A sewage epidemiology-based spatiotemporal multi-dimensional sewage drug concentration prediction method, device, storage medium and product

By constructing a multi-dimensional wastewater drug concentration prediction method based on Informer encoder and graph attention network, the problems of insufficient spatiotemporal coupling and noise processing in existing technologies are solved, and high-precision wastewater drug concentration prediction and classification assessment are achieved. This method is applicable to drug abuse monitoring and public health decision-making in wastewater epidemiology.

CN122201502APending Publication Date: 2026-06-12CHINA PHARM UNIV
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
CHINA PHARM UNIV
Filing Date
2026-03-25
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

Existing wastewater drug concentration prediction methods are inadequate in handling complex spatiotemporal coupling, noise and missing data, and insufficient physical correlation expression, leading to prediction bias and instability.

⚗Method used

A single-step regression model based on an Informer encoder is used for time dimension prediction, combined with a multi-layer graph attention network for spatial dimension prediction, and a spatiotemporal coupling prediction model is used to fuse time and spatial features to construct a multi-dimensional wastewater drug concentration prediction method.

🎯Benefits of technology

It improves the accuracy and stability of wastewater drug concentration prediction, and realizes the linkage between concentration prediction and classification assessment, which is applicable to drug abuse monitoring and public health decision-making in wastewater epidemiology.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
Patent Text Reader

Abstract

The application discloses a sewage epidemic-based spatiotemporal multi-dimensional sewage drug concentration prediction method and device, a storage medium and a product. The method faces multi-site sewage monitoring data, and based on the latitude and longitude information of the sites and monitoring information such as workdays or weekends, air temperature, humidity, air pressure, daily flow, domestic sewage proportion, service population and drug concentration, respectively constructs a time dimension prediction model, a space dimension prediction model and a spatiotemporal coupling prediction model; each model outputs a drug concentration prediction value at the next time, and according to the daily flow, the service population and the preset threshold conversion, obtains the consumption per thousand people and the threshold exceeding classification result.
Need to check novelty before this filing date? Find Prior Art