Chromatography Column State Prediction From Peak Resolution Trends

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing analysis systems rely on operator experience to determine the state of a column, which affects the accuracy of analysis, and lack precise methods for diagnosing column deterioration.

Innovation Solution

An analysis system that includes a storage unit to store peak resolutions from a chromatogram and a processor to predict the state of a column using methods like Kalman filters and linear regression, allowing for accurate diagnosis of column health and potential deterioration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If operator experience is used to determine column state, then the analysis system can operate without complex prediction mechanisms, but the diagnosis precision of column deterioration is insufficient

Engineering Contradiction:
Improvediagnosis precisionVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by storing historical peak resolution data and calculating average values before actual column deterioration occurs. The prediction unit uses this pre-stored data to forecast future column states, enabling proactive maintenance decisions rather than reactive responses based solely on operator experience.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms by continuously monitoring peak resolutions, comparing them against predicted values, and using the differences to refine column state assessments. The determination unit receives feedback from the prediction unit and adjusts replacement timing decisions based on whether actual measurements deviate from predicted trends.

Inventive Principle:
Principle #23Feedback

2Reliability

If column replacement is delayed to reduce costs, then operational expenses decrease, but analysis accuracy deteriorates due to column degradation

Engineering Contradiction:
Improveanalysis accuracyVSAvoidoperational cost
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The system performs preliminary prediction of column deterioration trends before actual performance degradation occurs. By calculating average peak resolutions and projecting future states, the system identifies the optimal replacement timing in advance, allowing scheduled replacements that prevent accuracy loss while avoiding premature replacements that would waste resources.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system monitors changes in peak resolution parameters over time and uses these parameter variations to predict when column performance will fall below acceptable thresholds. By tracking the evolution of resolution values and comparing them against predicted trajectories, the system determines the precise moment when replacement becomes necessary to maintain analysis accuracy.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If frequent column replacement is performed to maintain accuracy, then analysis precision is maintained, but system productivity decreases due to frequent interruptions

Engineering Contradiction:
Improveanalysis precisionVSAvoidsystem productivity
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The prediction unit performs preliminary assessments of column lifespan by analyzing historical data trends and calculating projected future performance. This allows the system to schedule replacements at optimal intervals that maintain precision while maximizing operational continuity, replacing columns based on predicted deterioration rather than fixed schedules that may be too frequent.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts replacement timing based on actual column performance characteristics and usage conditions. Rather than using static replacement schedules, the determination unit continuously evaluates whether current peak resolutions align with predictions and adjusts the timing of replacements to match actual column degradation rates, optimizing the balance between precision maintenance and productivity.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4257971B1Analysis system
Publication Date: 2025.12.24 YOKOGAWA ELECTRIC CORP
  • EP4257971B1 patent drawingFigure 1~2
  • EP4257971B1 patent drawingFigure 3~4
  • EP4257971B1 patent drawingFigure 5~6

AI summary

An analysis system includes: a storage configured to obtain resolutions of a column and to store each of the resolutions in association with a specific time, the column being configured to separate a component in a sample gas; and a processor configured to predict a state of the column based on resolution data indicating the stored resolutions.