长距离矿浆管道泄漏点检测分析研究

发布时间：2018-07-23 14:09

【摘要】：随着当代工业技术不断向前发展,长距离管道泄漏点检测技术取得了很大的研究成果,此类技术属于这个领域内的主要探索目标之一,但是到现在为止每种泄漏点检测的方法依然不全面完善,还有很多技术难点问题等待解决。本文主要针对的是长距离矿浆管道的泄漏点检测分析研究,针对其复杂的工况、特殊的介质以及准确性和快速性等,对长距离矿浆管道泄漏检测点检测技术进行了全面探索,采用了两种改进后的方法,并用某管道数据验证了方法的有效性。本文主要针对长距离矿浆管道泄漏点检测的准确性问题进行了以下工作：1、对传统的管道泄漏定位模型进行了改进,利用基于GPS和PAC的数据采集系统对声波信号进行同步采集,使信号到达首站控制机的时间差的误差最小；同时分析并确定了不同管道中介质流动的速度对声波波速的影响,利用改进的模型进行理论计算,较传统的定位模型定位精度有一定提高。2、针对长距离矿浆管道系统的噪声源进行了分析确定。并且针对这些噪声,使用了一种基于自适应算法的噪声抑制模型。该模型利用噪声和有用信号之间的相关长度的差异,对泄漏信号中的噪声进行抑制。提高了检测信号的信噪比。3、针对长距离矿浆管道使用了一种噪声分离及泄漏识别的方法。该方法利用盲信号理论下的核独立元分析法以及支持向量机理论下的最小二乘支持向量机的方法对信号进行去噪和泄漏识别。通过仿真研究比较系统的稳定性和准确性更好,可以更快的检测泄漏的发生,增强系统定位的准确性。文中利用某管道的历史数据对管道泄漏点检测系统进行分析研究,通过仿真结果可以表明在通过噪声抑制模型的性能有一定提高,信噪比最大提高6.1740dB。基于KICA和LS-SVM的噪声分离及识别仿真研究中,我们可以清楚的看出该方法能较准确的找到泄漏发生的时间以及对故障的类型进行识别,同时较其他方法在快速性和准确性的比较有一定提高。
[Abstract]:With the development of modern industrial technology, great achievements have been made in long distance pipeline leak detection technology, which is one of the main exploration goals in this field. However, up to now, every leak detection method is still incomplete, and there are still many technical difficulties to be solved. This paper mainly aims at the analysis and analysis of leakage point of long distance slurry pipeline. In view of its complex working condition, special medium, accuracy and rapidity, the leakage detection point detection technology of long distance slurry pipeline is comprehensively explored. Two improved methods are adopted and the validity of the method is verified by a pipeline data. In this paper, aiming at the accuracy of leak detection in long distance slurry pipeline, the following work is done: 1, the traditional pipeline leak location model is improved, and the sound wave signal is collected synchronously by using the data acquisition system based on GPS and PAC. At the same time, the influence of the velocity of medium flow in different pipes on the acoustic wave velocity is analyzed and determined, and the theoretical calculation is carried out by using the improved model. Compared with the traditional positioning model, the positioning accuracy is improved by 0.2. The noise source of the long distance slurry pipeline system is analyzed and determined. For these noises, a noise suppression model based on adaptive algorithm is proposed. The model uses the difference of correlation length between the noise and the useful signal to suppress the noise in the leakage signal. The signal-to-noise ratio (SNR) of the detected signal is improved. A method of noise separation and leakage identification is used for long distance slurry pipeline. In this method, the kernel independent element analysis method based on blind signal theory and the least squares support vector machine method based on support vector machine theory are used to Denoise and identify the signals. Comparing the stability and accuracy of the system with simulation research, the leak detection can be faster and the accuracy of the system location can be enhanced. In this paper, a pipeline leak detection system is analyzed and studied by using the historical data of a pipeline. The simulation results show that the performance of the model can be improved and the signal-to-noise ratio (SNR) can be increased by 6.1740dB. In the simulation research of noise separation and recognition based on KICA and LS-SVM, we can clearly see that the method can find the time of leakage and identify the type of fault. At the same time, compared with other methods in the comparison of speed and accuracy has a certain improvement.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD50

【相似文献】