End-to-end prediction method for dust collection efficiency of wet dust collector

By employing a multimodal image feature fusion and dynamic weight allocation strategy, an end-to-end prediction model was constructed, which solved the error and adaptability problems in the monitoring of dust removal efficiency of wet dust collectors, and achieved high-precision and robust dust removal efficiency monitoring.

CN121837883AActive Publication Date: 2026-04-10CHINA ACAD OF SAFETY SCI & TECH +1
4 Cites 0 Cited by

Patent Information

Application Number
CN202610057935.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-01-16
Publication Date
2026-04-10
Estimated Expiration
2046-01-16

AI Technical Summary

Technical Problem

Existing methods for monitoring the dust removal efficiency of wet scrubbers suffer from inherent errors and limitations in feature extraction, making it impossible to achieve high-precision and robust real-time monitoring.

Method used

By employing a multimodal image feature fusion and dynamic weight allocation strategy, one-dimensional pressure signals are converted into images and an end-to-end prediction model is constructed. The most effective feature combination is then selected to achieve high-precision and robust dust removal efficiency monitoring.

Benefits of technology

It achieves high-precision and robust dust removal efficiency monitoring, with an accuracy rate of 97.5%. It performs excellently under complex working conditions and has strong adaptability, solving the problems of error and poor adaptability in traditional methods.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses an end-to-end prediction method for the dust removal efficiency of a wet dust collector, and the method comprises the steps: firstly obtaining an actual pressure fluctuation signal, dividing the actual pressure fluctuation signal into a plurality of intervals according to the size of the dust removal efficiency, and selecting actual pressure fluctuation signal segments corresponding to the actual dust removal efficiency in different intervals through the actual pressure fluctuation signal corresponding to the actual dust removal efficiency; respectively converting the signal segments in different intervals into images by adopting two conversion methods; fusing the images by adopting different image feature fusion methods to form an image data set; a set network model is adopted to train different image data sets of each interval, then different initial discrimination models are formed, discrimination weights of the different initial discrimination models in each interval are dynamically calculated according to accuracy rates of the different initial discrimination models in each interval, performance advantages of each model in a specific interval are utilized to the maximum extent, and the discrimination accuracy of the different initial discrimination models in the specific interval is improved. And a dust removal efficiency fusion prediction model is obtained, so that high-precision and high-robustness dust removal efficiency monitoring is realized.
Need to check novelty before this filing date? Find Prior Art