Fermentation Process Data Modeling for Coupled Parameter Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The food fermentation process is complex and susceptible to various factors, making it difficult to establish reliable models for optimization and control due to its nonlinearity, time-delay characteristics, and strong coupling between parameters.
Innovation Solution
A data optimization method and system that utilizes a multi-scale cross-correlation feature filter (MCFF) for real-time feature extraction and processing, combined with machine learning and optimization algorithms, to create a reliable process model and optimize control data for the fermentation process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a white-box model is used to describe the fermentation process, then the model can incorporate fermentation mechanism knowledge, but the model establishment becomes particularly difficult due to process complexity and uncertain factors
Solution Approach 1:
The patent introduces a black-box model as an intermediary approach between mechanism-based white-box models and simple empirical models. This black-box model uses input-output variable relationships without requiring detailed fermentation mechanism knowledge, serving as a mediator that avoids the complexity of white-box model establishment while still providing reliable process description through data-driven approaches
Solution Approach 2:
The patent transforms the modeling approach by changing from mechanism-based parameters (white-box) to data-driven parameters (black-box). By using accessible process data and focusing on input-output variable relationships rather than internal mechanism parameters, the model establishment complexity is reduced while maintaining reliability through empirical evidence
2Ease of operation
If individual process parameters are regulated independently, then each parameter can be controlled separately, but chain reactions of other parameters occur resulting in uncontrollable fermentation process
Solution Approach 1:
The patent merges individual parameter controls into a unified optimization control system. By using a black-box model that captures the coupled relationships between multiple parameters, the system optimizes parameters simultaneously rather than independently, preventing chain reactions and maintaining overall process controllability while still allowing individual parameter adjustment
3Manufacturing precision
If the fermentation process is optimized to meet higher requirements for food safety, nutrition and quality, then product quality improves, but the process becomes more difficult to model and control due to strong nonlinearity and coupling
Solution Approach 1:
The patent replaces traditional mechanism-based modeling (mechanical/system-based approach) with data-driven black-box modeling. By substituting the need for detailed fermentation mechanism understanding with statistical and machine learning approaches that directly analyze input-output relationships, the system achieves high product quality control without the complexity of detailed process modeling
Data Source
AI summary
A data optimization method for a food fermentation process is provided. Process data of the food fermentation process is acquired, and a multi-scale cross-correlation feature filter (MCFF) is constructed. Feature data corresponding to the process data is extracted in real time based on the MCFF, and processed. A data prediction model corresponding to the feature data is created through a machine learning method, and based on the data prediction model and an optimization algorithm, predicted optimization control data corresponding to the food fermentation process is generated in real time. A data optimization system is further provided.