基于多模型融合的冷却液泄漏检测方法、装置及电子设备
By employing a multi-model fusion method for coolant leakage detection, which utilizes a self-attention prediction model and a random forest model to process engine operating data, the real-time performance and robustness issues of coolant detection in existing technologies are resolved, enabling accurate prediction and early warning of coolant levels.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHONGQING SOKON POWER CO LTD
- Filing Date
- 2025-07-04
- Publication Date
- 2026-07-17
AI Technical Summary
Existing coolant detection methods lack real-time performance and predictive capabilities, have poor environmental adaptability, and poor model robustness, leading to false alarms or missed alarms and failing to predict the trend of coolant level decline in advance.
A multi-model fusion approach is adopted, combining the self-attention prediction model Transformer-KAN and the random forest prediction model. By acquiring multiple engine operating data in real time, data processing and prediction value fusion are performed, and the problem is transformed into a classification problem to improve detection accuracy.
It enables real-time and accurate prediction of coolant level, reduces false alarms and missed alarms, improves detection efficiency and prediction accuracy, and can provide early warning of potential faults.
Smart Images
Figure CN120822178B_ABST