George Ifrim

Modeling and control of a multipurpose biotechnological plant. Photobioreactor modeling

The present paper is the first of a series of two papers that concerns the modeling of a complex multipurpose biotechnological plant that consist in two interconnected subsystems, namely an anaerobic digestor and a photobioreactor. The paper is presenting a complex analytical model for the photobioreactor that embeds the effect of pH on the photosynthetic growth of microalgae. The static characteristics of the multivariable non-square system are discussed along with the solution for its reduction when control strategies are aimed. Simplified dynamic models for the design of control loops were obtained through Hankel analysis.

QFT Control of Dissolved Oxygen Concentration in a Wastewater Treatment Pilot Plant

Abstract The paper deals with the robust control of the dissolved oxygen concentration in the case of wastewater treatment processes from milk and beer industry. The QFT robust control method has been chosen for the dissolved oxygen concentration control. Firstly the dissolved oxygen control loop was identified. The QFT controller and prefilter were designed so that the behaviour of the closed loop system be bounded by the limits (upper and lower) imposed to the system. The QFT control of the dissolved oxygen concentration was tested using an experimentally pilot plant that is also presented in the paper.

Modelling of a multipurpose biotechnological plant in view of automatic control. Process modelling and control properties analysis

This work continues the paper [1] and it deals with a multipurpose biotechnological plant which includes a photobioreactor interconnected with an anaerobic digester. In the paper linearized models of reduced order for anaerobic digester and for the entire biotechnological plant are obtained. These models of reduced order are determined from detailed models of 35th order for anaerobic digester and of 51th order for the entire plant using Hankel analysis. It is proposed a control structure of the plant in which the photobioreactors control loop drives the one of anaerobic digester. The results obtained in [1] and in the present paper offer the necessary information (i.e. models and control structure configurations) to the controllers design for the control loops of the multipurpose biotechnological plant.

Analysis of phosphorus removal performances in a municipal treatment plant

This paper addresses the performance modeling and analysis in the case of a wastewater treatment system for removal of phosphorus, consisting a biological reactor and a secondary settling tank. The mathematical model was developed in the SIMBA simulation environment. The system is based on an actual municipal wastewater treatment plant. The implemented architecture is similar to that in BSM1 (the digestion anaerobic process is not treated) so that the Activated Sludge Model 2d is used to describe the variables and processes of the wastewater treatment plant. The control strategies were simulated, using the model of an influent having loads corresponding to the real plant, for a dry regime, in open and closed loop. As part of the analyzed strategies, PI-type regulators were used for the control of dissolved oxygen concentration, internal recirculation and ferric chloride addition, which accelerates the phosphorus precipitation.

Modeling of a wastewater treatment process using neural networks

This research deals with the use of artificial neural networks for modeling a nonlinear biotechnological process, thus approximating a nonlinear function. In the absence of a mathematical model, an artificial neural network trained with a set of representative data can successfully replace and predict the systems behavior in time. Furthermore, this approximation is useful in the identification algorithms, prediction and diagnosis of different biotechnological systems. The biotechnological system considered in this research is a biological wastewater treatment process with active sludge.

Extremum seeking control for an anaerobic digestion process

The paper deals with the optimal control (extremum seeking algorithm) of an anaerobic digestion process. It uses a simplified version of the anaerobic digester model. Three cases of optimization criteria are considered taking into account the quality variable of the process: 1. the methane quantity in gaseous form, 2. the level of pollutants (it is expressed as the difference between the influent substrate and the substrate quantities accumulated in the digester) and 3. an aggregated quality variable resulted from the combination of the first two criteria. The analysis of anaerobic digestion process static characteristics shows that each characteristic reaches a point of maximum. The extremum seeking algorithm aims to keep the operating point in this point of maximum. The simulation results confirm the good behaviour of the extremum seeking algorithm.

Optimal Operation of a Lumostatic Microalgae Cultivation Process

This chapter proposes the optimization of batch microalgae cultures in artificially lighted photobioreactors . The strategy consists in controlling the incident light intensity so that the microalgae growth rate is maximized. Two approaches were developed and compared. In the first one, the ratio between the incident light intensity and the cell concentration (light-to-microalgae ratio) is optimized, either offline or online, and then maintained at its optimal value. In the second approach, the cells growth rate is maintained at its optimal value by means of nonlinear model predictive controller (NMPC). The proposed control strategies are illustrated and their efficiency is assessed, in simulation, for Chlamydomonas reinhardtii batch cultures. The proposed lumostatic operation strategies are shown to lead to a higher cell productivity and to a more efficient light utilization in comparison to conventional constant light operation approach.

Control of the light-to-microalgae ratio in a photobioreactor

This paper deals with the design of a controller that maximizes the biomass production in an artificially lighted photobioreactor operated in batch mode, by maintaining the ratio between the incident light intensity and the biomass concentration (light-to-microalgae ratio) at targeted light regime. The controller acts on the incident light intensity applied on the photobioreactor surface. First, the optimal ratio trajectory is determined and then the obtained optimal trajectory is tracked using a Nonlinear Model Predictive Control law. The proposed control strategy performances were illustrated and assessed in simulation for a Chlamydomonas reinhardtii batch culture.

Feedback linearizing control of light-to-microalgae ratio in artificially lighted photobioreactors

Abstract The present paper describes the design and the validation of a lumostatic controller for artificially lighted photobioreactors operated in discontinuous mode. The ratio between the incident light intensity and the biomass concentration, termed light-to-microalgae ratio, was selected as output variable, while its control was provided by manipulating the power supply of a light source and consequently the incident light intensity. The biomass yield on light energy was introduced in order to properly compare the batches operated under constant light intensities with the lumostatic batches. The results obtained in simulation show that a lumostatic batch can yield at least 10% more biomass per mole of supplied photons. The nonlinear controller, synthesized on the feedback linearizing technique, was implemented and validated on a laboratory torus photobioreactor inoculated with Chlamydomonas reinhardtii cells.

Control of the yeast growth process using an image processing-based transducer

The paper presents a control system for biotechnological processes, using an image processing-based biomass concentration transducer. The advantages of using such transducers are the low price and the response time, which is faster than any off-line measuring method. These advantages do not exclude the accuracy of the measurement. The chosen case study is a yeast growth process, a typical biotechnological process. An approximate model, then a refined one were used for tuning the proportional-integral (PI) controller of the control loop. This solution proved to be reliable and accurate and also validated the use of the mentioned transducer for control purposes. The biological data provided by the captured images also proved that the system reached the control objective.