基于支持向量机的发酵过程生物量在线估计

Biomass is a key factor in fermentation process, directly influencing the performance of the fermentation system as well as the quality and yield of the targeted product. Therefore, the on-line estimation of biomass is indispensable. The soft-sensor based on support vector machine （SVM） for an on-line biomass estimation was analyzed in detail, and the improved SVM called the weighted least squares support vector machine was presented to follow the dynamic feature of fermentation process. The model based on the modified SVM was developed and demonstrated using simulation experiments.     

面向可持续发展的半导体制造的集成产品和过程控制

Semiconductor fabrication is a manufacturing sequence with hundreds of sophisticated unit operations and it is always challenged by strategy development for ensuring the yield of defect-free products. In this paper, an advanced control strategy through integrating product and process control is established. The proposed multiscale scheme contains three layers for coordinated equipment control, process control and product quality control. In the upper layer, online control performance assessment is applied to reduce the quality variation and maximize the overall product performance (OPP). It serves as supervisory control to update the recipe of the process controller in the middle layer. The process controller is designed as an exponentially weighted moving average (EWMA) run-to-run controller to reject disturbances, such as process shift, drift and tool worn out, that are exerted to the op-eration. The equipment in the process is individually controlled to maintain its optimal operational status and maximize the overall equipment effectiveness (OEE), based on the set point given by the process controller. The ef-ficacy of the proposed integrated control scheme is demonstrated through case studies, where both the OPP (for product) and the OEE (for equipment) are enhanced.     

基于自适应神经模糊推理系统的间歇反应精馏组分估算(英文)

Composition estimation plays very important role in plant operation and control. Extended Kalman filter (EKF) is one of the most common estimators, which has been used in composition estimation of reactive batch distillation, but its performance is heavily dependent on the thermodynamic modeling of vapor-liquid equilibrium, which is difficult to initialize and tune. In this paper an inferential state estimation scheme based on adaptive neuro-fuzzy inference system (ANFIS), which is a model base estimator, is employed for composition estimation by using temperature measurements in multicomponent reactive batch distillation. The state estimator is supported by data from a complete dynamic model that includes component and energy balance equations accompanied with thermodynamic relations and reaction kinetics. The mathematical model is verified by pilot plant data. The simulation results show that the ANFIS estimator provides reliable and accurate estimation for component concentrations in reactive batch distillation. The estimated states form a basis for improving the performance of reactive batch distillation either through decision making of an operator or through an automatic closed-loop control scheme.     

间歇冷却结晶过程的模拟

The general mathematical model for batch cooling crystallization was established based on the population balance equation considering the change of slurry volume, and simulated with crystallization thermodynamics,kinetics and mass balance employing bed voidage. In the system of vitamin C-water-ethanol, reliability of this model was verified by comparison between simulation results and experimental data. The effects of operation parameters on product quality can be systematically investigated by modeling simulation.     

高纯度精馏过程的集成控制与在线优化策略

For high-purity distillation processes, it is difficult to achieve a good direct product quality control using traditional pro-portional-integral-differential (PID) control or multivariable predictive control technique due to some difficulties, such as long re-sponse time, many un-measurable disturbances, and the reliability and precision issues of product quality soft-sensors. In this paper, based on the first principle analysis and dynamic simulation of a distillation process, a new predictive control scheme is proposed by using the split ratio of distillate flow rate to that of bottoms as an essential controlled variable. Correspondingly, a new strategy with integrated control and on-line optimization is developed, which consists of model predictive control of the split ratio, surrogate model based on radial basis function neural network for optimization, and modified differential evolution optimization algorithm. With the strategy, the process achieves its steady state quickly, so more profit can be obtained. The proposed strategy has been successfully applied to a gas separation plant for more than three years, which shows that the strategy is feasible and effective.     

基于混合建模技术的复合肥养分含量MIMO软测量模型

In compound fertilizer production, several quality variables need to be monitored and controlled simultaneously. It is very diifficult to measure these variables on-line by existing instruments and sensors. So, soft-sensor technique becomes an indispensable method to implement real-time quality control. In this article, a new model of multi-inputs multi-outputs （MIMO） soft-sensor, which is constructed based on hybrid modeling technique, is proposed for these interactional variables. Data-driven modeling method and simplified first principle modelingmethod are combined in this model. Data-driven modeling method based on limited memory partial least squares（LM-PLS） al.gorithm is used to build soft-senor models for some secondary variables.then, the simplified first principle model is used to compute three primary variables on line. The proposed model has been used in practicalprocess; the results indicate that the proposed model is precise and efficient, and it is possible to realize on line quality control for compound fertilizer process.     

氯代烷烃对气相缩合法合成苯基三氯硅烷的影响(英文)

To enhance the process of phenyltrichlorosilane synthesis using gas phase condensation, a series of chloralkanes were introduced. The influence of temperature and chloralkane amount on the synthesis was studied based on the product distribution from a tubular reactor. The promoting effect of chloralkane addition was mainly caused by the chloralkane radicals generated by the dissociation of C–Cl bond. The promoting effect of the chloromethane with more chlorine atoms was better than those with less chlorine atoms. Intermediates detected from the reactions with isoprene and bromobenzene demonstrated that both trichlorosilyl radical and dichlorosilylene existed in the reaction system in the presence of chloralkanes. A detailed reaction scheme was proposed.     

MSMPR反应结晶器中地塞米松磷酸钠晶体生长模型的研究

The reactive crystallization process of dexamethasone sodium phosphate was investigated in a continuous mixed-suspension, mixed-product-removal(MSMPR) crystallizer. Analyzing experimental data, it was found that the growth of product crystal was size-dependent. The Bransom, CR, ASL, M J2 and M J3 size-dependent growth models were discussed in details. Using experimental steady state population density data of dexamethasone sodium phosphate, parameters of five size-dependent growth models were determined by the method of non-linear least-squares. By comparison of experimental population density and linear growth rate data with those obtained from the five size-dependent growth models, it was found that the MJ3 model predicts the growth more accurately than do the other four models. Based on the theory of population balance, the crystal nucleation and growth rate equations of dexamethasone sodium phosphate were determined by non-linear regression method. The effects of different operation parameters such as supersaturation, magma density and temperature on the quality of product crystal were also discussed, and the optimal operation conditions were derived.     

工业流化床反应器结块监视的动态PCA方法

Dynamic principal component analysis (DPCA) is an extension of conventional principal component analysis (PCA) for dealing with multivariate dynamic data serially correlated in time. Based on the fact that the measured variables in relation to chunk monitoring of the industrial fluidized-bed reactor are highly cross-correlated and auto-correlated, this paper presents a practical strategy for chunk monitoring by adopting DPCA in order to overcome the shortcomings of the conventional method. After introducing the basic principle of DPCA, both how to determine the time lagged length of data matrix and how to calculate the nonparametric control limits when the dynamic data are not subject to the assumption of independently identically distribution (IID) were discussed. An appropriate DPCA model based on the real data from a industrial fluidized-bed reactor was built, with parallel analysis and empirical reference distribution (ERD) method to select time lagged length and control limits, respectively. During data pretreatment, data smoothing was used to reduce noise and the serial correlations to some degree. The simulation test results showed the effectiveness of the DPCA based method.     

延迟焦化加热炉的自适应状态反馈预测控制和专家控制

An adaptive state feedback predictive control （SFPC） scheme and an expert control scheme are presented and applied to the temperature control of a 1200 kt·a^-1 delayed coking furnace, which is the key equipment for the delayed coking process. Adaptive SFPC is used to improve the performance of temperature control in normal operation. A simplified nonlinear model on the basis of first principles of the furnace is developed to obtain a state space model by linearization. Taking advantage of the nonlinear model, an online model adapting method is presented to accommodate the dynamic change of process characteristics because of tube coking and load changes. To compensate the large inverse response of outlet temperature resulting from the sudden increase of injected steam of a particular velocity to tubes, a monitoring method and an expert control scheme based on heat balance calculation are proposed. Industrial implementation shows the effectiveness and feasibility of the proposed control strategy.     

基于神经网络的软测量技术在精馏塔上的应用

针对扬子石化公司丁二烯精馏塔原控制系统存在的问题, 利用从集散控制系统(DCS)采集的大量现场数据和用机理模型得到的模拟数据, 运用前向反馈(BP)神经网络软测量技术，构造了产品丁二烯和总炔含量的自适应软测量仪表, 设计了一套控制系统. 实际监测数据表明, 这套控制系统可实现产品质量的闭环控制.     

基于改进的FasBack神经模糊系统的4-CBA软测量模型研究

提出一种基于改进的FasBack神经模糊系统的新型对羧基苯甲醛(4-CBA)软测量模型,用Leven-berg-M arquardt算法训练模型中的部分参数,经实际过程数据验证表明,提出的模型学习速度快、预测精度高、鲁棒性强,为实现精对苯二甲酸(PTA)生产过程中4-CBA含量的实时、精确控制提供了一条有效的途径。     

An experimental application of the inferential control scheme to the binary distillation column control

The inferential control scheme based on a linear estimator was applied to control of top composition of a pilot scale packed distillation column. Experimental and simulation studies were used to evaluate the inferential control scheme and to compare its performance with a conventional single temperature feedback control with proportional plus integral actions. Both digital simulation and experimental verification showed that the top composition control achieved with the inferential control scheme was superior to that achieved with the conventional control scheme.     

水泥熟料质量指标的软测量建模研究

针对水泥熟料质量指标的测量,提出一种基于最小二乘支持向量机的软测量建模方法;对于建模数据,提出了基于模糊聚类的数据预处理方法.实验研究表明,该数据预处理方法明显优于传统的拉依达准则方法,能够有效地去除现场测量数据中存在的异常数据;最小二乘支持向量机建模相比于RBF神经网络也具有明显优势,建立的软测量模型对于整个窑系统优化控制具有重要意义.     

EXPERIMENTAL EVALUATION OF THE INFERENTIAL SYSTEM FOR DISTILLATION CONTROL

An inferential control scheme has been applied to control the bottoms—product composition of a laboratory-scale n-propanol-methanol, distillation column. The closed-loop performance of the inferential control system is compared with the conventional proportional-integral control system. The application of inferential control resulted in improved response for disturbances in feed composition and feed flow rate.     

EXPERIMENTAL EVALUATION OF THE INFERENTIAL SYSTEM FOR DISTILLATION CONTROL

An inferential control scheme has been applied to control the bottoms—product composition of a laboratory-scale n-propanol-methanol, distillation column. The closed-loop performance of the inferential control system is compared with the conventional proportional-integral control system. The application of inferential control resulted in improved response for disturbances in feed composition and feed flow rate.     

Recursive data-based prediction and control of product quality for a PMMA batch process

In many batch processes, frequent process/feedstock disturbances and unavailability of direct on-line quality measurements make it very difficult to achieve tight control of product quality. Motivated by this, we present a simple data-based method in which measurements of other process variables are related to end product quality using a historical data base. The developed correlation model is used to make on-line predictions of end quality, which can serve as a basis for adjusting the batch condition/time so that desired product quality may be achieved. This strategy is applied to a methyl methacrylate (MMA) polymerization process. Important end quality variables, the weight average molecular weight and the polydispersity, are predicted recursively based on the measurements of reactor cooling rate. Subsequently, a shrinking-horizon model predictive control approach is used to manipulate the reaction temperature. The results in this study show promise for the proposed inferential control method.     

一种基于混合建模技术的MIMO软测量建模方法

针对复合肥产品中几种养分含量需要同时预报的一类多输入/多输出(MIMO)软测量建模问题,提出一种基于混合建模方法的复合肥养分含量MIMO软测量模型。该混合模型首先对几个不能实时测量的关键辅助变量采用基于限定记忆部分最小二乘算法的数据驱动建模方法建立自适应软测量模型,然后采用简化机理模型实时计算三种养分含量。基于实际工业过程数据的仿真结果表明,所建模型运算速度快、预测精度高,可以满足复合肥养分含量在线预报的要求。     

基于工况识别的注塑过程产品质量预测方法

针对多工况注塑过程的在线质量预测问题,考虑了过程数据高维、耦合、非线性等特点,采用拉普拉斯特征映射(LE)方法实现过程数据的非线性降维;在低维特征空间中采用Mean Shift聚类算法完成样本的工况聚类,以便注塑过程的工况分析和知识挖掘;同时运用Mean Shift原理,提出一种新样本的在线工况识别方法;最后应用基于混合粒子群(PSO)参数寻优的偏最小二乘支持向量机(PLS-LSSVM)方法,建立了多工况注塑过程的产品质量软测量模型。实验结果表明,相较于PLS-LSSVM方法,本文方法的预测精度和泛化性能均有明显提高,可为实际注塑企业提供一种效果良好的多工况产品质量在线预测方法。     

密炼机的推断控制系统

提出橡胶密炼机的一种基于神经网络软测量的推断控制系统,针对橡胶密炼机混烧胶的质量指标需要比较精确估计的一类软测量建模问题,研究了达用于MISO软测量模型的一种RBF逄法,并在此基础上实现了密炼机混炼胶粘度的推断控制。