Rotating Machine Abnormality Detection with Chronological Frequency Visualization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional systems for detecting abnormalities in rotating machines, such as turbines and compressors, fail to provide real-time trend analysis, making it difficult to early detect changes in operating states and potentially leading to increased severity of abnormalities and shutdowns.

Innovation Solution

A system comprising sensors for detecting vibrations, a sampler for periodic data transmission, a frequency analyzer for real-time analysis using Fast Fourier Transform, and a display unit for chronological and real-time visualization of results, allowing operators to recognize changes in machine states over time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional abnormality detection devices display only determination results without trends over time, then the device structure remains simple, but the ability to early detect abnormalities is reduced

Engineering Contradiction:
Improveabnormality detection capabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent adds a time dimension to the display by showing determination results chronologically over multiple periods. This transforms static single-point data into dynamic time-series visualizations, enabling operators to recognize trends and early signs of abnormalities without fundamentally changing the core detection mechanism.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system performs preliminary analysis by accumulating determination results over time and displaying trends before actual abnormalities occur. This allows operators to detect developing issues in advance, taking preliminary action to prevent severe abnormalities and shutdowns.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If multiple devices are used to detect abnormalities of multiple rotating machines, then detection coverage is improved, but the complexity of monitoring and analysis increases

Engineering Contradiction:
Improvedetection coverageVSAvoidmonitoring complexity
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent combines determination results from multiple detecting devices into a single integrated display system. By chronologically displaying results from multiple machines on one interface, it merges scattered information into a unified view, reducing monitoring complexity while maintaining comprehensive detection coverage.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The display unit is designed to handle and visualize data from multiple different rotating machines simultaneously. This universal interface can accommodate various machine types and detection results, providing a multi-functional monitoring solution that simplifies operation across diverse equipment.

Inventive Principle:
Principle #6Universality (Multi-functionality)

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

Enables early detection of abnormalities in rotating machines by providing real-time, chronological, and three-dimensional visualization of frequency analysis results, reducing the need for multiple operators and enhancing predictive maintenance.

Implementation Method 1

a sensor for detecting vibration of rotating machines to which the devices are allocated and generating a signal indicative of the detected vibration

Methodology Applied
Scientific EffectVibration detection: Vibration

Data Source

PatentEP3206003B1System and method for detecting abnormality of rotating machines
Publication Date: 2019.09.25 KOBE STEEL LTD
  • EP3206003B1 patent drawingFigure 1
  • EP3206003B1 patent drawingFigure 2
  • EP3206003B1 patent drawingFigure 3

AI summary

A system for detecting an abnormality of rotating machines includes two or more devices for detecting an abnormality and a computer. The respective devices include a sensor for detecting vibration of the respective rotating machines and generating signals indicative of the vibration, and a transmitting unit for transmitting the' signals to the computer. The computer includes a sampler for sampling the signals with a sampling frequency and outputting, at a time at intervals of a predetermined period, a plurality of the data detected and sampled within the predetermined period, a storage for storing the data, a frequency analyzer for analyzing the frequencies of the data, a determination unit for determining whether the machines have an abnormality based on the results of the frequency analysis, and a display unit for displaying, in chronological order and in real time, the results of the frequency analysis and displaying the results of the determination.