Microbial Biomass Production Using Digital Twins for Predictive Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current microbial biomass production systems face challenges in achieving optimal productivity and efficiency while minimizing resource consumption, with existing real-time control technologies lacking predictive capabilities and requiring excessive computing resources.
Innovation Solution
A microbial biomass production system utilizing a virtual environment for process emulation, combined with a trained model process to select and optimize control tools, including gas mixture regulation, biomass control, and aeration, to enhance productivity and reduce resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If real-time control technology is used to control microbial biomass production, then process control capability is improved, but predictive capability remains insufficient
Solution Approach 1:
The system performs preliminary actions by training machine learning models on historical process data before actual production. The trained models predict future process outcomes and optimize control parameters in advance, enabling proactive rather than reactive control of microbial biomass production
Solution Approach 2:
The system creates a virtual copy of the physical bioreactor process through digital twins and simulation models. This virtual model replicates the biological and chemical processes, allowing predictive analysis and optimization without interfering with the actual production system
2Loss of information
If complex control systems with high computing resources are deployed, then predictive capability is improved, but computing resource consumption increases excessively
Solution Approach 1:
The control system is segmented into distributed edge computing nodes within the bioreactor system that perform local predictive calculations. This distributes the computational burden across multiple low-power devices rather than relying on a single high-performance computing system, reducing overall energy consumption while maintaining predictive capability
Solution Approach 2:
The system dynamically adjusts model complexity and computation frequency based on process conditions and available computing resources. During stable phases, simpler models with lower computational requirements are used, while during critical transitions, more complex models are activated temporarily, optimizing the balance between predictive accuracy and energy consumption
Data Source
AI summary
The present disclosure relates to embodiments for solving the problem of microbial biomass production. The present disclosure describes systems and methods that increase the productivity of microbial biomass production by modeling the process of producing the specified biomass and selecting the parameters of the process of producing microbial biomass based on the results of this modeling. Also disclosed are systems and methods that reduce the resources used to produce microbial biomass by modeling the production process of this biomass using a process model improved by retraining.