多变量改进型预测控制在热力过程与感应电机中的应用研究

发布时间：2018-05-09 05:20

本文选题：锅炉燃烧系统 + PID控制　；参考：《南京信息工程大学》2017年硕士论文

【摘要】：锅炉燃烧系统是典型具有多变量、大滞后以及其他扰动因素特点的热力系统。它设计的合理与否,不仅影响到锅炉整体的效率,甚至影响锅炉的可靠性与经济性。本文采用新型的预测函数控制策略,以状态空间方程为模型,通过对目标指标添加预测误差项,构成了改进型PID结构的目标函数。该控制策略通过多变量预测函数控制与多变量PID控制相结合的预测模型,根据规定的预测模型,使得控制输入量根据其变化而不断进行滚动优化。仿真实验表明:多变量PID预测函数控制器(MPID-PFC)仿真结果分析显示,其应用于锅炉燃烧系统中具有较好的稳定性与较高的效率,拥有较好的实际应用价值。与此同时,不同于热力过程这一类大时滞的慢速系统,感应电机作为典型的快速系统,其是交流驱动系统的典型代表,同时其对于实时性具有很高的要求。本文在对感应电机效率优化问题进行研究的基础上,考虑到传统方法在电机动态时无法同时兼顾响应性能和效率优化的缺陷,本文设计了一种新型分数阶PI(FOPI)预测函数控制策略。该控制策略将预测函数控制和分数阶PI两种算法相结合,构建具有分数阶比例、积分性质的多变量预测函数控制器,兼顾了感应电机动态效率与转速响应速度的优点。应用于电动机效率优化的最大转矩电流比控制方面,采用前馈补偿解耦设计的思路,将系统分解成两个具有可测扰动的子系统。仿真实验表明:新型控制策略具有在线识别模型识别模型参数,跟踪效果好,抗干扰能力强,无超调,稳态误差小,取得了良好的控制效果。
[Abstract]:Boiler combustion system is a typical thermodynamic system with multivariable, large lag and other disturbance factors. Whether the design is reasonable or not not only affects the efficiency of the boiler as a whole, but also affects the reliability and economy of the boiler. In this paper, a new predictive function control strategy is adopted and the state space equation is taken as the model. By adding the prediction error term to the target index, the objective function of the improved PID structure is constructed. The control strategy is based on the predictive model of multivariable predictive function control and multivariable PID control. According to the prescribed prediction model, the control input is continuously rolled and optimized according to its variation. The simulation results show that the multi-variable PID predictive function controller MPID-PFC has good stability and efficiency in boiler combustion system and has good practical application value. At the same time, unlike the slow speed system with large time delay, induction motor is the typical representative of AC drive system, and it has a high requirement for real time. Based on the study of efficiency optimization of induction motor and considering the defect that the traditional method can not simultaneously consider the response performance and efficiency optimization in the dynamic state of induction motor, a new predictive function control strategy of fractional order PII-FOPI is designed in this paper. This control strategy combines predictive function control with fractional Pi algorithm to construct a multivariable predictive function controller with fractional proportional and integral properties, which takes into account the advantages of dynamic efficiency and speed response of induction motor. Applying to the maximum torque / current ratio control of motor efficiency optimization, the system is decomposed into two subsystems with measurable disturbance by the way of feedforward compensation decoupling design. The simulation results show that the new control strategy has the advantages of on-line identification of model parameters, good tracking effect, strong anti-interference ability, no overshoot, small steady-state error and good control effect.
【学位授予单位】：南京信息工程大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273;TM346

【参考文献】