Slug Flow Chemistry AI Control for Self-Optimizing Experiments

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for flow chemistry in slugs lack automation and optimization, leading to complex and inefficient process control.

Innovation Solution

A computer-implemented method using machine learning and artificial intelligence to determine a target parameter set for a flow chemistry setup in slugs, enabling self-optimization and automation by training a machine-learning model with process variables and optimizing parameters for improved control.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual experimentation and data analysis methods are used in flow chemistry research, then flexibility in experimental design is maintained, but productivity and time consumption are significantly reduced

Engineering Contradiction:
Improveexperiment throughputVSAvoidautomation system complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system performs self-optimization by automatically analyzing experimental data, identifying optimal parameters, and adjusting flow chemistry conditions without human intervention. The machine learning model learns from accumulated data to autonomously improve experimental outcomes, eliminating the need for manual data analysis and experimental design iterations.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system implements closed-loop feedback by continuously monitoring experimental results, comparing them against target values, and automatically adjusting process parameters. The machine learning model uses real-time and historical data to predict optimal settings, creating a self-correcting system that progressively improves productivity while maintaining controlled complexity.

Inventive Principle:
Principle #23Feedback

2Manufacturing precision

If extensive parameter space screening is performed manually, then comprehensive process optimization is achieved, but loss of time and experimental resources increase

Engineering Contradiction:
Improveprocess optimization accuracyVSAvoidscreening time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The machine learning model performs preliminary analysis of the parameter space by predicting promising regions before actual experiments are conducted. This pre-screening capability allows the system to focus experimental efforts on high-probability optimal conditions, achieving comprehensive optimization with significantly fewer experiments and reduced time loss.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system efficiently explores the parameter space by intelligently selecting which parameters to adjust and by how much, based on machine learning predictions. Instead of exhaustive manual screening, the AI system identifies critical parameters and optimal change magnitudes, achieving high manufacturing precision while minimizing the time required for parameter space coverage.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If automated control systems are implemented in flow chemistry, then productivity and consistency are improved, but device complexity and initial manual work increase

Engineering Contradiction:
Improveprocess control consistencyVSAvoidcontrol system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system achieves reliable and consistent process control through self-optimization capabilities. The machine learning model continuously learns from experimental data and automatically adjusts control parameters, eliminating the need for complex manual tuning and intervention. This self-service approach maintains high reliability while keeping the control system architecture relatively simple.

Inventive Principle:
Principle #25Self-service

4Productivity

If machine learning models are used for flow chemistry optimization, then productivity and data quality are enhanced, but device complexity and computational requirements increase

Engineering Contradiction:
Improveexperiment throughputVSAvoidcomputational system complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system replaces manual mechanical experimentation and data analysis with machine learning-based computational optimization. The AI model processes experimental data and generates optimization recommendations, substituting human cognitive and manual operations with automated computational processes. This substitution dramatically increases productivity while the modular implementation keeps computational system complexity manageable.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4157512B1Ai-system for flow chemistry
Publication Date: 2026.04.01 BASF SE
  • EP4157512B1 patent drawingFigure 1~2

AI summary

A computer implemented method for determining at least one target parameter set for a flow chemistry setup (110) for flow chemistry in slugs is disclosed. The method is a self-learning method. The method comprises the following steps: a) determining at least one process variable by using at least one sensor (122) of a flow chemistry setup (110); b) training of at least one machine-learning model (126) based on the process variable; c) determining the target parameter set by applying an optimizing algorithm in terms of at least one optimization target on the trained machine-learning model (126); d) providing the determined target parameter set and/or considering the determined target parameter set for evaluating a flow chemistry setup (110) and/or for evaluating at least one flow chemistry product.