| 遗传神经网络算法在应用中的优化策略研究 |
| |
| 引用本文： | 张蕾,张文明.遗传神经网络算法在应用中的优化策略研究[J].中国农机化,2006(1):77-79. |
| |
| 作者姓名： | 张蕾 张文明 |
| |
| 作者单位： | 1. 天津工程师范学院
2. 北京科技大学 |
| |
| 摘 要： | 通过分析废气的HC、CO、CO2和O2的含量,可以快速诊断发动机的常见故障。本文设计了神经网络诊断系统,通过融合分析废气含量、发动机转速、氧传感器及其它一些信息诊断发动机故障。本文设计的方法主要应用遗传算法的复制、交换、变异过程代替BP网络的反向传播过程,并对遗传算法进行了改进研究。实践证明这种基于遗传神经网络方法的故障诊断系统具有收敛速度快、推广性能强的特点,极大提高了发动机故障诊断系统的效率和准确性。
|
| 关 键 词： | 废气 神经网络 遗传算法 故障诊断 |
| 文章编号： | 1006-7205（2006）01-0077-03 |
| 收稿时间： | 2005-05-17 |
| 修稿时间： | 2005年5月17日 |
Optimization Strategies in Application Based on GA-BP Algorithm |
| |
| ZHANG Lei,ZHANG Wen-ming.Optimization Strategies in Application Based on GA-BP Algorithm[J].Chinese Agricul Tural Mechanization,2006(1):77-79. |
| |
| Authors: | ZHANG Lei ZHANG Wen-ming |
| |
| Institution: | 1.Automotive Engineering Department, Tianjin University of Technology and Education, Tianjin, 300222, Chain; 2.Civil and Environmental Engineering School, University of Science and Technology Beijing, Beijing, 100083, Chain |
| |
| Abstract: | Engine faults can be diagnosed by analyzing HC, CO, C02 and 02 of exhaust gas. For fault diagnosis system, BP-GA (Back Propagation and Genetic Algorithm) is designed by analyzing exhaust gas, engine rotate speed, oxygen sensor and some fault appearance. GA is also adopted and improved, which replaces the back propagation of BP network by reproducing, exchanging and varying. The convergence and generalization ability of the genetic and neural network system is improved remarkably and the accuracy and the efficiency of engine diagnosis system are added. |
| |
| Keywords: | exhaust gas neural network genetic algorithm fault diagnosis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
