Rolling Stand Vibration Spectra for Coupled Chatter Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for detecting chatter events in systems with multiple coupled rolling stands are inadequate, leading to reduced product quality, component damage, and potential equipment failure, as they often fail to reliably identify these events in complex systems.

Innovation Solution

A method that detects chatter events by simultaneously meeting two conditions: the presence of peaks in both amplitude spectra within a common frequency range and exceeding predetermined amplitude thresholds, optionally using additional criteria like signal correlation and amplitude gradients, to provide early warnings and improve detection accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing methods are used to detect chatter events in multiple coupled rolling stands, then the detection process is simple, but the reliability of chatter event detection is insufficient

Engineering Contradiction:
Improvechatter event detection reliabilityVSAvoiddetection method complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The detection method is segmented into multiple independent evaluation criteria: frequency range matching, amplitude threshold evaluation, and correlation coefficient analysis. Each criterion independently assesses a specific aspect of chatter detection, and their combined results improve overall detection reliability without requiring a completely complex integrated system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by requiring only two out of three criteria to be satisfied for chatter detection (frequency matching + amplitude threshold, or frequency matching + correlation coefficient). This partial satisfaction approach provides sufficient detection reliability while avoiding the excessive complexity of requiring all three criteria simultaneously.

Inventive Principle:
Principle #16Partial or excessive action

2Loss of time

If existing detection methods are used, then the implementation is straightforward, but early warning capability is insufficient

Engineering Contradiction:
Improvewarning time for chatter eventsVSAvoidchatter detection accuracy
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The method performs preliminary evaluation by continuously monitoring frequency ranges and correlation coefficients before amplitude thresholds are fully exceeded. When frequency matching and positive correlation are detected, the system prepares for potential chatter events, providing early warning capability while maintaining accurate detection through subsequent amplitude threshold evaluation.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If simple peak detection is used in amplitude spectra, then the method is easy to implement, but false detection of chatter events occurs

Engineering Contradiction:
Improvechatter detection accuracyVSAvoidspectral analysis complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent uses correlation coefficient analysis as a feedback mechanism to verify whether detected peaks represent genuine chatter events. The correlation coefficient provides feedback on the consistency between vibration signals from different rolling stands, allowing the system to distinguish true chatter from false positives while maintaining relatively simple spectral analysis procedures.

Inventive Principle:
Principle #23Feedback

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

Enhances the reliability of chatter event detection and provides timely warnings, reducing the risk of equipment damage and improving product quality by accurately predicting and preventing chatter in rolling mills.

Implementation Method 1

detecting the vibrations at the first rolling stand (1, 2) with a first sensor in the form of a first temporal vibration signal; and detecting the vibrations at the second rolling stand (1, 2) downstream of the first rolling stand in the rolling direction of the strip, simultaneously with the detection of the vibrations at the first rolling stand, with a second sensor in the form of a second temporal vibration signal

Methodology Applied
Scientific EffectVibration detection: Vibration

Implementation Method 2

determining a first amplitude spectrum for the first rolling stand (1, 2) and a second amplitude spectrum for the second rolling stand (1, 2) from the first and second temporal vibration signals, respectively

Methodology Applied
Scientific EffectFrequency domain transformation:

Data Source

PatentEP4599219B1Method for analysing the vibration behaviour of a system
Publication Date: 2026.04.08 SMS GROUP GMBH
  • EP4599219B1 patent drawingFigure 1
  • EP4599219B1 patent drawingFigure 2
  • EP4599219B1 patent drawingFigure 3~4

AI summary

The invention relates to a method and to a computer programme for analysing the vibration behaviour of a system consisting of at least a first and a second roll stand which are coupled to one another via a belt stretched into both roll stands. In order to better analyse the vibration behaviour of individual roll stands, it is known to first capture the vibrations of the roll stands as a temporal vibration signal using sensors and then to transform said temporal vibration signal into an amplitude spectrum in the frequency domain. The amplitude of said amplitude spectrum is then analysed to determine whether it is above a predefined amplitude threshold value. If so, an undesirable rattling event is detected. In order to also be able to apply such a rattling event to systems comprising at least two roll stands which are connected to one another via a common belt, according to the invention the amplitude spectra of the two roll stands is created and then a common frequency range is searched for and identified in which the two amplitude spectra are compared with one another. A rattling event is detected for said system only if peaks of said two amplitude spectra lie in the common frequency range and if both peaks have amplitudes above defined amplitude threshold values.