Research On Optimal Power Flow Of Power System With Large-scale Wind Farm Integration

The fast increasing of wind power installed capacity and large-scale wind power integration bring new challenges to power system scheduling and operation. It is essential for scheduling staffs to consider about how to make full use of wind energy,meantime ensuring safety and reliable power supply. As the large magnitude of the modern power system,the scheduling and operation is complex. Scheduling staffs need computer-aided tools such as optimal power flow to analyze and compute power system.Compared with the traditional energy sources, the biggest feature of wind power is uncertainty. It is a form of intermittent and random energy. At present, most optimal power flow(OPF) program is based on certainty principle, but it is difficult to deal with uncertainty caused by large-scale wind power integration. There are two trains of thought,one is to enhance the accuracy of wind power prediction,the other is to consider the uncertainty’s influence into model.Based on uncertainty programming theory, this paper starts with the second thought to study the model and algorithm of the large-scale wind power integration OPF and lay a good foundation to improve scheduling and operation level.First, this paper introduces the theory of uncertainty, and the two main programming models in uncertainty programming:chance-constrained programming and expectation model and their solution.It lays a good foundation to study the large-scale wind power integration OPF. Focuse on the double-fed induction generator(DFIG), this paper analyzes its operating principle and static mathematical model, and then studies the wind farms’ steady-state equivalent model. Based on the above model, the paper establishes the chance-constrained programming based OPF model and solve the model with particle swarm optimal method.Numerical testing on IEEE30bus system shows the confidence level of constraints have a significant impact to the final optimization results in chance-constrained programming OPF model. Therefore, we should set appropriate confidence level according to the specific constraints in practical applications. Considering the increasing of the wind power prediction accuracy, the paper proposes a concept of wind power forecast error cost, and establishes a OPF model which takes the cost of wind power forecast error into account(WFECI-OPF).A fast approximate solution method is proposed. Simulation examples validate that the proposed method has high accuracy and fast calculation speed, so it can meets the requirements in practical engineering applications.