Method for determining fluid density of fluid in an encapsulated electrical device
By using machine learning and digital models to correct the measured values of fluid density in electrical equipment, the problem of measurement deviation under the influence of weather conditions has been solved, enabling accurate and rapid monitoring of fluid density and improving the safety and maintainability of electrical equipment.
Patent Information
- Application Number
- CN202180034088.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2020-04-09
- Filing Date
- 2021-03-11
- Publication Date
- 2026-01-16
- Estimated Expiration
- 2041-03-11
AI Technical Summary
Existing technologies are insufficient to reliably monitor and correct changes in fluid density in electrical equipment in the short term, especially due to measurement deviations caused by weather conditions, which affect the operational safety and maintainability of electrical equipment.
Machine learning methods are employed to collect measurement data using sensors and train a digital model by combining it with weather data. The measured fluid density is then corrected using an artificial neural network (such as a recurrent neural network LSTM). The influence of weather conditions on the measurement is taken into account, and data cloud is used for calculation and correction.
It significantly improves the accuracy and speed of fluid density measurement, enabling timely detection of fluid density changes and enhancing the operational safety and maintainability of electrical equipment.
Smart Images

Figure CN115516592B_ABST
Abstract
Citation Information
Patent Citations
Gas compensation based full-temperature SF6 gas density relay checking device
CN102879733A
Wind power prediction method and system
CN104200280A