一种城市中未来PM2.5浓度预测方法和系统
By combining the spatiotemporal attention mechanism of temporal convolutional networks and long short-term memory networks, a combined model is constructed, which solves the complexity and error problems of PM2.5 concentration prediction in existing technologies and achieves efficient and accurate urban air quality prediction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XINJIANG UNIVERSITY
- Filing Date
- 2024-03-21
- Publication Date
- 2026-07-17
AI Technical Summary
Existing PM2.5 concentration prediction models suffer from high complexity, large computational load, large prediction errors, and insufficient applicability and universality, making it difficult to accurately predict urban air pollution.
A combined model is constructed by using a temporal convolutional network (TCN) and a long short-term memory network (LSTM) with a spatiotemporal attention mechanism. Features are selected by Pearson correlation coefficient, and data preprocessing and feature extraction are performed. The results are predicted using an adaptive weighted fusion model based on root mean square error.
It improves the accuracy and efficiency of PM2.5 concentration prediction, reduces model complexity, enhances the reliability and applicability of prediction, and achieves high-precision urban air quality prediction.
Smart Images
Figure CN118366564B_ABST