Research On Emergency Control For Transient Voltage Security Based On Mixed-integer Optimal Control Theory

Transient voltage security is one of the major security problems in modern power systems. Based on mixed-integer optimal control theory, the emergency control problem for transient voltage security can be solved based on stricter theoretical basis, which would bring an important guidance for engineerings. The optimal control problem for transient voltage security can be described as a mixed-integer optimal control model, how to deal with the integer control variables is a difficult problem to solve the model. With the present basis of engineerings, how to apply the optimal control strategy is also an important problem. Besides, under the background of large-scale power systems, it is a key point of the practical application of the optimal control for transient voltage security to build the optimal control algorithm. In view of the problems above, this thesis studies the optimal control problem for transient voltage security from the following aspects:1) A convex relaxation method for transient voltage security mixed-integer optimal control problem is proposed. By means of adjusting the generator excitation reference voltage and switching the capacitor/reactor banks, a multi-objective mixed-integer optimal control model for transient voltage security control is established. To solve the model, the normalized normal constraint method is used to transform it into a series of single objective mixed-integer optimal control model, and the convex relaxation technique is applied to transform it into continuous optimal control model. Then, we can get integer solution by optimizing the control variable’s magnitude and switching time. If this method couldn’t get integer solution, we add penalty function to handle the integer variable. The comparison between the proposed method and GAMS/SBB solver on IEEE 3-machine 9-bus and IEEE 10-machine 39-bus systems demonstrate the effectiveness of proposed method.2) An optimizing method for the control settings of the fast switching of capacitor banks is proposed to realize the optimal control for transient voltage security in a decentralized and coordinated manner. A mixed-integer dynamic optimization model considering multiple contingencies is established to depict the problem of tuning control settings, and the optimal range of these control settings are obtained by inverse derivation from the optimal capacitors’ outputs corresponding to each contingency. To solve the dynamic optimization problem that considered multiple contingencies, a two-level parallel framework, in terms of decoupling of contingencies and time intervals, is designed to improve computational efficiency based on the multiple shooting method and reduced space sequential quadratic programming algorithm. Furthermore, simulation results on IEEE 10-machine 39-bus test system and a real 165-machine 1003-bus provincial system demonstrate the effectiveness of the proposed method.3) An area decoupled multiple shooting method is proposed to solve the optimal control problem for transient voltage security in large-scale power system. Based on the multiple shooting method, with the introduction of area decoupling method, the optimal control problem of large-scale power system is transformed into a small-scale nonlinear programming problem and a lot of time domain simulation problems of decoupled small systems, which are solved alternately. Among them, the nonlinear programming problem is solved by reduced space sequential quadratic programming algorithm, and the simulation problems of small systems are computed parallel. In addition, the discrete orthogonal polynomials are used to fitting the curves of area boundary variables, which has further improved the computational efficiency of the area-based decomposition method. The comparison between the proposed method and the original multiple shooting method on a real 165-machine 1003-bus provincial system demonstrate the correctness and effectiveness of the proposed method.