潜油螺杆泵工作特性分析及在线故障诊断研究

发布时间：2018-01-17 22:15

本文关键词：潜油螺杆泵工作特性分析及在线故障诊断研究　出处：《西安石油大学》2015年硕士论文　论文类型：学位论文

【摘要】：潜油螺杆泵具有能耗低、效率高、占地面积少并且从根本上解决了地面驱动螺杆泵杆断、杆管偏磨的问题,在油田上数量逐年增加,然而,由于缺乏在线实时监控潜油螺杆泵井下机组工况的方法,因而难以及时准确地判断出井下机组的工况,这严重影响了机组的正常生产和使用寿命,大大限制了其在油田上的进一步推广,为了保障潜油螺杆泵安全高效生产、加大推广力度以及适应数字化油田建设的需要,本文开展了潜油螺杆泵工作特性分析及在线故障诊断研究。本文首先分析了潜油螺杆泵工作特性,并在此基础上,运用小波包和RBF神经网络进行了故障诊断,主要内容如下:(1)以潜油螺杆泵系统为研究对象,分析了泵的运动特性、力学特性、排量特性等工作特性,并运用节点分析法建立了潜油电机有功功率和泵功耗之间的数学模型。(2)应用ANSYS软件对潜油螺杆泵进行了有限元分析,得到了不同工况对泵工作特性的影响规律。(3)基于GPRS远程在线传输网络,依据地面驱动螺杆泵采油系统常见故障类型和诊断方法,结合潜油螺杆泵采油系统自身结构特点和工作特性,提出了以潜油电机有功功率为研究对象、小波包和RBF神经网络相结合的潜油螺杆泵在线故障诊断方案。(4)运用小波包提取了潜油电机有功功率信号,搭建了潜油螺杆泵故障样本库,利用Matlab软件建立了RBF神经网络,并确定了网络的训练参数,最终完成了网络的训练和测试。(5)以Visual Basic 6.0为平台,开发了潜油螺杆泵在线故障诊断系统软件,实现了人机交互、在线监测和故障诊断的功能,为潜油螺杆泵安全、高效生产提供了可靠保证。
[Abstract]:Submersible screw pump has low energy consumption, high efficiency, less area, and fundamentally solved the ground driven screw pump rod broken, rod pipe wear problem, the number of oil field increased year by year, however. Due to the lack of on-line real-time monitoring of submersible screw pump downhole operating conditions, it is difficult to determine the working conditions of downhole units in time and accurately, which seriously affects the normal production and service life of the units. In order to ensure the safe and efficient production of the submersible screw pump, to increase the promotion and adapt to the needs of the construction of digital oil field. In this paper, the working characteristics of submersible screw pump and on-line fault diagnosis are studied. Firstly, the working characteristics of submersible screw pump are analyzed, and on the basis of this, the working characteristics of submersible screw pump are analyzed. Wavelet packet and RBF neural network are used for fault diagnosis. The main contents are as follows: 1) taking the submersible screw pump system as the research object, the working characteristics of the pump such as motion, mechanical characteristics, displacement characteristics and so on are analyzed. The mathematical model between active power and pump power consumption of submersible motor is established by using node analysis method. The finite element analysis of submersible screw pump is carried out by using ANSYS software. The influence of different working conditions on pump performance is obtained. Based on GPRS remote online transmission network, common fault types and diagnosis methods of surface driven screw pump oil recovery system are obtained. Combined with the structural characteristics and working characteristics of the submersible screw pump oil recovery system, the active power of the submersible motor is proposed as the research object. Wavelet packet and RBF neural network combined submersible screw pump on-line fault diagnosis scheme. 4) using wavelet packet to extract submersible motor active power signal, build submersible screw pump fault sample database. The RBF neural network is established by using Matlab software, and the training parameters of the network are determined. Finally, the network training and testing. 5) based on Visual Basic 6.0, the online fault diagnosis software of submersible screw pump is developed, and the man-machine interaction is realized. The functions of on-line monitoring and fault diagnosis provide a reliable guarantee for the safety and efficient production of submersible screw pump.
【学位授予单位】：西安石油大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE933.3

【参考文献】