A secondary loop thermal parameter soft measurement method and system

By constructing mechanistic features and deep neural network models in steam generators, and combining mass conservation relations and composite loss functions, the accuracy problem of measuring thermal parameters in steam generators was solved, achieving high-precision and physically interpretable thermal parameter measurements, adapting to the complex operating conditions of nuclear power plants.

CN122364901APending Publication Date: 2026-07-10HARBIN ENG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
HARBIN ENG UNIV
Filing Date
2026-03-17
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

Existing technologies for measuring key thermodynamic parameters of steam generators frequently exhibit zero-point drift under high temperature, high pressure, and strong radiation environments, and are prone to false water level phenomena during transient processes, leading to inaccurate measurement results and misleading the control system.

Method used

By collecting multidimensional physical quantities in real time, constructing mechanistic features and introducing a deep neural network model, and combining the mass conservation relationship and composite loss function for training, a transfer adaptation mechanism is adopted to quickly adapt when the working conditions change, ensuring that the prediction results conform to the physical laws.

Benefits of technology

It achieves high-precision, physically interpretable thermodynamic parameter measurement, possesses strong robustness and low-cost cross-condition reuse capability, and adapts to the complex operating conditions of nuclear power plants.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122364901A_ABST
    Figure CN122364901A_ABST
Patent Text Reader

Abstract

A soft measurement method and system for secondary thermal parameters is disclosed, relating to the field of nuclear power plant secondary circuit operation monitoring and intelligent maintenance. The aim is to address the difficulty of achieving accurate measurement of key thermal parameters in existing steam generator technologies. The method includes: real-time acquisition of measured values ​​of steam generator water level, main steam pressure, feedwater flow rate, and regulating valve opening commands to form an initial sequence; obtaining characteristics of steam generator water level change rate, main steam pressure gradient, and instantaneous mass imbalance based on the acquired data to form a derived sequence; concatenating the initial and derived sequences to obtain a hybrid feature vector; and inputting the hybrid feature vector into a deep neural network model to obtain predicted values ​​of steam generator water level and feedwater flow rate. This model is constrained by the mass conservation principle during training.
Need to check novelty before this filing date? Find Prior Art