基于隐式广义预测控制算法的LNG出口压力控制系统研究

发布时间：2018-08-18 21:20

【摘要】：在现代工业和日常生活中LNG出口压力控制系统具有非常广泛的应用。伴随着工业的不断发展,人们对于LNG出口压力控制的要求也越来越高。然而,就工业现场的压力控制而言,是具有多变、复杂特点的过程,同时,在实际工业现场中,会有来自不同方面因素对LNG出口压力控制进行影响,再加上压力控制对象普遍具有非线性、大惯性、扰动因素多、大时滞等特点,这就会对压力控制算法提出了更高的标准和要求。随着科技的发展,出现了越来越多的先进控制技术。由于它能够克服传统控制技术的不足,因此被广泛地应用到工业控制中。本文将LNG出口压力作为研究对象,通过对广义预测控制算法进行研究,在此基础上,引出了通过直接辨识控制律参数的方式,对算法进行计算,使得此控制算法的实时性更高、快速性更强,称之为隐式广义预测控制算法,并建立了基于隐式广义预测控制算法的LNG出口压力控制模型,通过MATLAB仿真对比了当前PID算法和隐式广义预测算法对于压力控制的效果,还建立了以STM32为下位机的MCU的最小系统以及必要的外围电路设计,设计了压力采集电路、比例电磁阀驱动电路,同时利用c#和visualstudio-2010设计了lng出口压力控制上位机系统,方便了用户对于压力的显示和控制。论文的主要研究工作和创新点如下:1.隐式广义预测控制算法的分析和研究。本文针对一般的广义预测控制算法,需要重复对diophantine方程进行在线推导、求解,同时也要多次对矩阵进行求逆运算,这样增加了计算量,同时也使计算变得复杂;假设对系统的原模型参数进行直接辨识,同样需要重复对diophantine方程进行在线推导、求解,若想算出控制律,必须对控制器的参数进行求解,这样也会对计算时间有所加长,系统控制的实时性也会有所降低,考虑到以上因素严重制约了广义预测控制算法的实际应用,因此本文介绍了直接辨识控制律参数e、g、s。而不用对diophantine方程在线反复推理、求解,对于计算时间的减少具有明显的帮助,在此基础上发展了隐式广义预测控制算法。2.基于隐式广义预测控制算法的lng出口压力控制系统建模和研究。由于lng压力控制具有时变性、不确定性、大滞后性的特点,不仅仅是控制对象的不确定性、时变性、大滞后性,同时还具有储罐容量大、较大波动的用气量、小时供气量不匀称性等特点,且在压力控制过程中,lng会有状态的变化(液态转化为气态),因此现在用简单的pid算法已经无法达到控制的要求,利用隐式广义预测控制算法的carima模型,以及将lng压力控制抽象成一阶线性滞后环节,构建了基于隐式广义预测控制算法的lng出口压力控制框图和隐式广义预测控制器。通过仿真发现隐式广义预测控制对于压力控制的快速性、稳定性、鲁棒性会更加完美。3.LNG出口压力控制系统的硬件设计。基于隐式广义预测控制算法对本系统进行设计。在硬件上采用Cortex-M3为内核的STM32微处理器来作为嵌入式控制器的核心CPU,搭建了STM32最小系统,在此基础上,设计了RS-485通信接口电路、压力检测电路、比例电磁阀驱动电路,并采用MODBUS-RTU协议作为通信协议,为了实现各模块间的相互通信和信息传输的稳定性,在原有协议的基础上,针对本系统的需求,设计了控制协议的格式,方便上下位机信息交流。同时也方便了LNG出口压力进行实时数据显示、数据处理。4.LNG出口压力控制系统的软件设计。主要运用Visual Studio-2010开发平台及C#语言进行开发,通过参考其它控制系统设计的经验以及操作人员的使用习惯,设计了LNG出口压力控制上位机系统,其主要包括系统设置界面、退出系统界面、参数设置界面、报表参数界面、用户设置界面、设备设置界面等,实现对LNG出口压力控制系统下位机的监测、控制等功能。同时,上位机也能完成各种复杂的数据处理。
[Abstract]:LNG outlet pressure control system is widely used in modern industry and daily life. With the continuous development of industry, people have higher and higher requirements for LNG outlet pressure control. With the development of science and technology, more and more advanced control technologies have emerged, because of its ability to control the pressure of LNG outlet, which has the characteristics of non-linearity, large inertia, many disturbance factors and large time delay. In this paper, the LNG outlet pressure is taken as the research object, and the generalized predictive control algorithm is studied. On this basis, the method of directly identifying the parameters of the control law is proposed to calculate the algorithm, which makes the algorithm more real-time and fast. It is called implicit generalized predictive control algorithm, and the LNG outlet pressure control model based on implicit generalized predictive control algorithm is established. The effect of current PID algorithm and implicit generalized predictive control algorithm on pressure control is compared by MATLAB simulation. The minimum system of MCU with STM32 as its subordinate computer and its necessary are also established. Peripheral circuit design, design pressure acquisition circuit, proportional solenoid valve drive circuit, at the same time use C # and visual studio-2010 design of LNG outlet pressure control system, convenient for users to display and control the pressure. The main research work and innovation of this paper are as follows: 1. Implicit generalized predictive control algorithm analysis and research. For general generalized predictive control (GPC) algorithms, the Diophantine equation needs to be deduced and solved on-line repeatedly, and the matrix must be inverted many times at the same time. If you want to calculate the control law, you must solve the parameters of the controller. This will also lengthen the calculation time and reduce the real-time performance of the system control. Considering the above factors seriously restrict the practical application of the generalized predictive control algorithm, this paper introduces the direct identification of the control law parameters e, g, s. The implicit generalized predictive control algorithm is developed on this basis. 2. Modeling and research of LNG outlet pressure control system based on implicit generalized predictive control algorithm. Because LNG pressure control has time-varying, uncertain and large delay Posterior characteristics, not only the uncertainty of the control object, time-varying, large lag, but also has the characteristics of large tank capacity, large fluctuations in the amount of gas, the amount of gas supply is not symmetrical, and in the process of pressure control, LNG will change state (liquid into gas), so now using a simple PID algorithm has been unable to achieve In order to satisfy the control requirements, an implicit generalized predictive control algorithm based on Carima model and an implicit generalized predictive control algorithm based on the implicit generalized predictive control algorithm are used to abstract the LNG pressure control into a first-order linear hysteresis link. The hardware design of the LNG outlet pressure control system is based on the implicit generalized predictive control algorithm. The STM32 microprocessor with Cortex-M3 as the kernel is used as the core CPU of the embedded controller, and the STM32 minimal system is built. On this basis, the RS-M32 minimal system is designed. 485 communication interface circuit, pressure detection circuit, proportional solenoid valve drive circuit, and the use of MODBUS-RTU protocol as a communication protocol, in order to achieve the stability of communication and information transmission between the various modules, on the basis of the original protocol, according to the needs of the system, the control protocol format is designed to facilitate the exchange of information between the upper and lower computer. The software design of LNG outlet pressure control system is mainly developed by using Visual Studio-2010 development platform and C# language. By referring to the experience of other control system design and the usage habits of operators, the upper computer of LNG outlet pressure control is designed. The system mainly includes system setting interface, exit system interface, parameter setting interface, report parameter interface, user setting interface, equipment setting interface, etc. It can realize the monitoring and control functions of the lower computer of LNG outlet pressure control system.
【学位授予单位】：东华大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP273

【相似文献】