Wind power prediction system and method

Abstract

The invention provides a wind power prediction system, which comprises a data collection server (1), a database server (2), an application workstation (3), a wind power prediction server (4), a data interface server (5) and reverse physical isolation equipment (6), wherein the data collection server (1) is used for operating data collection software, is communicated with the integrated communication management terminal of a wind power plant, and collects data; the database server (2) processes, carries out statistical analysis and stores the data; the wind power prediction server (4) operatesa wind power prediction module, uses a neural network integrated algorithm based on weighted least squares support vector machine and quantum particle swarm prediction on the basis of a numerical value weather forecast collected or provided by a SCADA (Supervisory Control And Data Acquisition) system, and is combined with the real-time operation working condition of a wind power plant fan to carryout short-term and ultra-short term prediction on the output situation of a single fan and the whole wind power plant; the data interface server (5) is used for obtaining the numerical value weatherforecast; and the reverse physical isolation equipment (6) is used for guaranteeing network safety. The invention also discloses a wind power prediction method, which can guarantee that prediction data which is used in field can embody recent power generation power features.

Description

technical field [0001] The invention belongs to an electric power system, and in particular relates to a wind power prediction system and method. Background technique [0002] Accurate wind power forecasting can help the power grid dispatching department to make a good dispatch plan for various power sources, improve the stability of the power grid operation, improve the ability of the power grid to accommodate wind power, and then reduce the economic losses caused to wind power developers due to power cuts, thereby Increase the return on investment of the wind farm and provide auxiliary means for the management of the wind farm. [0003] Wind power prediction can be classified in many ways. According to the predicted physical quantity, it can be divided into predicting the wind speed first and then predicting the output power according to the wind turbine or wind power curve, and directly predicting the output power; according to the digital model, it can be classified into...

AI Technical Summary

Problems solved by technology

At present, due to the complex terrain, the difficulty of numerical forecasting of wind speed near the ground, the low accuracy of numerical weather forecasting, the late start of forecasting technology research, and the high requirements of forecasting methods for data quality, the current level of wind power forecasting is generally low. According to the National Energy Administration Regarding the requirements of the Interim Measures for the Management of Wind Farm Power Forecasting and Forecasting, the root mean square error of the forecast results for the whole day should be less than 20%.

Images