Research On Unit Commitment Problem Considering The Irregular Distribution Of Wind Power

As an economic and eco-friendly energy, wind power penetration in power systems has been increasing significantly in recent years. However, the uncertainty of wind power brings great challenge to power system. Nowadays, the forecast accuracy of wind power varies in a large range. However, the present power system operations are mainly based on those thermal power plants, which reduces the regulating ability of power system. As a result, in order to ensure the safety of power system, wind power will be cut down when it reaches peak value. The wind farm will be controlled to reduce its output, which will lead the wind power probabilistic distribution to be irregular. Unit commitment (UC) is a basic problem of power system researches. With the uncertainties increasing, the schedule of UC problem becomes increasingly complex. Under this background, the research on modeling and solving of UC problem has important theoretical significance and practical value. The main research contents and results are shown as follows:1. Based on traditional distribution model of wind power, the paper put forwards an irregular probabilistic distribution model of wind power considering the self-control abilities of wind farms.2. Reliable scenarios is an important guarantee for the accuracy of scenario UC model. Currently, a large number of researches about scenarios generation focus on single wind farm, and the research about multi-wind farms is rather less. By comprehensive analysis about the researches on scenarios generation, the paper proposes a new scenarios generation method of multi-wind farms.3. The paper builds the UC model considering the irregular distribution of wind power, and it contains two categories of scenarios:wind power compute scenario and wind power error scenarios. The traditional generators keep the same ON/OFF status under those scenarios. In order to improve the convergence speed of UC problem, the UC model is decomposed into three subproblem based on benders decomposition algorithm (BD-UC model) including:UC master model, security child model and scenarios child model. Through analysis about the impact of wind power irregular distribution to BD-UC model, the paper proposes a new method to choose compute scenario which can also accelerate the solving speed of UC model.4. Through analysis of main factors affecting the efficiency of BD-UC model, the paper proposes two methods to optimize the BD-UC model, which can further improve the convergence speed of UC model. Method one focuses on the reduction of iterative times by the adjustment of the wind scenarios order. Method two focuses on the reduction of computation time of each subproblem by the scale curtailment of subproblems. Both methods have strategic significance to UC model.