基于动态轨迹的统一性潮流法

发布时间：2018-06-09 22:51

本文选题：潮流计算 + 牛顿法　；参考：《天津大学》2014年博士论文

【摘要】：潮流计算是电力系统分析的一项基本工作。随着电力系统的快速发展，电网的规模与复杂度日益增加，电力系统的非线性程度越来越高（重载情况下尤为明显），传统潮流算法经常遇到不收敛的情况。因此有必要发展鲁棒并高效的潮流计算方法。本文研究了以非线性动力系统理论为基础的潮流方法。主要工作分为以下几个方面：（1）首先研究了牛顿潮流算法的状态空间收敛域特性。采用数值仿真的方法对牛顿法的收敛域进行了刻画，分析了牛顿法收敛域在基态和负荷变化时的特性，从收敛域的角度揭示了牛顿法对初值的敏感性和负荷条件对牛顿法收敛性的影响机理。（2）研究了基于动力系统轨迹的潮流方法，将潮流计算问题转换为动力系统的求解问题，，通过追踪动力系统轨迹获得稳定平衡点来得到潮流方程的解。首先介绍了基于动力系统轨迹的潮流方法的基本思想，分析了该方法的特点及优势。然后介绍了几种典型动力系统方程的构造形式，给出动力系统稳定平衡点和潮流解之间的关系，以及不同潮流问题下动力系统的适用情况。然后，介绍了该方法用于计算潮流多解问题时的求解思路。最后对这种方法进行了数值测试。测试结果表明基于动力系统轨迹的潮流方法可以用于计算初值问题和病态问题引起的牛顿法不收敛的情况，表明了这种方法的有效性和鲁棒性。（3）基于动力系统轨迹的潮流法，潮流解的收敛域对应着动力系统稳定平衡点的稳定域。本文首先通过数值方法刻画了几种典型动力系统的稳定域，分析了稳定域在负荷变化时的特性，从稳定域的角度揭示了基于动力系统轨迹的潮流法具有邻近收敛性，同时说明选取合适的动力系统形式（如QGS形式）时基于动力系统轨迹的潮流方法不会受到初值和病态条件的影响。其次证明了QGS形式的动力系统满足稳定边界的刻画定理，并依据定理设计了QGS形式动力系统的稳定边界的系统化刻画方法。（4）本文结合静态方法和动态方法的特点，提出了基于动态轨迹的统一性潮流法（TraJectory-based Unified Power Flow Method，简称TJU潮流法）。首先介绍了基于动态轨迹的统一性潮流法的基本思想，以及求解的三个阶段：准确轨迹、近似轨迹和快速求解。在此基础上根据静态方法和动态方法的结合情况设计了两种整体实现方案。其次，针对SDF形式和QGS形式的动力系统特点，提出了基于这两种动力系统的TJU潮流法。最后总结得出了TJU潮流法的特性。TJU潮流法兼具动态方法和静态方法的优点，具有良好的收敛性、邻近收敛性、鲁棒性和计算速度。（5）基于动态轨迹的统一性潮流法算法层的主要内容有：首先探讨了TJU潮流算法的设计中动态方法和静态方法的选取，根据TJU算法的基本思想和数值方法的特点，静态方法采用牛顿法，动态方法采用Pseudo-Transient方法。根据TJU潮流法的整体设计方案，提出了TJU基本潮流算法和TJU增强潮流算法，并详细介绍了具体的实现流程。然后通过数值计算的方式分析了TJU潮流算法中几个因素对计算速度的影响，对具体实现时参数的选取有一定的指导意义。最后，通过测试算例对TJU潮流算法进行了评估。测试结果表明TJU潮流算法具有良好的收敛性和计算速度，可以有效地求解初值问题和病态问题，并且对大系统同样有效。
[Abstract]:Power flow calculation is a basic work of power system analysis. With the rapid development of the power system, the scale and complexity of the power grid are increasing, the nonlinear degree of the power system is getting higher and higher (especially in the heavy load case). The traditional power flow algorithm often meets the situation of non convergence. Therefore, it is necessary to develop a robust and efficient power flow meter. The power flow method based on nonlinear dynamic system theory is studied in this paper. The main work is divided into the following aspects:
(1) first, the properties of the state space convergence domain of the Newton flow algorithm are studied. The numerical simulation method is used to characterize the convergence domain of the Newton method. The characteristics of the Newton's convergence domain in the ground state and the load change are analyzed. The sensitivity of the Newton method to the initial value and the convergence of the Newton method are revealed from the angle of the convergence domain. Influence mechanism.
(2) the power flow method based on the power system trajectory is studied. The power flow calculation problem is converted to the solution of the dynamic system. The solution of the power flow equation is obtained by tracking the trajectory of the power system to obtain the solution of the power flow equation. First, the basic idea of the power flow method based on the dynamic system trajectory is introduced, and the characteristics and advantages of the method are analyzed. After introducing the structural forms of several typical dynamic system equations, the relationship between the stable equilibrium point of the power system and the solution of the power flow, and the application of the power system under different power flow problems are given. Then, the solution of this method is introduced when it is used to calculate the multi solution of the power flow. Finally, the numerical test of this method is carried out. The results show that the power flow method based on the dynamic system trajectory can be used to calculate the non convergence of Newton method caused by the initial value problem and the ill conditioned problem, which shows the effectiveness and robustness of the method.
(3) based on the power system trajectory method, the convergence domain of the power flow solution is corresponding to the stable equilibrium of the dynamic system. In this paper, the stability domain of several typical dynamic systems is depicted by numerical method, the characteristics of the stable domain in the load change are analyzed, and the power flow method based on the dynamic system trajectory is revealed from the angle of the stable domain. At the same time, it shows that the power system based power flow method based on the dynamic system path is not affected by the initial value and the ill conditioned condition when selecting the appropriate dynamic system form (such as QGS form). Secondly, it is proved that the dynamic system of QGS form satisfies the stable boundary characterization theorem, and the stable boundary of the power system of QGS form is designed according to the theorem. A systematic portrayed method.
(4) in this paper, combining the characteristics of static and dynamic methods, the unified power flow method (TraJectory-based Unified Power Flow Method, called TJU tidal current method) based on dynamic trajectory is proposed. First, the basic idea of the unified tidal current method based on dynamic trajectory is introduced, and the three stages of the solution are described, the exact trajectory, the approximate trajectory and the fast track. Two overall implementation schemes are designed on the basis of the combination of static and dynamic methods. Secondly, based on the characteristics of SDF and QGS dynamic systems, the TJU flow method based on these two power systems is proposed. Finally, the special.TJU flow method of the TJU flow method is concluded with both dynamic and static methods. The advantages of the method are good convergence, convergence, robustness and computation speed.
(5) the main contents of the algorithm layer of the unified power flow method based on dynamic trajectory are as follows: first, the selection of dynamic and static methods in the design of TJU power flow algorithm is first discussed. According to the basic ideas of the TJU algorithm and the characteristics of the numerical method, the static method adopts the Newton method, the dynamic square method adopts the Pseudo-Transient method. The whole method is based on the TJU tidal current method. The TJU basic power flow algorithm and the TJU power flow algorithm are proposed, and the concrete implementation process is introduced in detail. Then, the influence of several factors on the calculation speed in the TJU flow algorithm is analyzed by numerical calculation, and the parameters are selected as a guide for the specific implementation. Finally, a test example is given to the TJU. The test results show that the TJU power flow algorithm has good convergence and calculation speed, and can effectively solve the problem of initial value and ill condition, and it is also effective for large systems.
【学位授予单位】：天津大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TM744

【参考文献】