变时滞系统的辨识自适应控制算法的改进与仿真研究

城市供水出水浊度过程控制是大时滞难控对象,Smith预估控制方法的研究有效地改善了时滞带来的控制困难的问题,但是当系统模型不精确时,很难获得好的控制品质。采用时滞辨识自适应方法,有好的控制效果,但当被控系统的时滞参数是时变时,在一定范围内可能存在的时滞辨识的效果不理想,所以提出采用变搜寻区域的辨识方法,可以实时地搜索出真实过程的时滞。该算法用于变时滞的系统,具有良好的控制品质和较强的自适应能力。     

大时滞过程控制方法及应用

论述了时滞系统的各种控制方法的原理和优缺点,从最早的Smith预估控制到新近发展的各种智能控制技术以及它们的融合,指出时滞系统的控制仍然是控制领域中的研究热点,特别是基于无参数模型的时滞过程控制的研究更是方兴未艾。可以认为,智能控制技术和传统的滞后控制方法的结合对时滞过程的控制是有效的,可望在工业实践中得到应用。     

一种改进的Smith预估控制系统

提出一种改进的Smith预估器 ,在常规Smith预估器的基础上增加了两个控制器 ,内环控制器用来稳定积分和不稳定过程 ,反馈控制器对干扰进行抑制 ,并把主控制器设为设定值加权PI控制器。仿真结果表明 ,该方案不仅可行 ,而且在模型失配的条件下 ,控制品质比常规的Smith预估控制有很大的改善。     

USE OF RTD AND LOCAL TRANSFER REPRESENTATION TO OBTAIN A TRANSIENT MODELLING OF A LIQUID-SOLID SYSTEM

The input-output thermal transient modelling of a multistage liquid-solid fluidized bed is proposed. This model is based on a flow modelling obtained from RTD measurements and a local heat transfer model. Experimental data are well represented by the model. Furthermore, sensitivity problems are studied and literature values of heat transfer coefficients are discussed.     

ROBUST CLOSED LOOP LOG MODULUS DESIGN OF SISO SMITH PREDICTORS FOR A SPECIAL CLASS OF PROCESSES

Linear open loop stable or prestabilized SISO processes with uncertain or changing time delays are controlled via a Smith Predictor. The processes under consideration have no other nonminimum phase characteristics than the time delay. Further, the process models have an inherent integration constant of zero or one. A controller design procedure is presented which guarantees a robust closed loop system, where the major tuning parameter of the controller is a nonlinear function of the maximum closed loop Log Modulus and the model-plant mismatch.     

Bulk Density,Porosity and Resistance to Airflow of Garlic Slices

ABSTRACT

The bulk density of garlic slices at different moisture levels (ranging from approximately 3 to 65% MC wet basis) was determined by weighing the contents of a container of known volume. The porosity was calculated using its relationship with bulk and apparent densities. An analysis of variance (ANOVA) revealed that bulk density and porosity were affected significantly by moisture and slice thickness as well as the interaction of these variables. Bulk density varied in a positive linear fashion with moisture and thickness while a negative linear correlation was found for the calculated bulk porosity. The linear model met the adequacy criterion for characterising the behaviour of garlic. Using a laboratory unit, the vertical resistance to airflow through the product and the effect of misture and slice thickness were investigated for an airflow rates of 0.09 to 1.2 m₃/s-m₂. A higher resistance to airflow was noted for the wet product with the experimental data fitting to Shedd's model when the aifflow range was divided into two sub-flow rates and to a modified Ergun's equation for the full range.     

INSTRUCTIONS FOR PREPARATION OF MANUSCRIPTS FOR DIRECT REPRODUCTION

ABSTRACT

The bulk density of garlic slices at different moisture levels (ranging from approximately 3 to 65% MC wet basis) was determined by weighing the contents of a container of known volume. The porosity was calculated using its relationship with bulk and apparent densities. An analysis of variance (ANOVA) revealed that bulk density and porosity were affected significantly by moisture and slice thickness as well as the interaction of these variables. Bulk density varied in a positive linear fashion with moisture and thickness while a negative linear correlation was found for the calculated bulk porosity. The linear model met me adequacy criterion for characterising the behaviour of garlic. Using a laboratory unit, the vertical resistance to airflow through the product and die effect of moisture and slice thickness were investigated for an airflow rates of 0.09 to 1.2 m³/s-m². A higher resistance to airflow was noted for the wet product with the experimental data fitting to Shedd's model when the airflow range was divided into two sub-flow rates and to a modified Ergun's equation for the full range.     

DRYING CHARACTERISTICS AND PRODUCT QUALITY OF OKRA

ABSTRACT

Effects of size (whole and sliced), pre-treatment (blanching in water and 0.5% NaCl solution at 95° C) and temperature of drying air (40, 55 and 70° C) on the drying characteristics and quality of okra were studied. Estimation of the drying rate established that the drying of okra takes place under the falling rate period. Page's model was found to adequately describe the drying behavior of okra over a wide range of drying conditions used in the study. The coefficients of Page's model were correlated with air temperature and it was found that the dependence of the rate constant on the drying air temperature can be described using the Arrhenius law. The quality of the dried product was found to be best when okra was sliced and blanched at 95° C in 0.5% NaCl solution for 5 min and then dried at 55° C.     

Predictions in time Series Using Multivariate Regression Models

Predictions in time series using multivariate regression models are studied with respect to their mean squared errors. Two new methods of prediction are proposed: the simple one and the method based on the kriging theory. The mean squared errors of these predictions are computed and it is shown that the first one can be regarded as a special case of the kriging approach.     

基于smith预估的水泥分解炉温度MRFAC控制器设计

针对水泥分解炉温度控制系统具有大滞后、模型不确定等特性,采用MRFAC(模型参考模糊自调整控制)的方法设计了其控制器。该方法在普通模糊控制器的基础上,增加了利用参考模型输出和实际被控对象输出之差,来在线修改常规模糊控制器规则库的辅助模糊控制器,从而提高了系统对参数不确定性的鲁棒性。为了削弱系统大滞后特性的不利影响,系统中还引入了基于神经网络的smith预估器,用其输出代替实际系统输出来进行反馈控制。仿真结果表明,该控制系统能有效抑制大惯性、大滞后的影响,具有较好的鲁棒性和动、静态性能,获得了满意的控制效果。     

Book Reviews

Books reviewed:
Robert Yaffee with Monnie McGee, Time Series Analysis and Forecasting
Robert J. Adler, Raisa E. Feldman, and Murad S. Taqqu, (eds.) APractical Guide to Heavy Tails     

MULTIRATE MULTIVARIABLE SELF-TUNING CONTROL OF DISTILLATION COLUMNS USING KALMAN PREDICTOR

A novel method for designing Kalman Predictor (KP) based multivariable self-tuning controllers for Multi Input Multi Output (MIMO) systems has been proposed, and also applied to the control of two distillation columns. The objective is to maintain constant terminal compositions despite disturbances entering the system even when the controlled variables are not measured at the same sampling rate. The KP generates minimum variance estimates of the output variables. Simulation results show that KP based multirate multivariable self-tuning controller exhibits better performance than the earlier reported multirate controller for set point tracking, even in the presence of nonstationary disturbances. KP based self-tuning controller is, therefore, suitable for industrial applications.     

暖通空调系统基于SMITH预估自校正控制算法

目前暖通空调控制系统中普遍存在运行效率低下、能耗浪费严重的现象,主要原因在于暖通系统本身具有时变、时滞等非线性特性,使得传统的控制方法无法取得良好的控制性能。如何克服暖通空调系统时变和时滞特性所带来的不利影响,是提高暖通空调系统控制回路性能的重要途径之一。本文设计了一种基于SMITH预估的自校正控制算法,首先通过在线辨识包含时滞参数在内的暖通空调对象模型,进而采用SMITH预估器对时滞进行补偿,而SMITH预估器中的PI控制器则用于提高鲁棒性,并实现控制器参数的实时更新。仿真结果表明该算法性能要优于传统的控制方法。     

Evaluation of Simple Control Strategies for Rotary Dryers

Rotary dryers are very versatile process equipment as it can accommodate a large variety of products over a wide range of flow rates. To obtain a uniform product quality and to operate the dryer economically with minimal environmental impact, it is necessary to use efficient control strategies. In this article, two simple control strategies are proposed to improve the feedback-only controller that is normally used. One strategy is a partial feedforward control loop addition to the feedback loop and the second is a moisture predictor that allows taking control actions on the estimates of the discharge moisture content rather than waiting to measure its value at the discharge. The performance of these two strategies was compared with the performance of a feedback-only controller and an ideal feedback-feedforward controller. It is shown that the two proposed control strategies significantly improve the performance of a feedback-only control strategy without having to install new sensors.     

串级控制与史密斯预估相结合在窑炉通路玻璃液温度控制中的应用

针对大滞后被控对象,介绍了一种串级PID控制器,并将史密斯预估补偿方法用于窑炉通路温度控制。     

Evaluation of Simple Control Strategies for Rotary Dryers

Abstract

Rotary dryers are very versatile process equipment as it can accommodate a large variety of products over a wide range of flow rates. To obtain a uniform product quality and to operate the dryer economically with minimal environmental impact, it is necessary to use efficient control strategies. In this article, two simple control strategies are proposed to improve the feedback-only controller that is normally used. One strategy is a partial feedforward control loop addition to the feedback loop and the second is a moisture predictor that allows taking control actions on the estimates of the discharge moisture content rather than waiting to measure its value at the discharge. The performance of these two strategies was compared with the performance of a feedback-only controller and an ideal feedback-feedforward controller. It is shown that the two proposed control strategies significantly improve the performance of a feedback-only control strategy without having to install new sensors.