Study On Load Model Aggregation Of Electric Power Systems

Power system transient stability simulation is the fundamental work for power system planning, design and operation, so the simulation accuracy is very important. Electric Power load model and its parameters have great impact on the simulation results. However, it is impossible to model all induction motors in the sub-transmission system, so the load models at the low-voltage level need to be aggregated to obtain the equivalent load model at the high-voltage level. Because of the deficiency of the traditional weighting-mean method and the need for the development of transient stability simulation software, the main work of this thesis is as follows:1. The traditional weighting-mean method is less accurate because it doesn't consider the operating condition of the motors. An improved weighting-mean method for aggregating a group of induction motors with considering load rates and critical slips is suggested. The improved method provides more accurate load model for transient stability simulation compared with the traditional weighting-mean method. The three different types of induction motors, composed of small industrial motor, large industrial motor and aggregated residential motor model, can be aggregated into one equivalent induction motor using this method in expectation to reduce the simulation burden. When the aggregation begins from the low-voltage level, the parameters of equivalent induction motor obtained by the so-called statistic synthesis based method could be acceptable usually considering their random characters of load rates. The effectiveness of the equivalent motor obtained by the statistic synthesis based method can be checked when the differences between the given random characters of load rates and the practical random characters of load rates can be evaluated.2. On the basis of previous work, we construct a development platform for transient stability simulation, using object-oriented programming. It is compatible and transplantable easily. The platform simulation functions rely on the dynamic component class. External data-mining module only scans component locations and simulation data are set linked with component parameters by pointers, which can reduce internal data operations largely. The platform can be used to simulate bulk power systems and give support to the calculation of the Dynamic Security Region.