节点文献
造纸过程先进控制算法及其软件实现
Advanced Control Algorithm and Software Realization on Papermaking Process
【作者】 孔亚广;
【导师】 孙优贤;
【作者基本信息】 浙江大学 , 控制科学与工程, 2002, 博士
【摘要】 在复杂工业过程中经常会遇到多种约束条件、多变量和强耦合、非线性、大时滞、多控制目标等问题。以上这些问题使用常规的控制技术已不能满足其要求。而采用先进控制,可以很好的解决上述问题。如何将先进控制技术应用到生产过程,并产生实际的经济效益,是当前工业自动化研究领域中所必须解决的问题。而随着计算机技术、硬件技术、网络技术的飞速发展,先进控制技术的产业化正逐步实现。 本文结合造纸过程,讨论造纸过程定量水分控制的算法设计和软件实现。充分考虑到造纸过程本身是一个复杂的传热传质过程,干扰多,耦合严重、纯滞后大等特点,利用OPC技术和多变量自学习解耦控制技术实现造纸过程的先进控制软件的设计,详尽阐述了造纸机定量水分专用控制装置的设计方案、特点、相关的软硬件资料和控制策略的具体实施。 本文的主要内容如下: 1、文详细分析了造纸过程先进控制的产生原因和发展过程,全面、系统地概述了造纸过程先进控制的研究现状和研究成果,并指出了理论研究与实际应用中所存在的困难和一些急需解决的问题。 2、造纸过程定量水分专用控制装置的调度、优化、控制等各个环节均给出了一个最新的研究成果。在调度层针对具有随机需求和纸机故障的造纸车间,建立了数学模型,并利用一致化转移方法将其转化为离散动态规划问题。分析了生产两种品种的纸机的最优策略的结构品质,提出了一种简易可行的次优控制策略:阈值控制策略。并给出了求解最优函数的方法。 3、在控制层则对定量水分的解耦控制设计了多变量自学习动态解耦控制方案,并分析了其他常规解耦方案,比较了各方案之间的优缺点,表明多变量自学习动态解耦方法控制效果好、鲁棒性强,适用于不易建立精确11 第一章综述 数学模型且比较复杂的工业过程。同时还给出了一个针对两种性能指标 的最优控制算法:H/V;混合优化控制算法,给出了求解最优控制器的 上逼近算法及其凸二次规划求解方法。 4、针对造纸过程的特点,给出了软件设计方案,介绍了OPC方法,分析了 OPC技术在先进控制软件中的应用方式,并指出如何在DCS 中使用‘。 OPC。同时给出了我们自主开发的高级过程控制算法软件包的实现方法。 5、本文对造纸过程定量水分环节设计了专用控制装置,集管理与控制于一 体,首先分别对造纸过程工艺流程及机理工况进行了分析,充分考虑造 纸过程的特点,采用两种控制方案对系统进行了建模和控制,给出了造 纸过程定量水分控制的综合自动化系统的设计方案。该方案能够根据系 统具体工况的改变实时调整控制决策,从而使得系统具有更快、更强的 适应能力,增加产量、节约原料和提高经济效益。 6、针对造纸过程水分难于测量的特点给出了水分的软测量模型和在线修正 方案。 最后对全文进行了概括性总结,并指出了理论和应用上有待进一步研究的问题。
【Abstract】 n the complex industrial process system, we always meet such problems: multiple constraint conditions; multi variable system; strong coupling; nonlinear system; large time delay; multi control target; and so on. Custom control technique can’t solve above problems. But Advanced control technique can deal well with those problems. How to use advanced control technique in industrial process and generate practical utility is a urgent problem in the industrial automate research realm. According as the galloping of computer technique, hardware technique, network technique, Industrialization of advanced control technique is realizable.In this dissertation, based on papermaking process, we discuss the application of advanced control software and advanced control algorithm in the control of basis weight and moisture of papermaking process. Papermaking system is a complicated transmit heat and character process, has many peculiarity such as disturb, strong coupling, great pure lag etc. We discuss the design of dedicated control equipment of basis weight and moisture for papermaking machine in detail including characteristic, correlative soft and hardware information and realization of control.The main contributions of the dissertation are as follows:1. We make an overview on the papermaking advanced control system in detail, including generation, development and the last achievement in the field. We also make a discussion on some difficulty and problem for advanced process control including theory analysis and application.2. The latest research pay-off are given including scheduling, optimization> control in the papermaking basis weight and moisture dedicated control system. In the scheduling layer, we discussed such papermaking workshop which has stochastic demand and failure. Mathematical model is setup. This problem are inverted into discrete dynamic programming. By exploring the properties of the optimal value function, several simple and realistic sub-optimal policies are presented. These sub-optimal policies are of threshold structure, which are featured only by a number of threshold parameters. The method which used to solve optimal value function are given in this paper.SN In the control layer, We design an algorithm which called multi variable auto learning dynamic decoupling control for basis weight and moisture system. Some general decoupling method are discussed in this paper. From the comparison of each method ,we can jump to conclusion that multi variable auto learning dynamic decoupling method has many properties such as good control effecU strong robust performance. It can be used in the complex industrial process which can’t build a mathematical model. An optimal control algorithm :H2/l1 mixed optimal control which about two index of performance is given. Lower approximation algorithm and its solve of convex quadratic programming are also given in this article.4. Based on papermaking process, we give its software design scheme. OPC is introduced in this paper. We also analysis the application method about how to use OPC in the design of advanced control software. How to use OPC in DCS is also given. We give the implement method of the advanced process control software which developed independently by us.5. Dedicated control system is designed for papermaking basis weight and moisture control. Management and control are integrated in this system. Firstly, crafts flow and mechanism are analyzed. We adopted two control method to model and control for papermaking system. The scheme can real-time rectify control decision-making based on the change of system operating model, so the system can get the following properties: quick and robust adaptive ability; increase turnout; save material; enhance economy utility.6. Based on the performance of measure moisture hardly, we give its soft measurement model and online adjust plan.Finally, we makes a conclusion and proposes the future research directions.
- 【网络出版投稿人】 浙江大学 【网络出版年期】2003年 02期
- 【分类号】TP273.5
- 【被引频次】5
- 【下载频次】926