Industrial Anomaly Diagnosis With Selective Sensor Data Upload
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing diagnostic devices face challenges in efficiently uploading large sensor data from industrial devices to servers due to network bandwidth constraints and security concerns, while also experiencing increased user load during data labeling, especially when dealing with a large number of target data points.
Innovation Solution
A diagnostic device and method that selectively transmits only influential sensor data to a server by using a classification learning model to diagnose anomalies and determine which data to upload, reducing network load and user burden through intelligent data selection and labeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all sensor data is uploaded to the server, then the server can perform comprehensive learning and improve diagnosis accuracy, but network bandwidth is consumed and security risks increase
Solution Approach 1:
The patent extracts and transmits only the essential elements (anomaly detection results and classification labels) from the complete sensor data to the server, rather than uploading all raw sensor data. This extraction approach maintains diagnosis accuracy while significantly reducing network bandwidth consumption.
Solution Approach 2:
The diagnostic device acts as an intermediary between the sensor data and the server, performing local anomaly detection and classification before transmitting results. This intermediary processing filters out unnecessary data, reducing network load while preserving critical information for server learning.
2Measurement precision
If all sensor data is uploaded to the server, then comprehensive learning can be performed, but user load increases during data labeling
Solution Approach 1:
The system extracts only the critical anomaly information and classification results for user labeling, rather than requiring users to label all sensor data. This dramatically reduces the user labeling burden while maintaining learning effectiveness.
Solution Approach 2:
The diagnostic device automatically performs anomaly detection and classification using the classification learning model, generating pre-processed data that requires minimal user intervention for labeling. This self-service approach reduces user load while preserving learning quality.
3Measurement precision
If classification learning model is continuously updated with all data, then model accuracy improves, but network transmission load increases
Solution Approach 1:
The system extracts only the essential learning elements (anomaly instances with classification labels) from the complete sensor data stream for transmission to the server. This extraction enables continuous model updating with high-value data while minimizing transmission volume.
Solution Approach 2:
The system transforms the data representation from raw sensor signals to structured anomaly records with classification labels. This parameter transformation condenses large volumes of raw data into compact, high-information-density records suitable for efficient transmission and model learning.
Data Source
AI summary
A diagnostic device, which is communicatively connected to a server that learns an abnormality of an industrial apparatus and generates a classification learning model for the abnormality, includes a sensor signal acquisition unit that acquires a sensor signal including a measurement value measured by a sensor in the industrial apparatus. The diagnostic device also includes an apparatus diagnosis unit that diagnoses whether the industrial apparatus is normal or abnormal based on the acquired sensor signal, a classification unit that classifies the abnormality based on the sensor signal and the classification learning model, when the industrial apparatus is abnormal, and a transmission unit that determines whether to transmit the sensor signal to the server based on the diagnosis result and/or and the classification result, and transmits the sensor signal to the server when it is determined that the sensor signal can be transmitted.