混沌遗传算法的神经网络在输油管道泄漏中的应用

发布时间：2018-03-04 18:42

本文选题：输油管道　切入点：特征提取　出处：《东北石油大学》2015年硕士论文　论文类型：学位论文

【摘要】：在当今的管道运输工业中,伴随着它的飞速发展,随之而来的是更多的环境被泄漏的石油所污染,同时造成能源的浪费和不可估量的经济损失。在最近几年中,大部分地区的管道出现不同程度的老化和锈蚀现象,加之一些不法分子的盗油事件发生,导致泄漏事故频发。传统的泄漏检测有着一定得误差,并且效率相对也低,所以一套有效的自动管道检测定位系统显得十分的重要,管道泄漏检测是一门涉及广泛的课题,有传感技术,信号检测,信号处理等学科。本文系统地归纳了国内外在输油管道泄漏检测方面的研究状况,较为详细的阐述了管道泄漏定位的工作原理、难点及解决办法;小波包算法以及神经网络技术,通过小波包变换实现压力信号消噪处理,同时从而提取特征向量作为神经网络的输入。为了提高对管道泄漏检测的准确性,本文提出了一种以混沌遗传算法为基础的BP神经网络的改进算法。BP神经网络应用广泛但是却容易陷入极小值,收敛速度相对也慢,经过改进后的算法是以混沌遗传算法为基础,优化BP神经网络的初始权值和阈值,并将混沌的遍历性和遗传算法的反演性等优点相结合。将混沌变量加入到遗传算法中,进一步的提高遗传算法的全局搜索能力和收敛的速度;混沌遗传算法优化以后的最优解作为BP神经网络的初始权值和阈值。利用改进后的CGA-BP算法进行管道泄漏预测,结果表明,该方法对管道泄漏的工况的分类效果明显好于单纯使用BP神经网络。现如今,单一的泄漏检测往往不能够满足我们的需求,在实际的应用中将多种方法相结合能够得到更为准确的提出系统的软硬件和总体设计。该系统通过管道两端的压力传感器采集到的信息,进行数据传输实现及时报警与定位的功能。
[Abstract]:In today's pipeline transportation industry, with its rapid development, more and more environment is polluted by leaking oil, which results in energy waste and incalculable economic loss in recent years. In most areas, pipeline aging and corrosion to varying degrees, coupled with the oil theft incidents of some illegal elements, lead to frequent leakage accidents. Traditional leak detection has certain errors, and the efficiency is relatively low. Therefore, a set of effective automatic pipeline detection and positioning system is very important. Pipeline leakage detection is a wide range of topics, including sensing technology, signal detection, This paper systematically summarizes the research status of oil pipeline leakage detection at home and abroad, and expounds the working principle, difficulties and solutions of pipeline leakage location in detail. In order to improve the accuracy of pipeline leakage detection, wavelet packet algorithm and neural network technology can realize the noise reduction of pressure signal by wavelet packet transform, and extract the eigenvector as the input of neural network, in order to improve the accuracy of pipeline leakage detection. In this paper, an improved BP neural network algorithm based on chaotic genetic algorithm is proposed. The BP neural network is widely used, but it is easy to fall into the minimum value, and the convergence rate is relatively slow. The improved algorithm is based on the chaotic genetic algorithm. The initial weights and thresholds of BP neural network are optimized, and the ergodicity of chaos is combined with the inversion of genetic algorithm. The global searching ability and convergence speed of genetic algorithm are further improved, the optimal solution of chaos genetic algorithm is used as the initial weight and threshold of BP neural network. The improved CGA-BP algorithm is used to predict pipeline leakage. The classification effect of this method is obviously better than that of using BP neural network alone. Nowadays, a single leak detection method can not meet our needs. In practical application, the software, hardware and overall design of the system can be more accurately proposed by combining various methods. The information collected by the pressure sensor at both ends of the pipeline can be obtained by the system. Data transmission to achieve timely alarm and positioning function.
【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE88;TP183

【相似文献】