基于Hopfield神经网络的谐波检测及DSP实现

Harmonic Detecting Based on Hopfield Neural Network and Realization on DSP

研究一种基于Hopfield神经网络的谐波检测方法,并且利用TMS320F2812实现该算法的训练学习。分析该算法检测的理论基础,阐述基于DSP实现的软硬件设计以及与上位机通信。结果表明,该方法具有很好的实时性、较高的检测精度以及自适应跟踪负载电流变化的能力。

Study on an approach of harmonic detection based on Hopfield neural network （HNN）. And realize its training algorithm on TMS320F2812. Analyze the approach＇s theoretical basis for harmonic detection and the software and hardware design of the controller based on DSP and the communication with PC. Results show the accuracy and practicability of the approach and the ability of adaptive change with load current.