A parameter optimization method for collaborative stress of a disassembly-free composite formwork and cast-in-place concrete

By constructing a spatial distribution probability model of connector position deviation using a weighted Gaussian process regression model and an active learning strategy, and combining stochastic finite element simulation and Bayesian optimization algorithm, the problem of decreased collaborative stress performance between the non-removable composite formwork and cast-in-place concrete caused by construction deviation was solved, achieving more efficient parameter optimization and improved stress performance.

CN122197640APending Publication Date: 2026-06-12YILI ZHENGHUA PLASTIC IND CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
YILI ZHENGHUA PLASTIC IND CO LTD
Filing Date
2026-04-29
Publication Date
2026-06-12

AI Technical Summary

Technical Problem

Existing parameter optimization methods have difficulty accurately inverting the global deviation field when dealing with construction deviations, resulting in a decrease in the collaborative stress performance of the non-removable composite formwork and cast-in-place concrete.

Method used

By combining a weighted Gaussian process regression model with an active learning strategy, a spatial distribution probability model of connector position deviation is constructed. The failure probability is calculated through stochastic finite element simulation, and key design parameters are identified by iterative optimization using a Bayesian optimization algorithm.

Benefits of technology

It significantly improves the accuracy and adaptability of parameter optimization, dynamically adjusts design parameters to adapt to construction deviations, and enhances the synergistic stress performance of the non-removable composite formwork and cast-in-place concrete.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122197640A_ABST
    Figure CN122197640A_ABST
Patent Text Reader

Abstract

The application discloses a parameter optimization method for collaborative stress of a disassembly-free composite formwork and cast-in-place concrete, and particularly relates to the technical field of machine learning assisted engineering optimization, and is used for solving the problem that the optimization result deviates from the engineering practice due to the complex construction deviation distribution and the sparse measured data in the existing parameter optimization method; the sparse measured data and the design parameters of the connecting piece position are acquired, the formwork area sensitive to the deviation is identified, and a weighted Gaussian process regression model is constructed as a spatial distribution probability model of the deviation; an active learning strategy is adopted to supplement the measuring points to update the model; the updated model is input into a stochastic finite element simulation to calculate the failure probability under different design parameter combinations; the gradient information of the failure probability with respect to each design parameter is calculated and coupled with the deviation uncertainty measure to obtain a coupling influence coefficient to identify key design parameters; finally, a Bayesian optimization algorithm is used to iteratively optimize the key design parameters with the minimization of the failure probability as the target.
Need to check novelty before this filing date? Find Prior Art