A method for imputing missing data in regional industrial water efficiency assessment

By fitting the Gompadour curve using the difference method and Fourier transform combined with the least squares method, the problem of missing data in the evaluation of regional industrial water use efficiency was solved, and more scientific and accurate evaluation results were achieved.

CN122309940APending Publication Date: 2026-06-30ZHEJIANG INST OF HYDRAULICS & ESTUARY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
ZHEJIANG INST OF HYDRAULICS & ESTUARY
Filing Date
2026-04-07
Publication Date
2026-06-30

AI Technical Summary

Technical Problem

The regional industrial water use efficiency assessment suffers from inconsistent data quality and missing data, affecting the scientific validity and accuracy of the assessment results.

Method used

The index data is preprocessed using the difference method, and the frequency with the largest contribution intensity in the frequency domain is extracted using Fourier transform to determine the period. The Gompaz curve is then fitted using the least squares method to predict the baseline value of the missing data, and corrections are made based on the periodic fluctuation pattern.

Benefits of technology

This improves the scientific rigor and accuracy of industrial water efficiency assessments, reduces the impact of data integrity issues on assessment results, and ensures that assessment results are more scientific and reasonable.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122309940A_ABST
    Figure CN122309940A_ABST
Patent Text Reader

Abstract

This invention discloses a method for missing data completion in regional industrial water use efficiency assessment. The method includes: acquiring an index dataset for regional industrial water use efficiency assessment and preprocessing the index data in the dataset using the difference method; using Fourier transform to transform the preprocessed index data from the time domain to the frequency domain, identifying and extracting the frequency with the strongest contribution from the frequency domain index data to determine the period of the index data; based on the period of the index data, selecting data within the period to calculate the mean, forming a new dataset; fitting a Gompattz curve to this dataset using the least squares method; based on the Gompattz curve, predicting the baseline value for the period corresponding to the missing value; and correcting the baseline value according to the periodic fluctuation pattern to obtain the predicted value of the missing data. This provides data support for the implementation of industrial water use efficiency assessment methods, reduces the impact of data integrity issues on the assessment, and ensures that the assessment results are more scientific, accurate, and reasonable.
Need to check novelty before this filing date? Find Prior Art