Edge Abnormality Detection Using Cloud-Selected Algorithms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current device-abnormality detection systems face challenges with high communication traffic and energy consumption due to large-scale data transmission between local and cloud terminals, which also results in slow response times and high latency in abnormality diagnosis.
Innovation Solution
A device-abnormality detection method and system that involves generating target performance index data by a local detection terminal and sending it to a cloud platform, where the cloud determines and sends back target detection algorithm information. The local terminal then performs abnormality detection using this information, reducing the need for large-scale data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If large amounts of real-time detection data are transmitted from local detection terminal to cloud terminal, then comprehensive processing and abnormality detection can be performed, but communication traffic and energy consumption increase significantly
Solution Approach 1:
The detection system is segmented into two parts: local preprocessing at the edge terminal (filtering, feature extraction) and cloud-based comprehensive analysis. This segmentation reduces the data volume transmitted to the cloud while preserving the essential information needed for accurate abnormality detection, thereby reducing energy consumption without sacrificing detection accuracy.
Solution Approach 2:
The local detection terminal performs preliminary processing actions on detection data before transmission, including filtering out normal data and extracting key features. This preliminary action reduces the amount of data that needs to be transmitted and processed by the cloud terminal, thereby reducing communication traffic and energy consumption while maintaining detection effectiveness.
2Measurement precision
If large amounts of raw data are processed in cloud terminal, then comprehensive abnormality detection can be achieved, but computation amount and data processing time increase
Solution Approach 1:
The processing workflow is segmented into local preprocessing (filtering, feature extraction) and cloud-based analysis. By dividing the processing tasks, the system reduces the computational burden on the cloud terminal and enables faster response times while maintaining comprehensive abnormality detection capabilities through the combination of local and cloud processing results.
Solution Approach 2:
The local detection terminal performs preliminary processing actions (filtering normal data, extracting features) before data reaches the cloud terminal. This preliminary action reduces the computation amount required at the cloud terminal and accelerates the overall response time, as the cloud terminal receives pre-processed data ready for rapid analysis.
3Measurement precision
If real-time transmission of large amounts of data is performed, then comprehensive detection can be achieved, but network resources and communication cost increase
Solution Approach 1:
The data transmission process is segmented into two stages: local filtering and feature extraction (reducing data volume) followed by selective transmission of only necessary data to the cloud terminal. This segmentation maintains the quality and completeness of transmitted data for accurate detection while significantly reducing the overall data transmission volume and associated network resource consumption.
Solution Approach 2:
The local detection terminal performs preliminary data filtering and feature extraction before transmission, removing redundant normal data and retaining only critical information. This preliminary action reduces the quantity of data that needs to be transmitted over the network, thereby reducing communication costs and network resource usage while preserving detection accuracy.
Data Source
AI summary
A device-abnormality detection method and a device-abnormality detection system are disclosed. The device-abnormality detection method includes: generating, by a local detection terminal, target performance index data of a device to be detected, and sending the target performance index data to a cloud platform. The method may include determining, by the cloud platform based on the target performance index data, target detection algorithm information corresponding to the target performance index data, and sending the target detection algorithm information to the local detection terminal. The method may also include implementing, by the local detection terminal, an abnormality detection of the device to be detected based on the target detection algorithm information and performance data of the device to be detected, and outputs an abnormality detection result.