基于模糊神经网络的点火提前角时间差软测量模型  

Soft-Measuring Model of Time Difference about Spark-Ignition Advanced Angle Based on Fuzzy Control and Neural Networks Theory

作　　者：谢常清[1] 鄂加强[1] 成志明[1] 龚金科[1] 蒋寿生[1,2] 袁文华[1,2] 

机构地区：[1]湖南大学机械与运载工程学院,长沙410082 [2]邵阳学院机械与能源工程系,邵阳422000

出　　处：《内燃机工程》2009年第2期73-77,共5页Chinese Internal Combustion Engine Engineering

基　　金：“985工程”二期--高效低排放发动机先进设计制造技术科技创新平台资助

摘　　要：基于模糊神经网络的理论方法,以点火提前角、发动机的功率和转速等特征参数作为二次变量,提出了点火提前角时间差软测量模型。仿真结果表明,软测量模型具有非常好的映射性能。试验结果表明,利用软测量模型对点火提前角进行控制时,该模型能在连续工作120 h内很好地实现成功点火,以及在冷起动条件下保证一次性点火成功,并且能满足排放性能要求。Based on the fuzzy neural networks theory,a soft-measuring model of time difference about spark-ignition advanced angle was proposed by using the spark-ignition advanced angle,engine power and speed as well as other related parameters.The simulation results show that this soft-measuring model is well mapped.The experimental results show that when the soft-measuring model is used for the spark-ignition advanced angle controlling,the successful ignition can be realized for a 120h continuous operation and the successful ignition in one spark can be achieved in cold start condition,and the engine emission can also meet the emission limit.

关 键 词：内燃机 模糊神经网络 点火提前角 软测量模型 

分 类 号：TK413.9[动力工程及工程热物理—动力机械及工程]