This paper presents the analysis and design of an original compact distillation system suitable to produce a high purity product from a mixture containing very small amounts of light and heavy impurities (typically be...
详细信息
ISBN:
(纸本)0852953933
This paper presents the analysis and design of an original compact distillation system suitable to produce a high purity product from a mixture containing very small amounts of light and heavy impurities (typically below 0.5%). The proposed set-up, called the Flash Compact System, uses a pre-flash/heater coupled to a single distillation column with side stream product. Separation performance and energy consumption of this set-up are compared against a direct and indirect two column sequence, as well as with the Petlyuk system. Dynamic properties of the design are evaluated in term of controllability and by closed loop dynamic simulation. The results demonstrate that this system is feasible for industrial applications.
Operator's models are attracting the interest of industry as they relate to monitoring and control of complex engineered processes. Such processes are usually relying upon the expertise and ability of human operat...
详细信息
Operator's models are attracting the interest of industry as they relate to monitoring and control of complex engineered processes. Such processes are usually relying upon the expertise and ability of human operators for control purposes;they are often though controlled sub-optimally owing to high personnel turnover, lack of fundamental understanding of plant dynamics, inaccuracy or unreliability of manual control actions, etc. Capturing and utilizing the expert's knowledge effectively and efficiently in an intelligent control framework promises to improve plant operational conditions particularly in the event of large-scale disturbances. Modeling the expert operator's experience is a perfect semiotic paradigm since the operator employs a variety of informational granules in order to establish an effect-cause relationship and take appropriate action in the event of an emergency. This paper focuses upon the role of the human in intelligent control, means to extract expert knowledge, represent it and employ it for control of complex systems. A neuro-fuzzy implementation strategy is suggested and the model is tested on a pH neutralization process.
A typical single course on dynamics and control is limited in the coverage that can be given to either topic. At Rensselaer we have had separate, one-semester courses on dynamic systems (processdynamics) and process ...
A typical single course on dynamics and control is limited in the coverage that can be given to either topic. At Rensselaer we have had separate, one-semester courses on dynamic systems (processdynamics) and processcontrol for a number of years. In this paper we present an overview of the course on processdynamics and discuss how learning modules are used to reinforce a number of concepts. A discussion of the topics covered and the educational delivery methods used is provided. We also discuss the Rensselaer initiative to integrate mathematics throughout the science and engineering curriculum.
An integrated structural/control design for the vibration suppression of a composite plate with segmented piezoelectric (PZT) actuators is examined, The electric power used by the PZT materials becomes a very importan...
详细信息
An integrated structural/control design for the vibration suppression of a composite plate with segmented piezoelectric (PZT) actuators is examined, The electric power used by the PZT materials becomes a very important factor for real application to the control of flexible structures, The method to predict power required for the control is presented, and the power is used as a part of the objective functions, The structural weight, the state error energy, control energy, stability robustness, and the electric power required for vibration control are considered as the objective function. The locations of the PZT actuators are considered as control design variables, The ply orientation and thickness coefficients of the laminated composite plate are considered as structural design variables, The improved compromise multiobjective optimization by using a reduction factor of performance indices is applied to solve this optimization problem, The sequential linear programming method with move limits is used, The sensitivity analysis, which is required in the optimization process, is performed for the eigenstructure assignment control scheme and the performance indices with respect to the design parameters, The optimized results showed a significant amount of reduction in the structural weight as well as the control performance indices. The required electric power is also decreased.
It is well known that material and energy recycle may have a profound influence on the plant dynamics. Previous studies have shown that the introduction of recycle typically increases the dominant time-constant and di...
It is well known that material and energy recycle may have a profound influence on the plant dynamics. Previous studies have shown that the introduction of recycle typically increases the dominant time-constant and disturbance sensitivity of the plant. In this work we consider the effect of recycle on the zero dynamics and show that recycle may introduce severe overshoots and inverse responses. Sufficient conditions for the existence of nonminimum phase transmission zeros caused by the presence of recycle are derived. The conditions may be evaluated based on steady-state information about the individual units only. The results are illustrated through application to a reactor-separator system.
control of an industrial hydrolysis batch process using a model-based control strategy is described. For this purpose, non-linear artificial neural network models of the processdynamics are identified with the aid of...
详细信息
control of an industrial hydrolysis batch process using a model-based control strategy is described. For this purpose, non-linear artificial neural network models of the processdynamics are identified with the aid of a semi-empirical model. Neural networks are trained using an augmented training data set. Neural network models predict the TiO2 precipitation rate for each batch. Using this prediction, some of the process parameters can be optimised through the application of the flexible recipes concept to control the quality of the intermediate product and the duration of each batch. A simulator has been designed to study the application of flexible recipe instructions. Different adaptation mechanisms which use on-line process measurements are discussed to improve model accuracy and to counteract changes in processdynamics.
control schemes have been developed for a continuous pilot plant for vitamin production using the approach recommended by Luyben. The plant was simulated using Speedup and accurate physical property prediction. The ap...
control schemes have been developed for a continuous pilot plant for vitamin production using the approach recommended by Luyben. The plant was simulated using Speedup and accurate physical property prediction. The approach was modified to fit the plant but it was confirmed that the basic premise of requiring one stream in the recycle loop under flow control was confirmed.
In this paper we present LP based MPC solutions to two classes of large scale problems: (i) plant-wide control, and (ii) CD control in a paper machine. The former problem represents a challenge of considerable dynamic...
In this paper we present LP based MPC solutions to two classes of large scale problems: (i) plant-wide control, and (ii) CD control in a paper machine. The former problem represents a challenge of considerable dynamic complexity given the interactions inherent in a multiple-unit plant. An application is presented for a DuPont benchmark problem. While the latter problem has relatively benign dynamics, the input-output dimensionality is significant (approximately 400x400), and the required solution times are on the order of seconds. A customized algorithm is presented for the latter problem to enable real-time implementation.
Plantwide inventories of main components and impurities in a complex plant are sharply affected by interactions between recycles which, in general, are not desirable. However, this paper demonstrates how the interacti...
Plantwide inventories of main components and impurities in a complex plant are sharply affected by interactions between recycles which, in general, are not desirable. However, this paper demonstrates how the interactions between some recycle loops may be exploited to create feasible plantwide control structures that are impossible to achieve simultaneously with stand-alone units. Thus, flowsheet architecture, equipment design and control system design must be interrelated. Simultaneous design is necessary for items involved in plantwide control structures. The paper presents a simulation based methodology for evaluating the effect of recycle interactions on dynamics and plantwide control of complex plants. Steady-state and dynamic simulations are combined with controllability analysis, both in steady state and in dynamic mode. A case study handling the removal of impurities in a plant with nested loops illustrates the approach. The controllability of two flowsheet alternatives is evaluated. The steady-state analysis is confirmed at low frequencies. Possible difficulties may occur at higher frequencies, where the period of disturbances and the time constants of the distillation columns are of the same order of magnitude. The relative direction of disturbances plays a significant role. Closed loop simulation validates the main trends of the controllability analysis, showing in the same time the difficulty in managing a perfect multivariable control of the material balance.
In this paper a new model predictive control (MPC) strategy, applicable to a set of nonlinear systems, is proposed and the use of it is demonstrated on a model of a waste treatment reactor. The MPC strategy is an exte...
In this paper a new model predictive control (MPC) strategy, applicable to a set of nonlinear systems, is proposed and the use of it is demonstrated on a model of a waste treatment reactor. The MPC strategy is an extension of earlier work in optimization-based control [2]. The motivation for the study is to search for approaches to nonlinear MPC without having to solve the full nonlinear problem. We restrict our problem by defining a nonlinear model set as a convex combination of a set of bounding linear models. The weighting factors between the models can be a function of the states and/or inputs. At a given time-instant we compute an optimal future control sequence for each of the bounding linear models. A novel feature is that all models must obey the constraints for each of the control sequences. The reason for these additional constraints is that they provide us with feasibility guarantees. It also is a means of robustifying the MPC. The final control sequence is found by interpolating the control sequences derived from the optimization problems. There are different possible approaches for choosing the interpolation variables. Provided the optimization criterion and the constraint sets for the control variables and states are convex, the proposed control algorithm involves only convex optimization problems. The interpolating MPC strategy is applied to a waste treatment reactor, where the processdynamics are nonlinear and time-varying depending on the disturbance. Linearization is carried out to obtain bounding models for the process. The interpolating MPC is designed based on the bounding models. Through the example we demonstrate significant improvements over a standard quadratic MPC strategy based on linear models.
暂无评论