June 1, 2017 (Vol. 37, No. 11)
An Open Approach for a Smart and Affordable Bioprocess Control System
The goal for upstream bioprocess development is to support and maintain the ability to rapidly develop and maintain manufacturing processes that reproducibly deliver high-quality products with high yields and competitive economics.
The degree of process knowledge required by the implementation of PAT and Quality by Design (QbD) in the biopharmaceutical arena, and the requirement for end-product pricing competing favorably with petroleum-based products in the biofuel and biologically derived chemical arena, bring additional challenges to bear and a drive toward the development of “smart” or intelligent automated processes.
Process development scientists must work under constant time pressure to deliver processes to pilot- and full-scale manufacturing processes. In many circumstances process delivery deadlines are dictated by business rather than scientific or technical factors so that the time available may have little or no relation to technical challenges presented by the particularly product and process in development.
Meeting these challenges requires high data throughput and integration of the most appropriate equipment and cotrol systems for the task at hand—ideally unrestrained by vendor integration restrictions.
Early-stage companies and academic process science departments often have significant funding restraints as an additional challenge to meeting their research and process science goals. They must be able to meet milestones to acquire additional funding or the transfer of processes to third-party collaborators which require the development and analysis of the same amount of high-quality data as larger well-funded organizations.
Academic groups need to be able to deliver similar data to receive additional funding from government or public sector sources. These organizations may have no choice but to purchase used equipment (e.g., fermenters, process controllers, analytical equipment) and still have to pay the original vendor’s full price for refurbishment and software upgrades with more data integration capability.
To meet these challenges, there is a need for cost-effective, “intelligent” software systems that can control processes and acquire and analyze process data and integrate information from instruments and devices acquired from multiple equipment manufacturers.
Intelligent Process Software
An example of a software product that meets these challenges is BlueVis offered by BlueSens (Figure 1). Taking the previously described challenges into account, BlueSens has developed its BlueVis bioprocess software to offer a solution for fermentation applications that is open to other manufacturers.
In the first step, various sensors, analyzers, probes, and actuators from many manufacturers can be connected to the software to gather all real-time information necessary to control the process. Based on the real-time measurement, BlueVis runs internal calculations and is able to control the process via the connected actuators, like agitators, pumps for the feeding and pH control, digital mass flow meters for the gas input control and thermostats and cryostats for the temperature control.
In accompaniment with BlueSens’ “Take what you have” philosophy the software enables the assembly of an intelligent and automated bioprocessing system by using single (already existing) hardware components. All data collected and processed can be offered to process lead systems via an integrated OLE for process control (OPC) server solution.
Figure 1. BlueVis scheme with the functions of the software.
Acting as an OPC client, the software can also receive data from process lead systems. Most of the process lead systems manufacturers support the approach of smart bioprocessing and importance of connectivity to other systems and offer OPC connectivity as well.
Process software holds a central position in the process, while the hardware is becoming exchangeable resulting in more flexible set-ups. Still probes and sensors, such as gas sensors, are essential to gain vital real-time information about the process (Table 1).
A scale-independent gas analysis is a must have for nearly every kind of smart fermentation process. Data like OUR, CER, or RQ are main indicators of the metabolic state and give important information if the process is running in its specifications or if actuators (feeding pumps, temperature, mass flow controllers, etc.) must be adjusted. Adjustment as well as execution of pre-programmed controls easily can be done via BlueVis.
Table 1. Examples of Sensors and Probes That Can Be Connected to BlueVis
Soft-Sensors & Real-Time Calculations
To spare the scientists elaborate calculation and analysis, the software introduced here is able to calculate additional key parameters via soft-sensors. Soft-sensors are virtual sensors, which calculate unknown key parameters of the process from known measurement data.
To achieve that, mathematical models based on existing measurements are used to predict process behaviors or calculates interesting parameters like biomass, growth rates, or specific yield rates. Values which cannot be measured in any other way without investment in additional process equipment can be calculated by integrated soft-sensors, which resemble a very reasonable and noninvasive way to receive these data.
In BlueVis, this is translated into practice using integrated soft-sensors (biomass, growth rate, or other) from EXPUTEC. The approach is to get more out of the data that is already monitored and to gain more information about the process by combining the data and creating a smarter process.
Besides the pre-programmed soft-sensors, the BlueVis software offers even more features to use real-time calculated data to gain more information about the process and to improve it. The software has an integrated math sensor that is easy to set up. In general, all functions in BlueVis can be used without the need of special programming skills. Real-time data can be exported easily to the numerical computing environment MATLAB for further calculations and analysis.
Conclusion
The challenges faced by upstream process scientists throughout the biotechnology sector require the implementation of cost-effective process-control system with real-time process data acquisition and analysis. These challenges can met by a software solution without limits to sourcing of integrated components that allows simple integration of existing equipment (Figure 2), integrated controllers, such as easy-to-use proportional–integral–derivative (PID) controllers or soft-sensors, and the capability to implement automated calculations and control logic in real time without the requirement for programming expertise.
Figure 2. “Take what you have” set-up.
Björn Friedritz ([email protected]) is a marketing & PR specialist for
BlueSens.
