【作者】 孙勇；

【导师】 于继来； 李志民；

【作者基本信息】 哈尔滨工业大学 ， 电力系统及其自动化， 2009， 博士

【摘要】 随着电网规模的不断扩大,低频振荡问题已成为电力系统稳定问题的一个重要方面。尤其是区域电网的互联,导致系统的阻尼变得薄弱,低频振荡的风险加大,严重影响了区域间的功率交换和系统的稳定性。阻尼控制是解决低频振荡问题的一种有效方法。目前,电力系统主要采用电力系统稳定器(Power System Stabilizer, PSS)和柔性交流输电系统(Flexible AC Transmission Systems, FACTS)的附加阻尼控制器来抑制低频振荡。因此,如何从控制策略和控制方法上提高控制器的阻尼水平,减少控制器的交互影响,无论在理论方面还是实践方面都具有重要的研究意义。本文研究了附加阻尼控制器的配置问题,在此基础上对电力系统传统附加阻尼控制器进行协调优化设计,最后分别对新型附加励磁控制器和FACTS附加阻尼控制器的设计方法及控制策略进行了深入的研究。控制器之间的不良交互影响会降低控制器的控制效果,因此,从工程角度,附加阻尼控制器的数目应适量,合理选择附加阻尼控制器的安装位置也就成为大区域电网阻尼控制器协调的重要内容。本文利用奇异值分解(Singular Value Decomposition, SVD)的方法结合模型的辨识,解决了阻尼控制器的配置问题。该评价指标不需要系统状态矩阵的QR分解和传递函数留数的计算,通过求解系统输入-输出的传递函数矩阵的最大奇异值及其奇异向量来识别每台发电机对振荡模态的贡献程度,找出附加励磁阻尼控制器的最佳配置点,从而简化了计算过程和运算量,为附加阻尼控制器的协调奠定了基础。算例验证了该方法能快速和有效地配置附加阻尼控制器。为了使传统PSS有效地发挥阻尼作用,本文对多机电力系统PSS协调问题进行了研究。首先将粒子群优化(Particle Swarm Optimization, PSO)算法作为一个变异因子嵌入到细菌觅食优化(Bacterial Forging Optimization, BFO)算法,综合了PSO算法的全局搜索能力强和BFO算法局部搜索能力强的优点,提出了BFO-PSO混合算法。在此基础上,提出了基于PSO-BFO混合算法的电力系统PSS协调优化的新方法,该方法将阻尼控制器参数设计问题归结为带有不等式约束的目标优化问题,通过非线性变换将有约束问题转化为无约束问题,然后采用PSO-BFO混合优化算法求解优化问题设计阻尼控制器参数。为了增强系统的鲁棒性,该方法在目标函数中考虑了多种运行方式。最后通过算例验证了该方法的合理性,有效地抑制了低频振荡。为了使设计的控制系统具有很好的鲁棒性,大范围保证电力系统运行的稳定性,提出了最小熵H∞鲁棒附加励磁控制器的设计方法。首先,将PS/T灵敏度设计方法引入控制器设计,解决了基于Riccati方程法在设计过程中的零极点对消问题。针对加权函数的选取,提出了一种基于BFO-PSO混合算法的最小熵H∞鲁棒附加励磁阻尼控制器设计方法,把多个不同的设计目标转换为不等式约束,将加权函数的选取表示为一个多目标优化问题。为了减小系统的复杂性,增加控制器的实用性,利用Hankel范数最优降阶方法分别对原系统和控制器进行了两次降阶处理,使得控制器阶次相对较低。最后通过算例验证了控制器的有效性。针对区域间的功率振荡,为了有效提高区域间的振荡模态阻尼,本文对FACTS的附加阻尼控制器设计问题进行了研究。首先分析了SVC电压控制和阻尼控制的相互作用,在此基础上建立了SVC附加阻尼控制器设计模型,提出了分数阶附加阻尼控制器设计方案。针对分数阶控制器参数设计问题,引入BFO-PSO算法进行参数求解,提出了阻尼系统模态振荡和稳定节点电压的协调控制目标函数,将控制器设计转化为一个多目标的优化问题。最后,通过仿真验证了分数阶附加阻尼控制器的有效性。更多还原

【Abstract】 With the growing expansion of power grid, low frequency oscillation problem has become an important aspect of power system stability. The damping of power system has become more and more weak with the regional power sytems interconnections. So the increasing risk of low frequency oscillation brings a serious problem of restricting the inter-regional power exchange and stability of power system. Damoing control is an effective way to solve the problem of low frequency oscillation. At present, Power System Stabilizer (PSS) and Flexible AC Transmission Systems (FACTS) auxiliary damping controller are widely used to suppress low frequency oscillation. Therefore, how to improve the damping level of controller and reduce the controller’s interaction are very important and significant from the control strategies and methods, not only for theorerical research but for practical applications.First the allocation method of auxiliary damping controller is studied in this paper. Then the conventional auxiliary damping controllers of power system are coordinated and parameters optimizing design. Finally, the new PSS and FACTS auxiliary damping controller are detailed in two aspects of the design methods and control strategied.The auxiliary damping controller should be adequate from engineering perspective, because adverse interaction will reduce the control performace of controller. Reasonable choice to install auxiliary damping controller will become an important aspect of controller coordination in major regional power system. A new configuration method of auxiliary damping controller based on singular value decomposition (SVD) and model identify is proposed. This method solves the problem of the damping controller allocation. Since both QR decomposition of system state matrix and the calculation of residues on transfer function are not needed, this method can reduce the computational burden. With the solution to the largest singular value and singular vector of system transfer function matrix, the method can identify each generator’s contribution to oscillation modes, and then find the optimum allocation of supplementary damping excitation controller. This method laid the foundation for the coordination of the auxiliary damping controllers. Test results show that the method of SVD can find the effective allocation of damping fastly.In this dissertation, the coordination problem of traditional PSS is studied in order to make full use of damping effective in multi-machine power system. First of all, the Particle Swarm Optimization (PSO) algorithm as a variation factor is embedded in Bacterial Forging Optimization (BFO) algorithm, and BFO-PSO hybrid algorithm is proposed, which has the benefits of the strong global search capability of PSO and the good local search ability of BFO. On this basis the new method of coordination of power system damping controller is proposed. The method makes parameter design of damping controller boil down to the goal optimization problem. Through the nonlinear transform the constraint problem transformed unconstrained problem, and then use PSO-BFO hybrid optimization algorithm to design parameter of damping controllers. In order to enhance the system robustness, the method considers varies operation conditions in the objective function. Finally, by calculation analysis the method is verified to be reasonable, and effectively inhibited the low-frequency oscillations.In order to make the control system having good robustness, and ensure the stability of power systems in large-scale operation, the minimum entropy H∞additional excitation controller is proposed. For avoiding the zero-pole cancellation between controller and target by solving the Riccati equation, PS/T sensitivity design method is introduced. For the selection of weighing function, a minimum entropy additional excitation robust damping controller is proposed which transforms different design goals into inequality constraints, and makes the selection of weighing function change as a multi-objective optimization problem. In order to reduce the complexity of the system and increase the practicality of the controller, the optimal Hankel norm reduced-order method is used to deal with the original system and the controller, making a relatively low order controller.In order to regional modal oscillation damping, the auxiliary damping controller design method is studies. First of all, the interaction of voltage control and damping control for SVC is analyzed. On this basis additional damping controller model for SVC is founded, and the fraction rank additional damping controller is proposed. BFO-PSO algorithm is introduced to solving the parameter design of the fractional order controller. And the coordination control objective function is proposed, which can damp system modal oscillations and keeping node voltage stability. So the thesis turns the design of controller into a multi-objective optimization. Finally, by calculation analysis the fraction rank additional damping controller is verified to be effective.更多还原