摘要
根据空间矢量脉宽调制(SVPWM)原理,提出一种基于人工神经网络(ANN)的SVPWM新算法。该网络系统结构分布化,利用神经网络的并行处理能力,将复杂的计算任务分散化,缩短了计算时间。在新的设计环境SystemGenerator/Simulink下建立了SVPWM的算法模型,并进行了硬件仿真,将模型下载到FP-GA芯片,产生了SVPWM波形,实验结果表明了该方法的有效性和可行性。
According to the principle of space-vector PWM,a novel SVPWM algorithm based on artificial neural network(ANN) is proposed. The structure of the network is divided into several parts to separate the calculating tasks, this way makes the most of the fast parallel calculating ability of ANN. The calculation time is shortened. The model is built up in a new method with System Generator/Simulink,and simulated. Then the model is downloaded to FPGA platform,the experiment results indicate the validity and feasibility.
出处
《电气传动》
北大核心
2006年第12期38-42,共5页
Electric Drive
基金
江苏省高技术基金资助项目(BG2005014)
关键词
空间矢量脉宽调制
人工神经网络
现场可编程门陈列
电机系统
space vector pulse width modulation(SVPWM) artificial neural network(ANN) field pro grammable gate array(FPGA) system generator (SG)