Xueyang Cheng

Graduation Semester and Year




Document Type


Degree Name

Doctor of Philosophy in Electrical Engineering


Electrical Engineering

First Advisor

Wei-Jen Lee


As one of the major renewable energy sources, wind power has experienced a fast growth in recent years. The installed capacity of wind farms has increased dramatically. Similar to other generator facilities, their impacts on power system transient stability and power quality should be carefully studied when large scale wind farms are integrated into the power grid. Therefore, it is necessary to establish an accurate dynamic model of large scale wind farm for researches and engineers. A large scale wind farm may have hundreds of WTGs and the structure of WTGs is very complicated and manufacture dependent. If each machine is represented by detailed model, it will aggravate the already existed "dimension disasters" problem and lead to a quite large, high-order and complex system. What's more, it is very difficult to know each subsystem's parameters. Though most models information are provided by the manufacturer, many parameters are always tuned on site. Also, it is often difficult for the utilities to know the operating status of the individual turbines within a farm. Therefore, detailed models of all generators in the farm will be difficult and impractical.Establishing dynamic equivalent model for a wind farm is a viable method for wind farm modeling. Dynamic equivalence model is less detail, yet accurate for dynamic studies, thus can significantly reduce model complexity with the major characteristics retained. This dissertation proposes a hybrid procedure for identifying the dynamic equivalent model parameters of a large scale wind farm. This proposed procedure is based on the newly published generic WTGs models. The generic models are standard, public and not specific to any vendor, so that it can be parameterized in order to reasonably emulate the dynamic behavior of a wide range of equipment.The proposed procedure utilizes a new and intelligent method, particle swarm optimization (PSO), to find an approximate solution of the generic WTGs' parameters in the first step. Then the gradient descent search analysis is applied to find more accurate results by using the solution from the first step as initial condition. The proposed PSO-Gradient Search method can provide the right balance between solution accuracy and computational burden. The dissertation also uses system reduction and key parameter identification approach to reduce the computation burden. Phasor measurement units (PMUs) serve as an on-line data collecting source to record the system response. The proposed procedure is applied on wind farm dynamic equivalence task on both PSS/E SAVNW case and Electric Reliability Council of Texas (ERCOT) system. The encouraging results of these two cases reveal the potential of proposed procedure on identifying the parameters of the dynamic equivalent model of a large scale wind farm.


Electrical and Computer Engineering | Engineering


Degree granted by The University of Texas at Arlington