Pressurized Water Reactor Control Trajectories for Multi-Goal Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional control technologies for pressurized water nuclear reactors face limitations in managing complex dynamics, long time-scale processes, and multiple control goals, making it difficult to balance and optimize operational safety, burnup, power density, and other factors simultaneously.
Innovation Solution
A method using computerized calculus of variations to predict and optimize reactor control trajectories over a 24-hour period, balancing multiple control goals by iteratively evaluating actuating variable trajectories based on a figure of merit system, allowing for real-time and comprehensive planning of reactor actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional control technology is used to govern the reactor, then real-time control of thermal power and coolant temperature is achieved, but the system cannot simultaneously balance multiple control goals including operational safety, burnup, power density distribution, and long-term operation
Solution Approach 1:
The control system segments the reactor governance into distinct functional modules: a predictor module that forecasts state variables, a navigator module that generates actuator trajectories, and an optimizer module that selects optimal trajectories based on multiple criteria. This segmentation allows each module to specialize in specific aspects of control, enabling comprehensive multi-goal optimization without overwhelming complexity.
Solution Approach 2:
The system performs preliminary prediction of reactor state variables and pre-calculates optimal actuator trajectories before actual control actions are executed. The predictor module forecasts future states based on current conditions, and the navigator module generates planned trajectories in advance, allowing the system to prepare and execute coordinated control actions that balance multiple goals simultaneously.
2Duration of action of moving object
If the reactor is controlled with focus on operational safety and burnup compensation, then reactor stability is improved, but the system lacks flexibility for long-term operation planning beyond real-time control
Solution Approach 1:
The system generates and displays comprehensive control plans for extended time horizons (24 hours or more) in advance. The navigator module creates detailed actuator trajectories that span long periods, and the optimizer module evaluates these trajectories against multiple criteria including safety, burnup compensation, and operational flexibility. This preliminary planning capability allows operators to view and understand long-term control strategies without complexity, as the system handles the computational burden of long-horizon optimization.
Solution Approach 2:
The system continuously monitors actual reactor states and compares them with predicted trajectories, adjusting the control plan as needed. The feedback mechanism allows the system to refine long-term predictions and trajectories based on actual performance, maintaining accuracy and adaptability over extended operation periods while keeping the interface simple for operators.
3Adaptability or versatility
If multiple control goals are considered simultaneously, then comprehensive reactor governance is achieved, but the computational complexity and time required for optimization increases
Solution Approach 1:
The system performs preliminary predictions of reactor state variables and pre-evaluates multiple potential actuator trajectories before final optimization. The predictor module forecasts future states in advance, and the optimizer module can quickly evaluate predetermined trajectories against multiple criteria, significantly reducing the computational time required to consider numerous control goals simultaneously.
Solution Approach 2:
The system creates and evaluates simplified representations or copies of potential control trajectories rather than exhaustively solving the full optimization problem from scratch. By working with approximated trajectory models and using efficient evaluation algorithms, the system can assess multiple control scenarios rapidly, balancing comprehensive goal consideration with computational efficiency.
Data Source
AI summary
A method of governing a pressurized water nuclear reactor can simultaneously consider and balance a large number of control goals. The method includes iteratively considering a large number of randomly varied possible trajectories (Ta) of actuating variables for controlling reactor core reactivity for a future time interval. Each trajectory (Ta) of actuating variables is assigned a figure of merit (Σ) on the basis of a Value Table which contains weighting or penalty values for a number of events or adverse reactor core states which are characterized by preset conditions or values of the actuating variables, the process variables and/or variables derived from them. The trajectory (Ta) of actuating variables is chosen such that the figure of merit (Σ) has a local extremum, and corresponding actuators are moved accordingly.