Wyniki wyszukiwania - BazTech

1

Improving efficiency of pH control by balance-based adaptive control application

Stebel K., Czeczot J., Metzger M.

Archives of Control Sciences

|

2013

|

Vol. 23, no. 1

19--40

EN

This paper deals with the efficient control of the pH process. Considering the PI + gain scheduling (PI+GS) controller as the benchmark and its control performance as the base, we investigate experimentally the overall improvement in the control performance obtained by the application of the Balance-Based Adaptive Controller (B-BAC), which requires only the measurement data of the flow rates and pH values. The improvement of the control efficiency is investigated not only in terms of the controlled variable performance but also in terms of the manipulated variable performance considered as the considerable control cost. The application of the B-BAController can ensure lower controlled pH variability at the price of the control effort similar to the PI+GS approach and thus it can improve the overall efficiency of pH control. The second important contribution is the experimental validation of the very simple and intuitive tuning procedure for the B-BAController.

2

Application of reduced order phenomenological model to pH nonlinear control

Stebel K., Metzger M.

Archives of Control Sciences

|

2005

|

Vol. 15, no. 2

199-216

EN

Changes in process sensitivity with pH make the design of the conventional controllers difficult. pH processes are very hard to control because of strong nonlinearity and usually time varying characteristics. A lot of work has been done in the field of phenomenological modelling of the pH processes. However, there is a need to simplify the model to have it suitable for control purposes. This follows from the fact that the internal model for the controllers must be as simple as possible. Wiener models consisting of a linear dynamic element followed by a static nonlinear element are considered to be perfect to represent a wide range of nonlinear processes. The proposed reduced order phenomenological model of pH has a similar form as the Wiener models but includes bilinear reduced order phenomenological dynamic element. This model was validated using the real world pilot plant. The possibility of combining the phenomenological and arbitrary approaches occurred. The aim of this paper is to present reduced order model of pH process which can be used for simulation purposes and to design process model based controller.

3

Application of the substrate consumption rate to the monitoring of distributed parameter bioreactors

Czeczot J., Metzger M., Babary J., Nihtilae M.

Control and Cybernetics

|

2001

|

Vol. 30, no 1

81-98

EN

This paper deals with the idea of estimation of the substrate consumption rate at each point of the tube of the classical distributed parameter bioreactor and its application to the monitoring of this system. It is shown how to approximate the profile of this parameter on the basis of the orthogonal collocation method and the recursive least-squares procedure with adjustable forgetting factor. Then, it is suggested how to apply this profile for monitoring of the bioreactor work (calculation of the current mass flux of the substrate being reacted in the reactor tube and of the total mass of the substrate reacted in the bioreactor tube during its activity). The idea presented in this paper ivas validated by means of the computer simulation and the results proving its very good performance complete the paper.

4

Monitoring and control of a class of distributed parameter bioreactors with application of the substrate consumption rate

Czeczot J., Metzger M., Babary J. P., Nihtilä M.

Archives of Control Sciences

|

2001

|

Vol. 11, no. 1/2

5-22

EN

This paper deals with the idea of estimation of the substrate consumption rate at each point of the tube of the classical distributed parameter bioreactor and with its application to the monitoring and control of this system. It is shown how to approximate the profile of this parameter on the basis of the orthogonal collocation method and of the recursive least-squares procedure with adjustable forgetting factor. Then, it is suggested how to apply this profile for monitoring of the bioreactor work (calculation of the current mass flux of the substrate being reacted in the reactor tube and of the total mass of the substrate converted in the bioreactor during its activity). Finally, the linearizing adaptive control law is derived on the basis of the substrate consumption rate estimated at the outlet of the bioreactor. This controller allows the outlet substrate concentration to be kept at its desired value by manipulating the velocity of the flowing medium. The idea presented in this paper has been validated by means of the computer simulation and the results, proving its very good properties, complete the paper.

5

Mathematical model of sequentially controlled activated sludge processes

Metzger M.

Archives of Control Sciences

|

1999

|

Vol. 9, no. 3/4

111-133

EN

Reduced order model of the activated sludge process dynamics, presented in the paper, can be treated as a compromise between simplicity (needed for control purposes) and complexity (needed for process behaviour description). Proper calibration of reduced order models is a very difficult task. A new strategy of the model calibration for two different-type sequentially controlled processes (sequentially operated continuous process and sequencing batch reactor) is presented. The proposed method accepts typical values of the model parameters and is based on the fitting of the measured process responses by manipulation of only two model parameters: the initial values of heterotrophic and autotrophic biomass concentrations. Discussion of the response fitting possibilities by changes of XBH and XBA in bio-acceptable limits, as well as validation results for experimental data from two pilot plants, show very promising application of the method.