Voltage And Reactive Power Optimization Based On Graph Theory Partition

Voltage and reactive power optimization is an important means for power resources configuration and improvement of system security and economic profit. This dissertation discusses the voltage and reactive power optimization algorithm, the power network partitioning and its practicability and application scheme. To overcome the shortcomings of existing algorithms, the sub-area division algorithm of the power systems based on graph theory is proposed according to the distributed and decentralized characteristics of the power system. Here, the sub-area division reactive power optimization problem is solved by the direct nonlinear primal-dual interior point algorithm. Combined above methods with the multi-agent technology, a multi-agent system based distributed voltage and reactive power system is proposed. The research includes following contents:1. This paper takes the direct nonlinear primal-dual interior point arithmetic to resolve the voltage and reactive optimization problem. The arithmetic starts at an interior point, gets the next interior point in the descending direction of object value, then from this point, gets another interior point in the descending direction of object value. A series of interior points are gotten by repeating the above steps. When the ending rule is satisfied, the optimized object value is gotten. This paper also discusses several problems in its solution, such as the choice of corrector steps and decision of obstacle gene. It also introduces the principle of advanced predictor-corrector primal-dual interior point arithmetic, that is by introducing predictor and corrector to correspond the optimization and feasibility, the convergence ability is improved.2. A good method of dividing control regions should make the buses in the same region close, and those in different regions less coupled. Moreover, the bus numbers in all regions are similar and the numbers of frontier buses are as small as possible. According to the characteristics of voltage and.reactive power control, the maximum-cardinality matching based fast sub-area division algorithm is presented in this dissertation. There are two steps in this algorithm. 1): Define the electrical distance that reflects the influence degree between buses. Then the method of ascending hierarchical classification is adopted to preliminarilypartition the power networks. Close buses with small electrical distance can be classified into one sub-area in this step. 2): On the basis of step 1, the improved multiple-way graph-partitioning algorithm is introduced to reduce the frontier buses of each sub-area in turn. For a sub-area, the algorithm iteratively establishes the corresponding bipartite graph, and then searches its maximum-cardinality matching and redivides power networks according to the unmatched node subset within the frontier node set.In this dissertation, the multiple-way graph-partitioning algorithm is first introduced to divide power networks into several sub-areas for voltage and reactive power control. Besides, several important deductions are educed for the first time to improve the multiple-way graph-partitioning algorithm. The improved algorithm can reduce not only frontier nodes as many as possible at each iteration but also total iteration number. The maximum-cardinality matching based fast sub-area division algorithm is applicable to voltage and reactive power optimization and control of large-scale power systems. It can minimize the mutual voltage control effect between sub-areas, and speed up the parallel optimization through reducing the number of frontier buses and balancing the numbers of interior buses of all sub-areas.3. Present voltage and reactive power control systems have some shortcomings such as bad coordination between sub-areas, slow reactive power optimization and huge cost. A novel voltage and reactive power optimal control system based on the multi-agent technique is proposed after the control sub-area division problem are solved. The distributed and parallel reactive power optimization algorithm based on sub-area division of the power systems is firstly combined with the multi-agent technology according to its characteristics. The multi-agent system based distributed voltage and reactive power system is proposed in this dissertation. On one hand, The distributed and parallel characteristic of the algorithm is suitable for multi-agent system, On the other hand, the multi-agent technology can coordinate the decentralized logical or physical systems to solve a problem in parallel. The multi-agent system provides a flexible intelligent platform for application of the algorithm. The multi-agent system proposed in this dissertation is of hierachical and distributed structure. Its functions and operation mechanism are introduced. Simulation results show that this system has the advantage of fast speed, and can reduce network power loss and improve voltagequality obviously. Besides, benefiting from its reliability, flexibility and adaptability, it is convenient to be realized on the basis of the existing voltage and reactive power optimization software in our country. Some instructive work about the multi-agent system based distributed voltage and reactive power system in this dissertation lays a foundation for further research and discussion in this aspect.