Fuzzy neural ignition timing control for a natural gas fuelled spark ignition engine
Fuzzy neural ignition timing control for a natural gas fuelled spark ignition engine
复制标题
天然气火花点火发动机的模糊神经点火正时控制
DOI：
10.1243/0954407011528833
复制
发表时间：
2001
期刊：
影响因子：
--
通讯作者：
C. Nwagboso
中科院分区：
文献类型：
--
作者：
W. Wang;E. Chirwa;E. Zhou;K. Holmes;C. Nwagboso
摘要 点火正时是影响几乎所有发动机输出的火花点火发动机的重要输入之一。众所周知，对于给定的发动机设计，给出最大制动扭矩的最佳点火正时随着气缸内火焰发展和传播的速率而变化。现代发动机显示点火正时通常由作为发动机速度和负载的函数的固定开环时间表控制。期望的是，该点火正时可以调节至产生最佳扭矩的最佳水平，以获得最小的燃料消耗和最大的可用功率。本文提出了一种基于模糊逻辑和神经网络理论的点火正时控制系统。开发了一种光纤传感器系统，用于测量发光强度，该系统将福特 1600 cm3 四缸天然气火花点火发动机的燃烧压力和点火正时控制关联起来。为了优化燃烧强度检测系统，进行了多次发动机测试。获得的结果为使用模糊神经控制技术的发动机智能控制提供了重要信息。而且，使用该技术进行的数据测试显示出良好的结果，与原始发动机的输出扭矩特性非常吻合。
Abstract One of the important inputs to a spark ignition engine which affects nearly all engine outputs is ignition timing. It is well known that the optimum ignition timing which gives the maximum brake torque for a given engine design varies with the rate of flame development and propagation in the cylinder. Modern engines show ignition timing being generally controlled by fixed open-loop schedules as functions of engine speed and load. It is desirable that this ignition timing can be adjusted to the optimum level which produces the best torque to obtain minimum fuel consumption and maximum available power. This paper presents an ignition timing control system based on fuzzy logic and neural network theories. A fibre optical sensor system was developed for measurement of the intensity of the luminous emission which correlates the combustion pressure and ignition timing control on a Ford 1600 cm3 four-cylinder spark ignition engine fuelled with natural gas. Several engine tests were carried out in optimizing the combustion intensity detection system. The results obtained provide important information compatible with intelligent control of the engine using fuzzy neural control technology. Moreover, tests carried out with data using this technology show good results that fit quite well with the original engine output torque characteristics.