Short-term power prediction method applied to dispersed wind power

Abstract

The invention discloses a short-term power prediction method applied to dispersed wind power, and belongs to the technical field of wind power fields. The prediction method comprises the following steps: (1) acquiring data required by power prediction of wind power fields in a specified area; (2) grouping the wind power fields according to the correlation among the output power of each wind power field; (3) selecting each group of representative wind power fields and building a real-time wind measuring tower; (4) partitioning wind direction sectors according to wind direction frequency distribution, building a wind power field power prediction model representing each sector, and calculating the power prediction result of the representative wind power field in each sector; (5) building an extrapolation model, and concluding the short-term output power of non-representative wind power fields according to the power prediction results of the representative wind power fields. In the short-term power prediction method, models of correlation grouping modeling and sector modeling are adopted instead of a single prediction model, so that the modeling time is shortened; the short-term output power of the non-representative wind power fields is extrapolated by adopting an extrapolation prediction method, so that repeated calculation of similar wind power field power prediction is avoided, the prediction efficiency is increased greatly, and real-time dispatching of a power grid is facilitated.

Description

technical field [0001] The invention belongs to the technical field of wind farms, in particular to a short-term power prediction method for distributed wind power. Background technique [0002] Distributed wind power refers to a wind power project that is located near the power load center and does not aim at large-scale long-distance power transmission. The generated power is connected to the grid nearby and consumed locally. At the end of 2011, the National Energy Administration issued the "Guiding Opinions on the Development and Construction of Distributed Access Wind Power Projects" (Guoneng Xinneng [2011] No. 374), which clearly shows the state's attitude towards encouraging decentralized wind power development. The operation of decentralized wind power connected to the grid will become another mainstream access method for subsequent wind power connection. [0003] In remote areas with sudden increase in load, installing distributed wind power on site can avoid a larg...

AI Technical Summary

Problems solved by technology

[0005] 1) Existing methods mostly use the NWP results at the anemometer tower to map the output of the entire wind farm. In the case of a vast wind farm area, complex terrain or scattered distribution of wind turbines, the representativeness of the location of the anemometer tower is reduced, which affects the prediction accuracy

[0006] 2) Currently, a single model is mostly used, which cannot be applied to areas with more complicated situations

For example, factors such as climate, terrain, wake, and unit type all have a great influence on the output characteristics of wind farms, and a single model cannot cover the influence of various factors, resulting in low accuracy of prediction results

[0007] 3) Computational efficiency needs to be further improved to meet the needs of real-time scheduling

NWP (Numerical Weather Prediction) calculations and CFD calculations are time-consuming, and model building is difficult, requiring high personnel and equipment

Images