Soft sensor has been extensively applied for online estimation of the key quality variables in modern industrial processes, which is extremely important for the process to achieve efficient monitoring and smooth control. To build an accurate data-driven model, the dynamic correlation and strong nonlinearity of process sequential data must be considered in soft sensor modeling. Therefore, a dynamic feature regression network (DFR) is proposed in this paper for industrial soft sensor modeling, consisting of a dynamic feature extraction network and a feature regression network to explore different industrial data features. First, the dynamic feature extraction network maps time series samples to a set of hidden dynamic features, while the feature regression network performs deeper feature extraction and output regression on key quality variables. Furthermore, since both networks consist of unsupervised feature extraction and supervised feature regression, a large scale of unlabeled samples can be utilized in the semisupervised learning of model parameters. Finally, the feasibility and efficacy of the proposed model are verified through the coal-to-ethylene glycol process data to predict the carbon monoxide content. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

本文针对工业过程中最常用的间歇补料反应过程,利用间歇过程的重复性和二维动态特性,以二维系统理论为基础,二维系统的状态空间模型,分析迭代学习控制的二维特性。将传统的PI控制器作为内环结构,使用迭代学习控制器优化其参考轨迹,得到基于二维系统的迭代学习PI控制方法（ILC-PI方法）,并通过稳定性分析确定控制器参数选择方式。将ILC-PI控制方法应用于一类多批次间歇补料反应过程中,结果表明,虽然初始收敛速率较低,但是随着批次不断优化,最终的跟踪性能较好。

A synergetic speed controller（SSC） is proposed for the multi-motor system of triple brushless DC（BLDC） motors with the relative coupling structure in order to improve the coordination and robustness.To achieve the desired synchronous control objective,an augmented dynamic model of the BLDC motor is firstly developed.Based on the synergetic control theory,a manifold is then constructed for the ith motor as the linear combination of the phase current,the speed tracking error and its integration,and the coordination error as well as its integration.The control law of the SSC is further deduced from the augmented model and the manifold.The effects of controller parameters on the synchronization performance and their selection principle are also discussed.Simulative results on the triple-BLDC motor system with relative coupling show that the SSC can provide better synchronization and robustness,by comparing with the conventional PID relative coupling control and the sliding mode control（SMC） relative coupling control.

The dynamic control allocation algorithm is proposed for a class of distributed control systems with hierarchical structure, in order to solve the real-time cooperative control problem in the case of sensor failure occurred in the subsystems. The augmented state-space model of the distributed control system is first established for the control allocation design, by using the dynamic model of sensor faults. According to robust control theory, the constrained and convex optimization is then formulated into linear-matrix-inequality form to design the gain matrix of the control allotter, which can be computed by various well-developed algorithms. Thus the suppression of sensor faults in the subsystem is realized and the expected cooperative control objective is achieved. Simulations on the four-corner leveling control system of an advanced hydraulic press demonstrate the effectiveness of the proposed control allocation algorithm. © 2018 Technical Committee on Control Theory, Chinese Association of Automation.

For the process of biogas pressurized water scrubbing, design and simulation of plant-level dynamic control system is researched based on the optimum structure and parameters of the steady state in this paper. According to the process requirements of the biogas pressurized water scrubbing pilot-scale plant, a steady state simulation system is established using Aspen Plus. The simulation research under different operation conditions is carried out to obtain the optimal operating parameters. Then plant-level dynamic control schemes and the dynamic simulation system are also designed to ensure the efficiency of biogas purification. The system's performance is tested under introducing biogas inlet flow disturbance and intake flow component disturbance. The simulation results show that the system has good dynamic response performance, such as the removal rate of CO2 is greater than 99.8% under various disturbances. (C) 2018, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

An enhanced non-dominated sorting genetic algorithm（NSGA-II） was applied to carry out the integer optimized design in this article.The steady simulation design was established based on the reaction kinetics analysis using Aspen Plus,and the feasible region of each optimized variable was calculated through sensitive analysis.Aiming at the total cost TAC as the objective function,the NSGA-II algorithm was used directly to obtain the Pareto optimal solutions under the constraints of feasible region of variables.The simulation results showed that the NSGA-II algorithm can effectively reduce the total cost of TAC.

本文针对醋酸甲酯水解生产工艺优化设计问题,采用改进的非支配遗传算法（NSGA-II算法）开展厂级多变量经济优化设计。在反应动力学分析基础上进行Aspen Plus稳态模拟设计,通过灵敏分析计算出每个优化变量的可行域;以年总成本TAC为优化目标函数、优化变量可行域为约束,运用非支配遗传算法NSGA-II直接对Aspen Plus中反应精馏塔和甲醇回收塔的稳态模型进行多变量优化计算,将优化结果和序贯优化结果相比较,NSGA-II多变量同步优化算法更易寻找到全局最优解,大大降低了生产总投资成本。

本文针对工业过程中最常用的间歇补料反应过程,利用间歇过程的重复性和二维动态特性,以二维系统理论为基础,二维系统的状态空间模型,分析迭代学习控制的二维特性。将传统的PI控制器作为内环结构,使用迭代学习控制器优化其参考轨迹,得到基于二维系统的迭代学习PI控制方法（ILC-PI方法）,并通过稳定性分析确定控制器参数选择方式。将ILC-PI控制方法应用于一类多批次间歇补料反应过程中,结果表明,虽然初始收敛速率较低,但是随着批次不断优化,最终的跟踪性能较好。