Spectroscopic Bioreactor Control for Real-Time Parameter Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional bioprocess control methods in bioreactors are hindered by the need for cumbersome off-line measurements and models that are often scale-specific, making it difficult to accurately control bioprocessing variables like glucose concentration and viable cell density, requiring extensive data sets and manual analysis.
Innovation Solution
A computer-implemented method using spectroscopy to obtain measurement results, generate bioprocessing parameters, and control the bioprocess through a controller configured with trained models, allowing for real-time adjustment of additive fluid and gas flows based on bioprocessing parameters, thereby reducing the need for manual off-line measurements and improving scalability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If off-line measurements are used to determine bioprocessing variables, then measurement capability is provided, but operator workload and complexity increase significantly
Solution Approach 1:
The patent replaces manual mechanical sampling operations with an automated optical detection system. A probe continuously measures bioprocessing variables in-situ using light absorption spectroscopy, eliminating the need for operators to manually withdraw samples and perform offline analysis. This substitution of mechanical/manual operations with optical automation directly resolves the contradiction between measurement capability and operational complexity.
2Measurement precision
If models are trained on extensive reference data sets from large bioprocessing conditions, then prediction accuracy improves, but data collection time and resource requirements increase
Solution Approach 1:
The patent implements a progressive model training approach where the prediction model is initially trained on a limited reference data set covering a subset of bioprocessing conditions. The model is then iteratively refined by adding new reference measurements as they become available during actual bioprocess runs. This partial action approach allows the system to achieve functional prediction accuracy without requiring complete reference data collection upfront, thereby reducing initial data collection time while maintaining improving precision over time.
3Measurement precision
If conventional sensors are used in bioprocessing fluid, then direct measurement is possible, but cell cultivation compatibility is compromised
Solution Approach 1:
The patent introduces light as an intermediary measurement medium that can penetrate the bioprocessing fluid without physically contacting or contaminating the cell culture. Instead of inserting conventional electrical sensors that may affect cell viability, the system uses optical probes that measure light absorption characteristics of the fluid. This intermediary optical approach enables direct in-situ measurement while maintaining cell cultivation integrity, resolving the contradiction between measurement capability and biological compatibility.
4Measurement precision
If spectroscopic models are generated for specific bioreactor systems, then acceptable accuracy is achieved, but model scalability to different scales is limited
Solution Approach 1:
The patent develops a universal spectroscopic prediction model that can be applied across multiple bioreactor scales and system configurations. Rather than creating separate models for each specific bioreactor type and scale, the system uses a standardized optical probe and spectroscopic measurement approach that captures fundamental bioprocessing variables applicable to diverse systems. The model is trained on reference data from various conditions and scales, enabling it to function universally across different bioreactor implementations, thus achieving both accuracy and scalability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances bioprocess control by providing real-time, scalable adjustments to bioprocessing parameters, improving bioprocess outcomes and reducing operator-intensive activities, while enabling more precise control of bioreactor conditions.
Implementation Method 1
A sensor S is configured to perform spectroscopy of a bioprocessing fluid FL
Data Source
AI summary
The present invention relates to a computer implemented method performed by a controller (C) configured to control a bioprocess comprised in a bioreactor (BR), the method comprising obtaining (410) measurement results by performing spectroscopy of a bioprocessing fluid (FL) comprised in the bioreactor (BR), generating bioprocessing parameters using the measurement results, one or more bioprocessing target parameters and one or more trained models, and, controlling the bioprocess using the generated bioprocessing parameters. The method wherein the one or more trained models are neural networks, wherein the measurement results comprise a spectrum, wherein the spectrum is split to a number N parts used to calculate N average values, wherein the N average values and the corresponding values of bioprocessing parameters are used § as features in the neural network.