Blast Furnace Fault Detection Using Sensor Synchronization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional methods for determining faults in blast furnaces fail to distinguish between operational faults and sensor faults, leading to delayed detection, especially in shaft-pressure sensors, due to the use of statistical fault indices and thresholds that do not consider synchronicity between sensor outputs.

Innovation Solution

A blast furnace fault determination apparatus that calculates a fault index using output values from multiple sensors and a ventilation index, with separate thresholds for early detection of operational and sensor faults, allowing for simultaneous monitoring of ventilation resistance and fault indices to accurately identify faults.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a statistical fault index is used to determine faults in blast furnaces, then fault detection coverage is improved, but the ability to distinguish between operational faults and sensor faults deteriorates

Engineering Contradiction:

Improvefault detection coverageVSAvoidfault type discrimination accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent segments the fault detection task into two independent evaluation systems: one for operational faults (using ventilation index and operational parameters) and another for sensor faults (using synchronization degree and sensor-specific statistics). This segmentation allows each system to specialize in detecting its target fault type, resolving the contradiction between comprehensive detection and accurate discrimination.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces synchronization degree as an intermediary metric that specifically measures the consistency between shaft-pressure sensors and other sensors. This intermediary serves as a dedicated indicator for sensor faults, separating the detection of sensor issues from operational issues and enabling accurate fault type discrimination while maintaining comprehensive coverage.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If a threshold is set for sensor output values to detect faults, then simple fault detection is achieved, but detection is delayed until the fault becomes severe

Engineering Contradiction:

Improvefault detection simplicityVSAvoidfault detection timing

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The patent replaces static threshold-based detection with dynamic statistical evaluation that continuously monitors multiple parameters including mean, variance, and synchronization degree. This dynamic approach adapts to normal operational variations and detects faults earlier by identifying subtle deviations from expected statistical patterns rather than waiting for extreme threshold violations.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes the detection parameter from simple output value magnitude to composite statistical metrics including synchronization degree, variance, and correlation coefficients. These parameter changes enable detection of gradual fault development and early anomalies before they reach severe threshold levels, reducing detection delay while maintaining operational simplicity.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If mean and variance of nearby sensors are used to detect target sensor faults, then sensor fault detection is enabled, but synchronicity between sensors is not considered causing delayed detection

Engineering Contradiction:

Improvesensor fault detection capabilityVSAvoidfault detection timing accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent introduces synchronization degree as a dedicated intermediary metric that specifically measures the temporal and statistical consistency between shaft-pressure sensors and other sensors. This intermediary captures the synchronicity relationship that simple mean-variance analysis misses, enabling timely and accurate sensor fault detection by identifying asynchrony as an early fault indicator.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentEP3950966B1Blast furnace abnormality assessment device, blast furnace abnormality assessment method, and blast furnace operation method

Publication Date: 2024.06.19 JFE STEEL CORP

Figure 1
Figure 2(a)~2(c)
Figure 3~4

AI summary

A blast furnace fault determination apparatus according to the present invention includes a fault index calculation unit configured to calculate a fault index indicative of the degree of fault in a blast furnace, a ventilation index calculation unit configured to calculate a ventilation index of the blast furnace, and a determination unit configured to determine a fault condition in the blast furnace using the fault index calculated by the fault index calculation unit and the ventilation index calculated by the ventilation index calculation unit. The fault index calculation unit may calculate the fault index using output values of a shaft-pressure sensor group installed around the furnace body of the blast furnace.