牙轮钻机钻进控制研究

发布时间：2018-02-03 08:47

本文关键词： 牙轮钻机钻进控制 BP神经网络PID 仿真分析　出处：《吉林大学》2017年硕士论文　论文类型：学位论文

【摘要】：钻进系统是牙轮钻机的主要组成部分,钻机性能的好坏很大程度上取决于钻进系统的性能,钻进系统主要完成两个动作:钻杆的给进-提升动作和钻杆的回转动作,这两个动作分别对应钻进系统的两个液压控制回路,即给进-提升回路和回转回路,对牙轮钻机钻进的控制就是对这两个液压回路的控制。本文首先对牙轮钻机钻进系统的两个液压回路进行了分析,并且对液压回路常用的两种调速形式节流调速和容积调速进行了对比分析,通过对比选用了功率损失小、效率高的容积调速形式作为钻进系统液压回路的调速形式,并用AMESim液压仿真软件画出了两个液压回路的原理图,计算了液压回路的部分性能参数。钻机钻进控制系统的主要控制对象就是钻进系统的液压回路,通过液压回路控制钻杆的运动实现钻进,为了对液压回路进行控制仿真,本文中对液压回路的各液压元件进行了原理及其相互作用分析,建立了其简化的数学模型,并用Simulink建立了回路整体数学模型,用AMESim液压仿真软件建立了加压回路与回转回路的物理仿真模型,分别用两种模型对回路进行仿真分析,对比两种模型的仿真分析结果,验证模型的正确性。钻进系统液压回路在开环状态时稳定性较差,并且在负载突变时震荡较明显,需要进行闭环控制,本文选用了神经网络PID控制算法,相比于传统PID控制器,神经网络PID控制器不仅继承了其结构简单的优点,还能实现参数的自整定,并且具有较强的学习能力力,能够边工作边根据传感器反馈的地层数据进行学习,使钻进控制更好的适应地层变化。用Simulink控制仿真软件分别搭建了传统PID控制器和神经网络PID控制器的模型,并且分别用这两控制器的模型与用AMESim搭建的液压回路模型进行了联合仿真,仿真结果表明神经网络PID控制下钻进系统的性能更好。
[Abstract]:Drilling system is the main part of rotary drilling rig. The performance of drilling rig depends on the performance of drilling system to a great extent. Drilling system mainly completes two actions: the feeding and lifting of the drill pipe and the rotary movement of the drill pipe. These two actions correspond to the two hydraulic control circuits of the drilling system, namely, the feed-lift loop and the rotary loop respectively. The control of drilling wheel drilling is the control of these two hydraulic circuits. Firstly, two hydraulic circuits of drilling system of rotary drilling machine are analyzed in this paper. And the hydraulic circuit commonly used two types of throttling speed regulation and volume speed regulation were compared and analyzed, through the comparison of the choice of small power loss. The high efficiency volumetric speed regulation form is used as the speed regulation form of hydraulic circuit of drilling system, and the schematic diagram of two hydraulic circuits is drawn with AMESim hydraulic simulation software. The main control object of drilling control system is the hydraulic circuit of drilling system, which controls the movement of drill pipe to realize drilling. In order to control and simulate the hydraulic circuit, the principle and interaction of the hydraulic components in the hydraulic circuit are analyzed in this paper, and its simplified mathematical model is established. The whole mathematical model of the loop is established with Simulink, and the physical simulation model of the pressurized loop and the rotary loop is established by using the AMESim hydraulic simulation software. The simulation results of the two models are compared to verify the correctness of the model. The hydraulic loop of drilling system has poor stability in open loop state. And the vibration is obvious in the sudden change of load, so the closed-loop control is needed. In this paper, the neural network PID control algorithm is used, compared with the traditional PID controller. Neural network PID controller not only inherits the advantages of simple structure, but also realizes the self-tuning of parameters, and has a strong learning ability. Ability to learn from formation data feedback from sensors while working. The model of traditional PID controller and neural network PID controller is built by using Simulink control simulation software. The simulation results show that the performance of the drilling system under the control of neural network PID is better.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TD422.1

【相似文献】