Multisensor Facility Diagnosis for Temperature-Based Abnormality Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for diagnosing production facilities do not accurately identify variables associated with abnormalities due to the lack of external observation data, leading to potential misalignment between the arm and sensor, which can result in inaccurate diagnosis.
Innovation Solution
A learning device that acquires data through external observation, including image, temperature, and distance data, to generate a learning model for inferring the condition of workpieces and production facilities, using machine learning methods such as supervised, unsupervised, and reinforcement learning, with deep learning combinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If external observation data (image, temperature, distance data) is acquired and used in the learning model, then the measurement precision and reliability of abnormality identification is improved, but the device complexity increases due to multiple sensors and data processing requirements
Solution Approach 1:
The patent combines multiple types of sensors (imaging device, temperature sensor, range sensor) into an integrated diagnostic system that collects various data types simultaneously. This merging approach allows the system to capture comprehensive information about the production facility's state, improving abnormality identification accuracy while managing complexity through unified data processing architecture.
Solution Approach 2:
The learning model is designed to process multiple data types (image data, temperature data, distance data) and identify various abnormalities in production facilities. This multi-functional approach enables a single system to diagnose different types of issues across various production equipment, improving measurement precision without requiring separate specialized systems for each diagnostic task.
2Reliability
If multiple types of data (image, temperature, distance, setting data) are collected and processed, then the reliability of diagnosis is improved, but the loss of time for data processing and model generation increases
Solution Approach 1:
The system performs preliminary data processing by collecting and preprocessing multiple data types (image, temperature, distance, setting data) before they are fed into the learning model. This preliminary action prepares the data in advance, reducing the processing time required during actual diagnosis operations and improving overall system reliability through thorough data preparation.
Solution Approach 2:
The learning model continuously processes incoming data streams from multiple sensors without interruption, maintaining continuous diagnostic monitoring of the production facility. This continuous processing ensures that abnormalities are detected promptly while efficiently utilizing computational resources, balancing reliability improvement with time management.
Data Source
AI summary
A learning device and other techniques allow accurate diagnosis of a production facility. A learning device (10) includes a data acquirer that acquires data for learning, and a model generator that generates a learning model for inferring a condition of a workpiece (3) handled in a production facility (2) on the basis of the data for learning. The data for learning includes setting data indicating a setting of the production facility (2), image data indicating an image of the production facility (2) captured by a camera (4), temperature data indicating a surface temperature of the production facility (2) measured by a temperature sensor (5), distance data indicating a distance from a range sensor (6) to the production facility (2) measured by the range sensor (6), and condition data indicating the condition of the workpiece (3) handled in the production facility (2).