【作者】 邵剑；

【导师】 金晓明；

【作者基本信息】 浙江大学 ， 控制理论与控制工程， 2012， 硕士

【摘要】 本文针对纯碱煅烧过程的时变、大时滞、多变量强耦合等复杂过程特性,在单台煅烧炉预测控制的基础上,提出了基于经济指标优化的煅烧过程MPC优化控制策略,并进一步针对实际煅烧生产过程多台煅烧炉并联运行的情况提出了生产负荷变化的智能MPC优化控制策略。主要研究工作包括：1、简单介绍了纯碱煅烧过程工艺和设备的发展过程,重点分析了煅烧过程的工艺原理、控制特点,总结了国内外纯碱煅烧过程控制策略研究和应用现状。2、通过工业现场调研和工艺资料的分析,阐述了纯碱煅烧过程的主要控制指标,研究了纯碱煅烧过程常规控制策略的原理和特点,分析了常规控制策略实施中存在的主要问题,并在此基础上提出了在煅烧过程实施先进控制的必要性。3、针对单台煅烧炉纯碱煅烧过程的动态模型与优化目标,将多变量DMC优化控制策略应用于煅烧过程中,并通过MATLAB仿真验证。仿真结果表明,该优化控制策略能够将出碱温度、出气温度等关键工艺指标维持在正常范围内,保证产品质量,达到较理想的控制效果。然后,在上述控制策略的基础上提出了基于经济指标优化的多变量模型预测控制优化控制策略,即在滚动优化时先进行经济指标最小化的稳态优化再进行动态优化,仿真结果表明,该优化控制策略不仅能够实现煅烧过程关键工艺指标的控制和优化,保证了产品质量,还能实现“卡边”优化,降低了能耗。4、分析工业现场遇到的各种工艺状况,针对实际工业现场多台煅烧炉并联工作的控制和优化问题,在单台煅烧炉多变量模型预测控制的基础上,结合模糊控制、专家控制策略,提出了一种针对生产工况变化动态分配湿重碱进料量的智能MPC优化控制策略。仿真结果表明,该优化控制策略在故障诊断和炉况监视的支持下,能够根据各煅烧炉的工况动态分配进碱量,在保证产品质量的前提下,实现总进碱量的整体优化。最后,将该优化控制策略应用到某纯碱厂轻灰煅烧工序项目中,取得了较好的控制效果。更多还原

【Abstract】 Soda calcination processes invariably emerge complicated characteristics such as time-varying, large time-delay and strong-coupling, etc. In this thesis, a MPC optimization strategy is studied for the single-calciner MPC system. Furthermore, an intelligent MPC optimization strategy for the real parallel multi-calciner MPC system is proposed.The main contributions of the research work are summarized as follows:1. The basic principles and equipments of soda calcination process are briefly introduced, followed which the principles and characteristics of the process are mainly analyzed. In addition, the state of art of soda calcination process is summarized.2. According to the investigation of practical industrial application and theoretical analysis, the main control objects of soda calcinations are presented. The shortage of conventional control strategies is analyzed based on the study of the process’s principals and characteristics. Consequently, the necessity of implementation of advanced process control (APC) strategy is demonstrated.3. A multivariate DMC optimization control strategy is designed for a single-calciner process based on its dynamic model and optimization objects. This strategy is further validated through MATLAB simulation examples, which illustrates the effectiveness and superior performance of the proposed approach. According to the simulation results, the key indicators such as the temperature of outlet soda and outlet air can be successfully controlled in the normal range. Afterwards, a multivariate MPC optimization control strategy is proposed on the basis of the above mentioned one. Simulation results indicate that the new strategy can not only keep the key variables in the normal range but also realize the bonder control. This advantage can reduce the energy consumption and finally bring benefit for the company.4. In practical applications, the soda calcinations process is parallel of multi-calciner. Besides, of the varying production states in real cases should be considered. Hence, a new intelligent MPC optimization control strategy is proposed combined with fuzzy control and expert control technologies. A mechanism of dynamic distribution of the feed of wet soda according to different conditions is invented in the new strategy. Simulation results state that this strategy can optimize the whole feed of soda on condition of good products quality. The good performance is also validated by the implementation to the light soda calcination process in a soda plant.更多还原