Wind power system modeling and DSP (Digital Signal Processor) realizing method based on neural network

Abstract

The invention relates to a wind power system modeling and DSP (Digital Signal Processor) realizing method based on a neural network, comprising the following steps of: determining input and output signals by analyzing working mechanisms of a wind power system and the neural network, wherein the input signal includes wind speed and a propeller pitch angle, the output signal includes power, wind wheel rotating speed and a wind wheel torque. The invention can achieve any training accuracy to determine the weight value of each layer by combining the BP neural network with the wind power system, establishing a BP neural network model and setting a hidden layer number to be large enough, thereby well fitting the property of a modeling object; in order to realize the application possibility, the invention designs a neural network real-time simulation system realized by adopting DSP as a neural network computation coprocessor (SEED-C30PS); in addition, the invention realizes the BP neural network used for the parameters of the wind power system on a DSP system according to fast arithmetic capability determined by the specific hardware structure of a DSP chip, thereby enhancing the arithmetic speed of the system and enabling the system to meet the requirements for instantaneity.

Description

【Technical field】: [0001] The invention belongs to the fields of wind power generation, intelligent control, power electronics technology and digital signal processing technology. It specifically relates to a neural network-based wind farm modeling method and DSP implementation. The typical BP neural network is used in complex and nonlinear wind power systems. 【Background technique】: [0002] With the continuous introduction of China's renewable energy stimulus policies and the improvement of power grid supporting construction, China's huge wind power development potential will be further stimulated, and China has the ability to become a world's largest renewable energy country. From the characteristics of wind power generation itself, it can be seen that the magnitude and direction of natural wind speed are random and uncontrollable, so the wind energy generated by acting on the blades of the wind turbine is also random and uncontrollable; The complex mechanism makes the ...

AI Technical Summary

Problems solved by technology

Due to the complexity of the wind power generation system, there is a certain deviation between the established mechanism and the actual

Images