Research On The Engineering Realization And Application Of Electrical Load Modeling

Against the background of engineering project, this thesis lays emphasis on the following questions in the application of electrical load modeling:1. Particle Swarm algorithm, which is quite immune to local optima and is fairly efficient in solving problems with complex hyperspace is introduced into the field of electrical load parameter identification. The convergent efficiency and searching ability of PS algorithm , genetic algorithm (GA) and evolutionary strategy (ES) are compared. The best w which fits for the problem of load parameter identification is presented. Furthermore, a composite PS algorithm is proposed, and expected results are obtained when the new algorithm with optimal w is applied to the field data.2. A criteria of selecting suitable data for load modeling is given. The idea of deciding the construction of load model according to the dynamic degree of field data is also presented in detail. Meanwhile, the measurement of dynamic degree of the data is given.3. A new method of dynamic load classification is proposed. It is based on the eigenvalue of load model and has less vector dimension and reflects more intrinsic characteristics of dynamic load.4. The influence of induction-motor ration on the problem of transient stability is discussed. The case of HeNan provincial power system using actual load parameters is studied. Comparative results show that the influence of dynamic load characteristics on the stability of power system is important and profound.5. Software of Electrical Load Modeling which including the interface between the actual load parameter database and PSASP database is introduced in this thesis. It makes the use of actual load parameter easier and the theoretic research of load modeling practical.Research work of this thesis help solving the difficulties of diversification of actual load parameters and has great significance in the application of electrical load modeling.