Research On Method Of Reactive Power Compensation Based On Geometric Algebra

The research of reactive power is always the focus in power system. It is important both in theory and industrial practice. The use of many nonlinear loads such as the power electronic devices and electric arc furnace brings on more and more serious unbalance of reactive power that result in power quality problem. It's necessary to study the physical mechanism of reactive power, and give a universal, correct mathematical description, which used to better compensate for reactive power and to improve the power factor effectively.Firstly the background, development and the problems of the theory of reactive power in recent years is described in detail. A new definition theory of reactive power based on geometric algebra is introduced in this thesis, then the mathematical model called power multi-vector is proposed. Based on the reactive power definition in single-phase circuits, the definition of reactive power respectively in three-phase three-wire circuit and three-phase four-wire circuit are proposed. In this method, the decomposition of current and voltage is carried out in vector space, which can clearly display the multidimensional nature of power. It overcomes the defects and limitations in traditional definitions and models, and it is well used in the passive filter compensation circuit as well as active filter. According to the theory proposed in this paper, the models of single-phase and three-phase active filter compensation circuit are established respectively. Some principles and performance analysis are given in the last, which can provide some reference to the detected method based on geometric algorithm using in active filters. A comparison between the new detective method and the algebra based on the instantaneous reactive power theory is taken. In this paper, several examples in single-phase, three-phase three-wire and three-phase four-wire circuit with nonlinear load are presented respectively, through the MATLAB simulation, the results confirmed that the method proposed in this paper has obvious advantage in compensation. It is useful to serve practical purposes, such as power factor improvement, and to ensure the stability of power system.