An intelligent prediction method for centrifugal pump performance considering performance constraints

CN114398838BActive Publication Date: 2025-04-15CHINA JILIANG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210149668.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-02-18
Publication Date
2025-04-15
Estimated Expiration
2042-02-18

AI Technical Summary

Technical Problem

The prior art is difficult to accurately predict the performance of centrifugal pumps under multiple operating conditions, especially in non-designed operating conditions, with large prediction errors and failure to effectively consider the relationship between head, power and efficiency.

Method used

A centrifugal pump performance intelligent prediction method considering performance constraints is adopted. By collecting multi-case data, a multi-output regression model is constructed, and a support vector regression model and particle swarm optimization algorithm are used to optimize model parameters to reduce the constraint error between performance parameters.

Benefits of technology

Accurate prediction of the head, power and efficiency of centrifugal pumps under multiple operating conditions is achieved, reducing prediction errors, meeting engineering practice requirements, and reducing designer operation complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114398838B_ABST
    Figure CN114398838B_ABST
Patent Text Reader

Abstract

The present invention discloses a centrifugal pump performance intelligent prediction method considering performance constraints. The method collects geometric parameters of the impeller and the water pressure chamber of the centrifugal pump with different specific speeds, and the real values ​​of the head, efficiency and power under different flow rates, and respectively constructs a training set and a test set; then, based on the constraint relationship of the performance parameters, a multi-output regression model is constructed and trained; the geometric parameters and flow rate of the impeller and the water pressure chamber of the centrifugal pump in the test set are input into the trained multi-output regression model to obtain the predicted values ​​of two performance parameters; according to the constraint relationship, the predicted value of another performance parameter is calculated; the minimum value of the error between the predicted value and the real value of the calculated performance parameter is used as the objective function of the heuristic optimization algorithm, and a multi-output regression model with optimal structural parameters is obtained, thereby obtaining the predicted value of the unknown performance parameter. The method of the present invention accurately and reasonably predicts the three performances of the centrifugal pump under multiple working conditions.
Need to check novelty before this filing date? Find Prior Art