Wind Power Prediction Based On Xinjiang Wind Operating Rules

Wind’s volatility and intermittent shortcomings, make the large-scale wind power connected will bring great challenges to the stability of power system and the normal scheduling.In order to improve the ability to coordinate wind power farm with power system, help the development of wind power sustained, stable and healthy, need a deeper research to the wind power forecasting. In this paper, based on the similarity with the wind the windward area of wind data is of great significance to the wind power forecast, proposed Xinjiang wind operation rules, and carries on the scientific demonstration. According to Xinjiang wind operation rules, and the improved particle swarm optimization automatic optimizing of parameters support vector machine, built Xinjiang wind operation rules with improved particle swarm optimization support vector machine wind prediction model and use Xinjiang measured data verifies it’s feasibility.The main innovation points of this paper are as follows:According to the prediction model of support vector machine, Analysis of the function of punishment vector and kernel function parameter and their effect on the support vector machine regression prediction, then establish the improved particle swarm optimization model to automatic optimize the kernel function parameter and penalty vector. Use the examples to verify that the accuracy of the improved particle swarm optimization support vector machine is better than the existing support vector machines in the 15 min short-term wind power.We put forward the theory of Xinjiang wind operation rules, with the analysis of Xinjiang Dabancheng, grass lake,thirteen rooms wind area wind data, established the model of Xinjiang wind operation rules, and summary the system of Xinjiang wind operation rules. built Xinjiang wind operation rules with improved particle swarm optimization support vector machine wind prediction model, and use the examples to verify the model can effectively improve the prediction accuracy of the wind forecast in 15 min and 24 h.