A resin regenerant formulation optimization method, device, equipment, medium and product

By optimizing the ion exchange resin regenerator formulation through machine learning models and the SHAP algorithm, the problems of regenerator formulation complexity and environmental risks have been solved, achieving efficient and safe regeneration of regenerator combinations suitable for various resins and PFAS chain lengths.

CN121415926BActive Publication Date: 2026-04-10SHANGHAI JIAOTONG UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-10-23
Publication Date
2026-04-10

AI Technical Summary

Technical Problem

In existing technologies, the optimization of ion exchange resin regenerator formulations is complex and carries environmental risks. It is difficult to achieve efficient regeneration and cannot cover all scenarios with different resin types and PFAS chain lengths. Traditional methods are inefficient and pose safety risks.

Method used

By employing machine learning models (such as gradient boosting regression models) combined with the SHAP algorithm and Bayesian optimization, key features are identified and regeneration condition combinations, including single regenerators and combinations of two regenerators, are optimized to reduce environmental risks by predicting regeneration recovery rates.

Benefits of technology

It improves regeneration efficiency, reduces environmental risks, is suitable for PFAS-resin systems in different scenarios, reduces the use of high-concentration organic solvents, and lowers safety risks and costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121415926B_ABST
    Figure CN121415926B_ABST
Patent Text Reader

Abstract

The application discloses a resin regenerant formula optimization method, device, equipment, medium and product, relates to the field of formula optimization, and comprises the following steps: acquiring actual influence factors; predicting the regeneration recovery rate by using a regeneration recovery rate prediction model according to the perfluoroalkyl compound structure characteristics, the resin properties and the regeneration conditions; performing feature importance sorting on the perfluoroalkyl compound structure characteristics, the resin properties and the regeneration conditions by using a SHAP summary graph based on the regeneration recovery rate prediction model, obtaining key features, and selecting a key feature optimal value range based on a SHAP dependence graph; and searching for a regeneration condition combination corresponding to the maximum regeneration recovery rate in the key feature optimal value range by using an optimization algorithm; and the regeneration condition combination comprises a single regenerant and a two-by-two regenerant combination. The application can realize directional setting of a regenerant formula, improve regeneration efficiency and reduce environmental risks at the same time.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Resin synthesis method and system based on machine learning, medium and product

    CN118430675A

  • A prediction model for valproic acid trough concentration in adults based on machine learning and its application

    CN119763717A