A gas pressure regulator fault diagnosis method based on difference perception anti-decoupling
The gas pressure regulator fault diagnosis method based on differential perception and adversarial decoupling utilizes outlet pressure and instantaneous flow signals to construct an end-to-end fault diagnosis model. This solves the problems of online monitoring and multi-category identification in the existing technology for gas pressure regulator fault diagnosis, and achieves high-precision and robust fault diagnosis results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SOUTHWEST PETROLEUM UNIV
- Filing Date
- 2026-06-15
- Publication Date
- 2026-07-17
AI Technical Summary
Existing fault diagnosis methods for gas pressure regulators are difficult to achieve online monitoring and multi-category fault identification. In particular, there is insufficient research on the enhancement of weak fault features, and the model generalization ability is insufficient, making it unable to meet the diagnostic needs under complex operating conditions.
By employing a differential perception adversarial decoupling method, an end-to-end fault diagnosis model is constructed using dual-channel time-series data. By utilizing outlet pressure and instantaneous flow signals, combined with differential perception feature enhancement and operating condition adversarial branching, deep decoupling of fault features and operating condition features is achieved, thereby improving diagnostic accuracy and robustness.
It achieves high-precision and robust fault diagnosis of gas pressure regulators, and can monitor and identify multiple types of faults online. It is especially effective in diagnosing minor faults under variable flow load conditions, improving the model's generalization ability and diagnostic accuracy.
Smart Images
Figure CN122413084A_ABST