Dew point meter measurement data processing method and device

By constructing a dynamic sensing window and a fault-window morphology mapping library, adaptive diagnosis and correction of dew point meter measurement data are achieved, solving the problem of decision space explosion under multiple concurrent faults, improving the accuracy and real-time performance of the diagnostic system, and extending equipment life.

CN121542674BActive Publication Date: 2026-05-26HEFEI COMATE INTELLIGENT SENSOR TECH CO LTD +1
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
HEFEI COMATE INTELLIGENT SENSOR TECH CO LTD
Filing Date
2026-01-15
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing dew point meter measurement systems are unable to effectively separate and attribute abnormal symptoms when faced with multiple complex concurrent faults, resulting in a decision space explosion in the diagnostic system, making it impossible to quickly locate the root cause, and the correction effect lacks specificity.

Method used

By constructing a dynamic perception window and detecting fault probability entropy and instantaneous abnormal events, combined with a fault-window shape mapping library and a correction strategy library, an adaptive window switching and targeted data correction are achieved, forming an intelligent closed-loop diagnostic system.

Benefits of technology

By effectively decomposing complex diagnostic problems into manageable sub-problems, we can improve the accuracy of anomaly identification, reduce false alarm rates, meet real-time requirements, extend equipment lifespan, and promote intelligent operation and maintenance models.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121542674B_ABST
    Figure CN121542674B_ABST
Patent Text Reader

Abstract

This invention discloses a method and device for processing dew point meter measurement data, relating to the fields of industrial automation measurement and control and fault diagnosis technology. It constructs a dynamically sensing window with variable shape to preprocess time-series data from the dew point meter, calculates fault probability entropy in real time, detects instantaneous anomalies, and combines a fault-window shape mapping library to determine the window shape and adaptively switch it to obtain anomaly identification results. A corresponding algorithm is selected from a correction strategy library to correct the anomaly. This invention constructs a dynamically sensing window with variable shape, driven by real-time calculated fault probability entropy, enabling adaptive switching between panoramic, sniper, and multi-focus shapes. Utilizing a built-in fault-window shape mapping library, it associates anomaly symptoms with fault physical mechanisms, separates multiple concurrent faults, infers root causes, and avoids decision space explosion. Based on the diagnostic results, it calls targeted algorithms from the correction strategy library to correct anomalies, forming an intelligent closed loop of identification-diagnosis-correction.
Need to check novelty before this filing date? Find Prior Art