Channel PI parameter setting method and system based on context Bayesian optimization

By employing a context-based Bayesian optimization-based channel PI parameter tuning method, which utilizes operating condition context modeling and stability feasible region constraints, the problem of low PI parameter tuning efficiency in long-distance open channel water conveyance systems is solved, achieving efficient and safe control under varying operating conditions.

CN122308046AActive Publication Date: 2026-06-30WUHAN UNIV
4 Cites 0 Cited by

Patent Information

Application Number
CN202610413463.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-31
Publication Date
2026-06-30
Estimated Expiration
2046-03-31

AI Technical Summary

Technical Problem

Existing PI parameter tuning methods are difficult to balance system stability, safety and optimization efficiency under varying operating conditions in long-distance open channel water conveyance systems, and do not fully utilize the correlation information between different operating conditions, resulting in low parameter tuning efficiency.

Method used

A channel PI parameter tuning method based on contextual Bayesian optimization is adopted. By combining operating condition context modeling, stability prior learning and stability feasible region constraints, and offline and online optimization, a stability knowledge base and Gaussian process model are constructed to achieve adaptive tuning of PI control parameters.

Benefits of technology

It improves the efficiency of parameter tuning under new operating conditions, ensures system stability and safety, and optimizes efficiency while ensuring control accuracy and stability. The tuning method is continuously improved through data closed-loop updates.

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

Abstract

The invention discloses a channel PI parameter setting method and system based on context Bayesian optimization, and the method comprises the steps: firstly, carrying out the Bayesian optimization with the minimization of stabilization time as a target under a plurality of typical working conditions, constructing an offline stability knowledge base fusing the context information of the working conditions, training a context-aware stability prediction Gaussian process model; secondly, for a target working condition, based on model prediction uncertainty, adaptively selecting a knowledge migration or local exploration path, and constructing a stability feasible region; then, in the feasible region, dynamically constructing a comprehensive performance objective function and carrying out Bayesian optimization by combining channel water level control precision and gate action stability indexes, so as to obtain an optimal PI (Proportional Integral) parameter; and finally, realizing continuous updating of the knowledge base and the model through a data closed loop. According to the method, safe, efficient and self-adaptive setting of the channel PI parameters under the variable working conditions is achieved, and the operation stability of a water conveying system is fundamentally guaranteed.
Need to check novelty before this filing date? Find Prior Art