本文系统地研究了在约束条件可调整的情况下CMMO问题中可行性与目标协调的关系，论述了当系统优化不可行时，在进行软约束调整的过程中要兼顾系统的期望目标，以获得满意的优化结果。本文运用混合逻辑的方法来描述软约束调整的优先级，并将多目标协调问题转化为逻辑约束满足问题，从而系统地解决了稳态优化中软约束调整与目标 协调的问题，并以壳牌重油分馏塔标准问题为例，进行了仿真，仿真结果表明了本文算法的有效性。

本文充分考虑大多数复杂热工控制对象非线性特性与运行工况密切相关的实际特点，采用多模型动态矩阵控制方法。并将该方法应用于某电厂300MW机组锅炉过热汽温对象，在典型工况下通过试验数据获得其局部三阶子模型集，基于每个局部子模型分别设计子DMC控制器。通过跟踪实际工况变化来对子控制器加权以获得合适的控制增量。实验结果表明该方法对参数突变适应快，可取得令人满意的控制效果。

This paper is concerned with air handling units (AHUs), the performances of which directly influence those of heating, ventilation and air conditioning systems. An autotuning recedinghorizon optimization method is proposed to synthesize a proportional-integral-derivative (PID) type controller for AHUs. This algorithm is composed of two levels of control. The lower level adopts a conventional PID controller to obtain an acceptable, but not necessarily optimal, performance. The higher level provides optimal low-level controller parameters through minimization of the generalized predictive control criterion. Because the method does not require changes in hardware and the definitions of conventional controller parameters, it can be both easily accepted by process engineers and widely applied to industrial areas. Compared with the performance of a well-tuned conventional PID controller, simulation and experimental results show that the proposed method for AHU systems can achieve a better performance under a wide range of operating conditions.

Performance assessment of control loops has got the attention of many researchers. This study focuses on performance assessment of model predictive control. An MPC-achievable benchmark is proposed based on subspace identification; the advantage of this benchmark is that only system input and output data is used and the state space model need not explicitly calculated during identification, in addition, it can be implemented to both univariate and multivariate process. Two performance measures can be constructed to assess the control loop, which can evaluate the potential benefit to update the new identified model. To find the potential benefit by tuning the parameter, tradeoff curves similar to LQG benchmark can be draw as reference. The effect of constraints imposed on process variables can be evaluated by the installed controller benchmark. Simulation about an industrial example was done using the proposed method.

Knowledge-based control tries to integrate the knowledge of human operators or process engineers into the controller design. Fuzzy control, one of the most popular intelligent techniques, has been successfully applied to a large number of consumer products and industrial processes. Model predictive control (MPC) has been used in process control systems with constraints; however, the constrained optimization problem involved in control systems has generally been solved in practice in a piece-meal fashion. To solve the problem systemically, the multi-objective fuzzyoptimization (MOFO) is used in the constrained predictive control for online applications as a means of dealing with fuzzy goals and fuzzy constraints in control systems. The conventional MPC is integrated with the techniques from fuzzy multicriteria decision making, translating the goals and the constraints to predictive control in a transparent way. This paper investigates how to use the fuzzy goal programming in predictive control and how to use the fuzzy goals and fuzzy constraints in predictive control. The presented method allows for a more flexible aggregation of the control objectives than the usual weighting sum of squared errors. It is shown that the model predictive controller based on MOFO allows the designers a more flexible aggregation of the control objectives than the usual weighting sum of squared errors in MPC. The visual robot path planning validates the efficiency of the presented algorithm.