Sensor Condition Detection Using Signal Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing monitoring systems for sensors, such as volume sensors, often malfunction, leading to inaccurate data that can cause material shortages or overflows, impacting productivity and customer satisfaction, as they require redundant sensing technologies to determine sensor conditions, increasing complexity and cost.

Innovation Solution

A system that uses a monitoring module with a sensor condition determination module (SCDM) to analyze sensor signals and determine if a sensor is operating normally, erratic, or stuck, employing feature extraction engines and support vector machines (SVM) to identify sensor conditions based solely on the sensor signal, eliminating the need for redundant sensing technologies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If redundant sensing technologies are used to determine sensor conditions, then sensor reliability is improved, but device complexity increases

Engineering Contradiction:
Improvesensor condition determination accuracyVSAvoidmonitoring system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The sensor monitoring system uses the sensor's own output signal to detect its own malfunction conditions. The system analyzes characteristics of the sensor signal itself (such as rate of change, pattern recognition) to determine if the sensor is malfunctioning, rather than requiring separate redundant sensing technologies. This self-diagnostic capability eliminates the need for additional hardware while maintaining reliable sensor condition determination.

Inventive Principle:
Principle #25Self-service

2Reliability

If redundant sensing technologies are used to determine sensor conditions, then sensor reliability is improved, but cost increases

Engineering Contradiction:
Improvesensor condition determination accuracyVSAvoidsystem cost
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The system leverages the existing sensor signal and processing infrastructure to perform self-diagnosis. By analyzing the sensor's own output characteristics through pattern recognition and rate-of-change calculations, the system eliminates the need for expensive redundant sensors or specialized diagnostic equipment, thereby reducing overall system cost while maintaining reliability.

Inventive Principle:
Principle #25Self-service

3Productivity

If sensor malfunctions are not detected, then operational continuity is maintained, but measurement precision deteriorates

Engineering Contradiction:
Improveoperational continuityVSAvoidvolume level data accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system continuously monitors sensor signal characteristics and provides feedback about sensor health status. By analyzing the rate of change and patterns in the sensor signal, the system can detect malfunctions in real-time and alert operators, ensuring that inaccurate measurements are identified and addressed while maintaining operational continuity through timely intervention.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3215815B1System for determining sensor condition
Publication Date: 2020.05.20 INTEL CORP
  • EP3215815B1 patent drawingFigure 1A
  • EP3215815B1 patent drawingFigure 1B
  • EP3215815B1 patent drawingFigure 2

AI summary

The present disclosure is directed to a system for determining sensor condition. A sensor signal generated by a sensor to communicate the current condition of an aspect being monitored by the sensor may also be employed to determine the condition of the sensor itself. For example, a device capable of determining if the sensor condition is normal or malfunctioning (e.g., erratic, stuck, etc.) may comprise a monitoring module (MM) to receive the sensor signal. The MM may comprise a sensor condition determination module (SCDM) to determine sensor condition. The SCDM may include a feature extraction engine to determine various characteristics of (e.g., to "extract features" from) the sensor signal and a model to determine sensor condition based on the extracted features. The model may include a support vector machine (SVM) taught to determine sensor condition utilizing sampled sensor signals correlated with annotations of sensor condition.