Nuclear power plant distributed control system and control method

By constructing a three-level control architecture consisting of a global optimization layer, a coordination control layer, and an execution control layer, and utilizing transfer learning and multi-agent attention networks, the problem of multi-objective conflict in nuclear power plants was solved, and dynamic coordination of power generation efficiency, equipment lifespan, and safety margin was achieved.

CN121165677AActive Publication Date: 2025-12-19CHINA NUCLEAR CONTROL SYST ENG
6 Cites 1 Cited by

Patent Information

Application Number
CN202511695838.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-11-19
Publication Date
2025-12-19
Estimated Expiration
2045-11-19

AI Technical Summary

Technical Problem

Traditional proportional-integral-derivative (PID) control strategies are difficult to effectively balance the multi-objective conflicts between power generation efficiency, equipment lifespan, and safety margin in nuclear power plants, especially when there are sudden load changes, which may lead to equipment stress overload and reduced safety margin.

Method used

A global optimization layer is adopted to generate a cross-unit general control strategy library through transfer learning algorithm. The coordination control layer uses a multi-agent attention network to dynamically allocate the control weights of the two-loop subsystems. The execution control layer calculates the safety constraint matrix through the safety manifold projection module, and a three-level control architecture is constructed to coordinate multiple objectives.

Benefits of technology

It enables a dynamic balance between power generation efficiency, equipment lifespan, and safety margin in nuclear power plants based on actual operating conditions, avoiding the limitations of traditional PID control strategies and improving the robustness and adaptability of the system.

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

Abstract

The invention discloses a nuclear power plant distributed control system and control method, and relates to the technical field of nuclear power plant digital control, and the method comprises the steps: a global optimization layer generates a cross-unit general control strategy library based on a transfer learning algorithm according to the decision period of the global optimization layer; when the coordination control layer operates by using the transferable control strategy corresponding to the unit equipment where each two-loop subsystem is located, the corresponding two-loop subsystem control weight is dynamically allocated based on the multi-agent attention network according to the decision period of the coordination control layer; when the execution control layer operates by using the dynamically distributed two-loop subsystem control weight, the first-loop temperature, the first-loop pressure and the coolant flow are obtained according to the decision period of the control layer, a safety constraint matrix is calculated through a safety manifold projection module, and a quadratic programming problem with a relaxation factor is solved; and the corrected control weight of the two-loop subsystem is obtained and converted into a control instruction to be issued to a DCS field control station.
Need to check novelty before this filing date? Find Prior Art