一种黄土粘度的检测方法及系统
By using a long short-term memory neural network model based on the meta-learning algorithm MAML and combining it with pit data to establish a loess viscosity detection model, the quality problem of pits caused by neglecting environmental factors in existing detection methods is solved, achieving more accurate and efficient loess selection and improving the construction quality of pits.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JINAN BAOTU SPRING BREWING CO LTD
- Filing Date
- 2023-07-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for testing loess viscosity fail to consider factors such as temperature, sand content, and pH in the cellar environment, resulting in loess selection that does not meet the requirements of the cellar and affecting the quality of the cellar.
A long short-term memory neural network model based on the meta-learning algorithm MAML was adopted. Combined with data on loess viscosity, particle size, sand content, temperature and pH value of the pit, a dataset was established and trained to find the optimal initialization parameters and construct a loess viscosity detection model.
It improves the accuracy and efficiency of loess viscosity testing, provides standardized guidance for the construction of pits, ensures that loess selection meets the requirements of pits, and improves the quality of pits.
Smart Images
Figure CN116680574B_ABST