Fuzzy logic control of an unstable bioreactor

A rule based fuzzy controller (FLC) is developed for stabilization of an unstable continuous stirred tank bioreactor (CSTBR) from various start-up conditions. The output variable is the reactor substrate concentration and the manipulated variable is the dilution rate. The performance of the FLC is evaluated by simulating a mathematical model of an unstable CSTBR. FLC is robust to perturbations in the specific growth rate, specific consumption rate and also to a disturbance in the feed substrate concentration. The performance of the FLC is superior to that of a conventional proportional controller.     

Control of a bioreactor using a neural network

This paper describes an experimental investigation concerning the use of neural networks to achieve the non-linear control of a continuous stirred tank fermenter. The influent dilution rate and the substrate concentration have been selected as control variables. The backpropagation learning algorithm has been used for both off-line and on-line identification of the inverse model which provides the control action. Experimental results show the performance and the implementation simplicity of this control approach.     

Control of a Thiobacillus denitrificans bioreactor using machine vision

A PC-based machine vision system has been used to continuously monitor changes in biomass concentration and to control the undesirable production of colloidal elemental sulfer (a reactor upset condition due to an excessive concentration of inhibitory sulfide substrate) in a bioreactor containing Thiobacillus denitrificans. A field of view of a video camera was established which contained regions of different background lighting. Mean values of the distribution of red, green, and blue intensity components within corresponding regions of a digital image image captured from the camera were used to monitr color changes associated with changes in biomass concentration, and to determine if the reactor was in an upset condition. The ration of red to blue intensity components was an important parameter in detecting the formatin of an elemental sulfur precipitant. Using a stepper motor-driven pressure regulator, intelligent process control was performed by altering the hydrogen sulfide feed flow rate setpoint on the vision system measurements.     

Fuzzy self-tuning PI controller for bioreactors

A fuzzy self-tuned PI controller for regulation of a nonlinear bioreactor is presented. The basic idea is to parameterize Ziegler-Nichols like tuning formula by two parameters and and then to use an on-line fuzzy inference mechanism to tune the PI controller parameters k _c and _I . The fuzzy self-tuning method takes the process output error as input and the tuning parameters and as outputs. Simulation studies on the nonlinear bioreactor model equations show that the present method is superior to that of fixed parameters conventional PI controller (based on transfer function) for both servo and regulatory problems. The present fuzzy logic controller is robust to process parameters uncertainties and to changes in magnitude and direction of the disturbances.     

A fuzzy logic controller.

This paper describes a fuzzy sets method which is very useful for handling uncertainties and essential for knowledge acquisition of a human expert. Kinetics of a reactor is often complex and not trivial to describe by mathematical equations. Reactor control by traditional control technology is therefore difficult. A novel technology is presented. In the following a fuzzy inference (approximate reasoning) is used for decision making in analogy to human thinking, facilitating a more sophisticated control. Readers of this paper do not need any advanced mathematics beyond the four basic operations in arithmetic (+, -, x, divided by) and using the maximum and minimum values. This fuzzy inference is introduced to construct a fuzzy logic controller which is suitable for a nonlinear, multivariable and time variant system applied to a bioreactor.     

Control model of human stance using fuzzy logic

 A control model of human stance is proposed based on knowledge from behavioral experiments and physiological systems. The proposed model is based on the control of global variables specific to body orientation and alignment, rather than on the control of the body’s center of mass within the base of support. Furthermore, the proposed control model is not based on purely inverted pendulum body mechanics where only motion at one joint is controlled, as for instance the ankle. In the proposed model, the degrees of freedom are controlled by using reciprocal and synergistic muscle actions at multiple joints. The control model is based on three sets of different global variables which act in parallel: (1) limb length and its derivative, (2) limb orientation and its derivative, and (3) trunk attitude and its derivative. An important feature of the control model is the use of fuzzy logic, which enables us to model experimental findings and physiological knowledge in a meaningful and explicit way using fuzzy if-then rules. In the control model, 36 fuzzy if-then rules are implemented and applied using a four-linked segment model consisting of a trunk, thigh, shank and foot. Uni- and biarticular limb muscles and trunk muscles are represented as torque actuators at each individual joint. In the model, three sets of global variables act in parallel and make corrective and coordinated responses to internal, self-induced perturbations. The data show that the use of global variables and fuzzy logic successfully enables us to model human standing with sway about a point of equilibrium. Small changes in, for example, total body sway are comparable to those seen during natural sway in human stance. The selected controllers—limb length, limb orientation and trunk attitude—seem to be appropriate for human stance control. Received: 30 October 1996/Accepted in revised form: 7 April 1997     

Application of fuzzy multiobjective optimization for self-sucking impellers in a bioreactor

For three types of self-sucking impellers (fourand six-pipe and disk impellers) mixing power, initial point, amount of gas leaving the impeller and mass transfer coefficient were determined experimentally. Investigations were performed for two systems: water and biomass solution.From the point of view of a minimum mixing power and maximum mass transfer coefficient the best impeller has been chosen. Fuzzy multiobjective optimization for determination of optimum operating conditions is proposed.List of Symbols c concentration of oxygen - D tank diameter - d impeller diameter - g acceleration of gravity - H height of liquid in the tank - H height of liquid above impeller, H=H-y - k consistency coefficient - k _L a volumetric mass transfer coefficient - N rotational speed of impeller - n flow behaviour index - P mixing power for pure liquid - P _G mixing power for aerated liquid - V _G volumetric air flow rate - y distance of impeller from the tank bottom - v _a apparent kinematic viscosity of liquid - density of liquid - time - gas hold-up - Eu=P/N ³ d ⁵ or Eu_G=P _G /N ³ d ⁵ Euler Number for non-gassed or aerated liquid - Fr=N ² d/g Froude Number - Fr^*=N ² d ² /g(H -y) modified Froude Number - K_G=V _G /N d ³ gas flow number - Re=N d ² /v _a Reynolds Number - Sh=k _K a/(g ² /v _a )^1/3 Sherwood Number     

Control of continuous fed-batch fermentation process using neural network based model predictive controller

Cell growth and metabolite production greatly depend on the feeding of the nutrients in fed-batch fermentations. A strategy for controlling the glucose feed rate in fed-batch baker’s yeast fermentation and a novel controller was studied. The difference between the specific carbon dioxide evolution rate and oxygen uptake rate (Q _c − Q _o) was used as controller variable. The controller evaluated was neural network based model predictive controller and optimizer. The performance of the controller was evaluated by the set point tracking. Results showed good performance of the controller.     

Dynamic fuzzy model based predictive controller for a biochemical reactor

The kinetics of bioreactions often involve some uncertainties and the dynamics of the process vary during the course of fermentation. For such processes, conventional control schemes may not provide satisfactory control performance and demands extra effort to design advanced control schemes. In this study, a dynamic fuzzy model based predictive controller (DFMBPC) is presented for the control of a biochemical reactor. The DFMBPC incorporates an adaptive fuzzy modeling framework into a model based predictive control scheme to derive analytical controller output. The DFMBPC has the flexibility to opt with various types of fuzzy models whose choice also lead to improve the control performance. The performance of DFMBPC is evaluated by comparing with a fuzzy model based predictive controller (FMBPC) with no model adaptation and a conventional PI controller. The results show that DFMBPC provides better performance for tracking setpoint changes and rejecting unmeasured disturbances in the biochemical reactor.     

Design and simulation of a fuzzy controller for fed-batch yeast fermentation

In baker's yeast fermentation, the process is non-linear and the response of the system to changes in glucose feeding has a very long delay time. Therefore, a conventional system can not give satisfactory results. In this paper, a fuzzy controller designed to control a fed-batch fermenter is presented. The fuzzy controller uses Respiratory Quotient (RQ) as a controller input and produces glucose feeding rate as control variable. The controller has been tested on a simulated fed-batch fermenter. The results show that the maximum yeast production is possible by keeping the specific growth rate (μ) and the glucose concentration (C _s) at preset values (μ _Cand C _s,c) and minimizing the ethanol production.     

An adaptive fuzzy logic controller using the respiratory quotient as an indicator of overdosage in the baker's yeast process

The baker's yeast process was optimised with a fuzzy logic controller, which is capable of detecting (with the respiratory quotient as indicator) and eliminating overdosage. The controller was developed to enable automatic modification of the set value for the respiratory quotient according to glucose concentration in the broth. With this controller, a cell yield of 55% (w/w) from glucose and a maximum specific growth rate of 0.16 h^–1 were obtained.     

Wastewater nitrification kinetics using reciprocating jet bioreactor

Kinetic investigations on growth parameters of nitrifying and COD oxidizing bacteria were carried out with recourse to a three stage reciprocating jet bioreactor system using real life wastewater. The system employed in this investigation essentially consisted of separate aerobic oxidation stage along with nitrification stage and anaerobic denitrification stage with facility for biomass recirculation whenever necessary. Steady-state COD oxidation reactor performance was assessed for various values of residence time. Yield coefficient and decay coefficient of COD oxidizing biomass were obtained as 0.3329 kg BM/kg COD and 0.0032 (1/h) respectively.It was observed that COD oxidizing bacteria co-existed with nitrifying bacteria during nitrification process due to the nature of wastewater used. Steady-state nitrification reactor performance was also assessed for various residence time values. Exact concentration of nitrifying and COD oxidizing biomass in the nitrification reactor was then estimated with the help of kinetic growth parameters of COD oxidizing biomass and extent of COD oxidation achieved in nitrification reactor. This further enabled evaluation of corrected kinetic growth parameters estimated as 0.4272 kg BM/kg NH ₄ ⁺ -N and 0.00626 (1/h) for nitrifier biomass yield coefficient and decay coefficient respectively.     

Biological hydrogen production using a membrane bioreactor

A cross-flow membrane was coupled to a chemostat to create an anaerobic membrane bioreactor (MBR) for biological hydrogen production. The reactor was fed glucose (10,000 mg/L) and inoculated with a soil inoculum heat-treated to kill non-spore-forming methanogens. Hydrogen gas was consistently produced at a concentration of 57-60% in the headspace under all conditions. When operated in chemostat mode (no flow through the membrane) at a hydraulic retention time (HRT) of 3.3 h, 90% of the glucose was removed, producing 2200 mg/L of cells and 500 mL/h of biogas. When operated in MBR mode, the solids retention time (SRT) was increased to SRT = 12 h producing a solids concentration in the reactor of 5800 mg/L. This SRT increased the overall glucose utilization (98%), the biogas production rate (640 mL/h), and the conversion efficiency of glucose-to-hydrogen from 22% (no MBR) to 25% (based on a maximum of 4 mol-H(2)/mol-glucose). When the SRT was increased from 5 h to 48 h, glucose utilization (99%) and biomass concentrations (8,800 +/- 600 mg/L) both increased. However, the biogas production decreased (310 +/- 40 mL/h) and the glucose-to-hydrogen conversion efficiency decreased from 37 +/- 4% to 18 +/- 3%. Sustained permeate flows through the membrane were in the range of 57 to 60 L/m(2) h for three different membrane pore sizes (0.3, 0.5, and 0.8 microm). Most (93.7% to 99.3%) of the membrane resistance was due to internal fouling and the reversible cake resistance, and not the membrane itself. Regular backpulsing was essential for maintaining permeate flux through the membrane. Analysis of DNA sequences using ribosomal intergenic spacer analysis indicated bacteria were most closely related to members of Clostridiaceae and Flexibacteraceae, including Clostridium acidisoli CAC237756 (97%), Linmingia china AF481148 (97%), and Cytophaga sp. MDA2507 AF238333 (99%). No PCR amplification of 16s rRNA genes was obtained when archaea-specific primers were used.     

Cardiac tissue engineering using perfusion bioreactor systems

This protocol describes tissue engineering of synchronously contractile cardiac constructs by culturing cardiac cell populations on porous scaffolds (in some cases with an array of channels) and bioreactors with perfusion of culture medium (in some cases supplemented with an oxygen carrier). The overall approach is 'biomimetic' in nature as it tends to provide in vivo-like oxygen supply to cultured cells and thereby overcome inherent limitations of diffusional transport in conventional culture systems. In order to mimic the capillary network, cells are cultured on channeled elastomer scaffolds that are perfused with culture medium that can contain oxygen carriers. The overall protocol takes 2-4 weeks, including assembly of the perfusion systems, preparation of scaffolds, cell seeding and cultivation, and on-line and end-point assessment methods. This model is well suited for a wide range of cardiac tissue engineering applications, including the use of human stem cells, and high-fidelity models for biological research.     

Control of dissolved oxygen concentration in mammalian cell culture using a computer-coupled mass flow controller

Summary A simple proportional control system for dissolved oxygen (DO) concentration in cell culture medium was developed by using a computer-coupled mass flow controller. The DO levels were very stable during the cultivation of Vero-6, while flow rates of air and/or oxygen enriched air were gradually changed depending on the DO concentration and the preset DO level. Vero-6 cells could grow normally to the confluence in the range of 30% and 50% of DO. Growth of Vero-6 at 10% of DO was markedly retarded.     

Optimization of fed-batch bioreactor using neural network model

An algorithm using feedforward neural network model for determining optimal substrate feeding policies for fed-batch fermentation process is presented in this work. The algorithm involves developing the neural network model of the process using the sampled data. The trained neural network model in turn is used for optimization purposes. The advantages of this technique is that optimization can be achieved without detailed kinetic model of the process and the computation of gradient of objective function with respect to control variables is straightforward. The application of the technique is demonstrated with two examples, namely, production of secreted protein and invertase. The simulation results show that the discrete-time dynamics of fed-batch bioreactor can be satisfactorily approximated using a feedforward sigmoidal neural network. The optimal policies obtained with the neural network model agree reasonably well with the previously reported results.     

Optimization of bioreactor using metabolic control analysis approach

A new methodology based on a metabolic control analysis (MCA) approach is developed for the optimization of continuous cascade bioreactor system. A general framework for representation of a cascade bioreactor system consisting of a large number of reactors as a single network is proposed. The kinetic and transport processes occurring in the system are represented as a reaction network with appropriate stoichiometry. Such representation of the bioreactor systems makes it amenable to the direct application of the MCA approach. The process sensitivity information is extracted using MCA methodology in the form of flux and concentration control coefficients. The process sensitivity information is shown to be a useful guide for determining the choice of decision variables for the purpose of optimization. A generalized problem of optimization of the bioreactor is formulated in which the decision variables are the operating conditions and kinetic parameters. The gradient of the objective function to be maximized with respect to all decision variables is obtained in the form of response coefficients. This gradient information can be used in any gradient-based optimization algorithm. The efficiency of the proposed technique is demonstrated with two examples taken from literature: biotransformation of crotonobetaine and alcohol fermentation in cascade bioreactor system.     

Disposable bioreactor for cell culture using wave-induced agitation

This work describes a novel bioreactor system for the cultivation of animal, insect, and plant cells using wave agitation induced by a rocking motion. This agitation system provides good nutrient distribution, off-bottom suspension, and excellent oxygen transfer without damaging fluid shear or gas bubbles. Unlike other cell culture systems, such as spinners, hollow-fiber bioreactors, and roller bottles, scale-up is simple, and has been demonstrated up to 100 L of culture volume. The bioreactor is disposable, and therefore requires no cleaning or sterilization. Additions and sampling are possible without the need for a laminar flow cabinet. The unit can be placed in an incubator requiring minimal instrumentation. These features dramatically lower the purchase cost, and operating expenses of this laboratory/pilot scale cell cultivation system. Results are presented for various model systems: 1) recombinant NS0 cells in suspension; 2) adenovirus production using human 293 cells in suspension; 3) Sf9 insect cell/baculovirus system; and 4) human 293 cells on microcarrier. These examples show the general suitability of the system for cells in suspension, anchorage-dependent culture, and virus production in research and GMP applications. This revised version was published online in July 2006 with corrections to the Cover Date.     

Hydrogenotrophic denitrification of synthetic aquaculture wastewater using membrane bioreactor

A hydrogenotrophic denitrification system was evaluated in removing nitrate from synthetic aquaculture wastewater for recirculation purposes. Two membrane bioreactor (MBR) systems, namely, aeration–denitrification system (ADS) and denitrification–aeration system (DAS) were studied with 50 mg/L of influent concentrations for both organic matter and nitrate nitrogen. The DAS achieved better removal efficiency of 91.4% total nitrogen (T-N) and denitrification rate of 363.7 mg/L.day at a HRT of 3 h compared to ADS. Further, there was no nitrite accumulation in the DAS effluent. The nitrite accumulation in ADS effluent was lesser when CO₂ was used as buffer rather than K₂HPO₄ and KH₂PO₄. Estimation of kinetic parameters of hydrogenotrophic bacteria indicated lesser sludge production compared to heterotrophic denitrification. In the DAS, membrane fouling was nonexistent in the aeration reactor that was used to produce the recirculating effluent. On the contrary, membrane fouling was observed in the denitrification reactor that supplied hydrogen to the mixed liquor. Thus, this study demonstrated DAS capability in maintaining the acceptable water quality appropriate for aquaculture, in which a closed recirculating system is typically used.