoid transient instability caused by sudden load changes;Finally,research virtual inertia adaptive and centralized optimization methods from both decentralized and centralized control levels to further improve the stability and dynamic characteristics of DC voltage.The main content and innovative work of this article are as follows:(1)A DC microgrid model with virtual inertia has been established,laying the foundation for the stable operation and optimization control research of the system.Establish a traditional control model for a DC microgrid that includes grid connected inverters,energy storage systems,new energy generation systems,and loads;On this basis,in order to suppress the rapid fluctuation of DC voltage and reduce the impact of power disturbance on the higher-level power grid,virtual capacitor and virtual DC motor(VDCM)control are respectively applied to the energy storage system;Finally,the dynamic characteristics of DC microgrids under different control strategies were compared and analyzed.(2)A large signal stable operation method based on attraction domain estimation considering VDCM and flexible load regulation is proposed to address the issue of transient instability that may occur even when switching between steady-state operating points due to significant load changes.For unknown general load disturbances and known flexible load increases,estimate the target steady-state point of the DC microgrid and estimate its attraction domain based on the Takagi-Sugeno(TS)fuzzy model.If there is a risk of instability,analyze the impact of VDCM virtual damping and virtual inertia control parameters or flexible load increase schemes on system stability,and then optimize them so that the estimated domain of attraction(EDA)of the target steady-state point directly or indirectly includes the initial steady-state point,thereby eliminating the risk of instability.The research results indicate that using domain of attraction estimation to regulate VDCM and flexible loads is a practical and effective method to ensure stable system operation.(3)A virtual inertia adaptive control method based on droop curve intercept adjustment is proposed to address the issues of insufficient inertia support capability of the inverter and long transient response time of the system.Reveal the mechanism of virtual inertia formed by adjusting the intercept of the droop curve.On this basis,the intercept limit of the droop curve is considered,and a power function is nested in the arctangent function to adjust the intercept of the droop curve in real time based on the DC voltage and its rate of change,providing adaptive inertia support for the DC microgrid.The research results indicate that this method can fully utilize the inertia support capability of the inverter at the moment of load sudden change,and can reduce transient response time according to dynamic performance requirements after load sudden change.(4)Aiming at the problem that traditional virtual inertia centralized control is difficult to balance DC voltage stability,dynamic characteristics,and safety,a virtual inertia centralized optimization method based on model predictive control is proposed.Establish a linear discrete model of a DC microgrid,and then design a model predictive controller with the goal of minimizing the weighted sum of squares of the DC voltage tracking error and virtual inertia coefficient.It is hoped that the DC voltage will remain stable and the intensity of inertia response will not be too large;Constrained by the DC voltage and its rate of change,to prevent machine cutting,load shedding,and even system collapse caused by dynamic voltage exceeding the safety threshold;At each sampling moment,solve the virtual inertia coefficients of each control unit and output them to achieve centralized optimization control of virtual inertia.The research results indicate that this method improves the stability and dynamic characteristics of DC voltage while also taking into account safety.更多还原