An edge AI gateway dedicated to online life inference. It loads trained HI and RUL models and infers in real time across many sensing points, suited to wide, large-scale life prediction deployments.
The RULe.XI series connects 16, 32 or 64 Aurora smart multivariate sensors. Model training is done centrally on a back-end AI training server (life model training module plus edge gateway management module); the gateway only collects data and runs inference. It sends damage-mode features and health data back to the server for storage, receives updated life inference models, and keeps outputting the Health Index (HI), feature contribution and Remaining Useful Life (RUL) on site. Results can be sent to FDC, SCADA or PHM systems via standard protocols such as Modbus TCP.
Key features
- Edge gateway module: sensor connection management, and collection of health data and damage-mode feature data.
- Life prediction module: HI inference and visualization, damage-mode feature contribution analysis, and RUL inference and visualization.
- Central management (with the AI training server): gateway performance monitoring, feature data collection and storage, model storage and management, and RUL model training and distribution.
- Three-color health light: green 80–100%, yellow 60–79%, red below 60%, paired with an SOP to guide maintenance.
Scenario
We suggest the Life Prediction Inference Gateway when you:
- Have many pieces of equipment (16 to 64 points) needing online life prediction;
- Want to train and manage life models plant-wide and distribute them to each edge gateway;
- Need HI and RUL sent straight to the fab FDC / SCADA, with alarms or interlocks when RUL falls below a threshold.
Models
This series is inference-only. Model training is done by the AI training server or the RULe Life Prediction (Training and Inference) Gateway:
- RULe.XI016: 16 sensors, Intel Core i3-N305, 8 GB memory, 64 GB storage, 6 serial ports, 2.5GbE x 2, 3-pin DC 9–36V, 214 x 115.5 x 65.8 mm
- RULe.XI032: 32 sensors, same specifications as XI016
- RULe.XI064: 64 sensors, same specifications as XI016
Case Study
- TSMC EFEM wafer robot life prediction (Level 4)
- FATC DISCO dicing saw spindle and blade life prediction
- FATC DISCO grinder/polisher spindle life prediction
Contact us
If you don't know whether your application is suitable for the Life Prediction Inference Gateway, feel free to contact us.