| 基于DCPSO的模糊神经网络的管道泄漏检测方法 |
| |
| 引用本文： | 李炜,张美玲,李庆卿. 基于DCPSO的模糊神经网络的管道泄漏检测方法[J]. 工业仪表与自动化装置, 2010, 0(6): 3-7 |
| |
| 作者姓名： | 李炜 张美玲 李庆卿 |
| |
| 作者单位： | 兰州理工大学,电气工程与信息工程学院,兰州,730050;甘肃第一安装工程有限公司,兰州,730060 |
| |
| 基金项目： | 教育部春晖计划项目,兰州理工大学特色学术梯队基金 |
| |
| 摘 要： | 提出了一种基于发散-收敛PSO(DCPSO)优化模糊神经网络的管道泄漏检测与估计方法。该方法采用广义概率积、广义概率和模糊算子代替普通神经网络中的传递函数,并用DCPSO算法对该模糊神经网络的权值进行优化。通过实际管道泄漏数据对网络进行仿真研究,结果表明文中所述方法在管道泄漏的检测与估计中,不仅比BP算法具有更快的收敛速度,其结果也更优,进而也昭示出该方法在管道泄漏检测与估计中的可用性。
|
| 关 键 词： | 管道 泄漏检测与估计 广义概率积 广义概率和 模糊神经网络 DCPSO优化算法 |
Study on pipeline leak detection methods of fuzzy neural network based on DCPSO optimization |
| |
| LI Wei,ZHANG Meiling,LI Qingqing. Study on pipeline leak detection methods of fuzzy neural network based on DCPSO optimization[J]. Industrial Instrumentation & Automation, 2010, 0(6): 3-7 |
| |
| Authors: | LI Wei ZHANG Meiling LI Qingqing |
| |
| Affiliation: | 1.College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050,China;2.The First Installation Work Limited Company of Gansu,Lanzhou 730060,China) |
| |
| Abstract: | This paper puts forward a kind of new method applied in pipeline leak detection and estimation.That is fuzzy neural network based on divergent-convergent PSO（DCPSO） optimization algorithm.This method used the generalized probability and the probability and generalized fuzzy operator to substitute for the transfer function of neural network,and used DCPSO optimization algorithm to optimize the fuzzy neural network weights.Through the actual pipeline leak of network data simulation,indicated that this method avoided the defects of slow in network training and easy in the local superior of the BP neural network in the pipeline leak detection and estimation,and verified the validity of this method. |
| |
| Keywords: | pipeline leakage detection and estimate general probability probability and generalized fuzzy neural network DCPSO optimization algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
