Computational Design Modules
Fermentors Operating Conditions Evaluation Module

Needs Assessments

The scope of a biotechnology process may include Microbes Immobilization Carriers Fermentors when such a bioprocess adopts at least one heterogeneous bioreactor: The immobilization of the microbes is often carried out in a vessel that is called Fermentor. The operations of these Fermentor for supporting the continuous biotechnology process often entails effecting a precise re-initialization of the microbes in the immobilized state. The re-initialization is the resetting of the carriers of immobilization back to the original state, but beginning with reference statistically meaningful states that require different extents of restoration or re-initialization, before being introduced into the bioreactors. This re-initialization, which is of critical importance for process performance referencing, is expected to be supported with a computation-intensive software module. Assessing the functionalities that must characterize such computation intensive module for evolving an effective set of specifications for the development of such software, is based on the operations analysis of a such concept fermentor.  

The fundamentals of the assessment, of course, for the re-initialization stems from the non-uniformity with which the microbes grow during the stay in the bioreactors. The exit state of the microbes is evaluated  by one of two approaches, one is empirical and the other rigorous: The former measures the outlet state of the immobilized microbes for each carrier under real-time operating conditions, The latter models - possibly with Kolmogorovs stochastic methods; though both are based on the use of Neural Network Analysis as the same core method. In this assessment of the needs for the development  of specifications for a computational module for the operations of fermentors, the measured data approach is adopted.

Several batches of immobilized microbes beads at the end of each operation of bioreactor cycle are produced to generate a reference exit state.

Operations real-time exit state are compared to the weighted reference state results for each batch operation of microbes state due for re-initialization in fermentors, by inputting the real-time data into a Neural Network system for association with statistical meaningful data constructed from the weighted reference state.

The evaluated exit state properties are then used to set the Fermentor operating conditions for the re-initialization of the immobilized microbes for the given batch of beads or carriers.

Finally the restoration from that evaluated statistically meaningful state of immobilized microbes is performed, and the beads readied for the next batch of operation. 

Program Method
By virtue of the above proposed method of operation, the computational module for accomplishing the tasks must follow as closely as possibly the steps presented as Fig 1. Obviously the stipulation in the process of a Neural Network system places a requirement of decision making system for the definition of the exit state of each bead depending on measurable properties of the bead or carrier. indisputably, the accuracy of the prospective performance of the system improves with the number of test data made available for training, so as large sample as possible for the training is suggested.

The exit state as used involves properties or characteristics of the beads that are measurable, and must be inferred from the measurable data. Such data the evidently stands out include the

kinetic data changes of the immobilized microbes due to the immobilization, and the reaction dynamics consequent on these changes. These non-measurable data may effectively be evaluated by identifying and developing develop the dimensionless parameters that impact the reaction dynamics and kinetic data of the immobilized microbes. These are expected to be evaluated as part of the exit states of the immobilized microbes.

     Fig. 1 List of Computational Functional Tasks for Immobilized
               Microbes Re-initialization

• Perform several production-quality operations of the process/bioreactor and obtain a set of exit states representative immobilized microbes beads; for Tubular Flow Fixed Bed Bioreactors the data are obtained by intermittent interruption of the operation,
• Construct Fuzzy Math relationship between these exit states and the operations of the bioreactor/process and generate statistical representative exit state for the operation of the bioreactor
• Test the performance of the trained systems by using educated data and cross-check the results with real-time results until achieving stipulated level of confidence obtains,
• Acquire Beads exit data and relay data into computer module
• Compare the exit states of real-time operation batches against the weighted data based on the Fuzzy Math to construct a set of statistically representative data
• Relay data into Fermentors operation control system
• Enable weighting refinement with self-training of an intelligent system using new exit state data for correlation.

Specifications Development
The software developed to support the above computational program method must clearly be made a component of an existing Green technology computational System, and as such, be a computational module for adoption with a Green Technology Computational Platform. The Computational Modules by specification must be multi-threaded with threads for each of the tasks of data acquisition, measured data analyses, data relay to the Fermentors controller(s). The data analysis thread should by development be able to analyze and assess the role of each of the critical parameters defining the exit state of a bead or carrier of immobilized microbes. The Module should be trainable and trained for varying conditions that implement the evaluation of the dimensionless parametric studies with a constructed Fuzzy Math relationship. The module should by development continually update itself with each new measurements submitted for analysis. At a minimum, a thread should be provided for each of the Program Method sets given in the Table 1.

Table 1. List of Module Tasks Applications

The Module should adopt the median values of the exit state properties as the statistically meaningful data for each batch of effluent immobilized microbes, and use these statistical median values of the exit state properties to set the Fermentor operating conditions for the re-initialization of the immobilized microbes for the given batch of beads or carriers.

However, Client applications for interacting with the Module must also be part of the software specification; and the Client Software actually should be a collection of software pieces, one for the instruments for data logging of the measurable data, and for receiving re-initiation data for the Fermentor(s) Controllers.

  Company