摘要
在氢燃料发动机试验的基础上,利用RBF神经网络良好的输入输出映射关系,建立起以最佳点火提前角为转速、负荷函数的优化控制数学模型。数字信号处理(DSP)芯片具有快速实时控制的优势,将其用于优化控制氢燃料发动机的点火提前角。基于RBF神经网络进行了仿真并与试验结果进行对比,结果表明:点火提前角最大绝对误差0.41°CA,最大相对误差1.5%。该优化控制模型能准确地获取氢燃料发动机最佳点火提前角。
On the basis of experiment on hydrogen-fueled engine, an advanced ignition angel optimization model, which was the function of rotating speed and loading, was established by the input-output satisfactory relation of radial basis function （RBF） neural network. Digital signal processor （DSP） with quick operation to digital signal was applied to keep optimized control over the advanced ignition angel. Moreover, the calculating results with RBF neural network were contrasted with the experimental results. The results indicated that the optimized control model could accurately forecast the ignition timing of the hydrogen-fueled engine.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2008年第3期19-22,共4页
Transactions of the Chinese Society for Agricultural Machinery
基金
国家自然科学基金资助项目(项目编号:50322262)
河南省科技计划项目(项目编号:0524260028)
河南省高等学校青年骨干教师资助项目(项目编号:教高[2007]335号-88)
关键词
氢燃料
发动机
点火正时
控制优化
径向基函数神经网络
Hydrogen fuel, Engine, Ignition timing, Control optimization, Radial basis function neural network
