A machine-learning tool learns from sensors' data of a nuclear reactor at steady state and maps them to controls of the nuclear reactor. The tool learns all given ranges of normal operation and responses for corrective measures. The tool may train another learning tool (or the same tool) that forecasts the behavior of the reactor based on real-time changes (e.g., every 10 seconds). The tool implements an optimization technique for differing half-life materials to be placed in the reactor. The tool maximizes isotope production based on optimal controls of the reactor.

A method to determine a global core reactivity bias and the corresponding estimated critical conditions of a nuclear reactor core prior to achieving reactor criticality. The method first requires collection and evaluation of the inverse count rate ratio (ICRR) data; specifically, fitting measured ICRR vs. predicted ICRR data. The global core reactivity bias is then determined as the amount of uniform reactivity adjustment to the prediction that produces an ideal comparison between the measurement and the prediction.

Surrogate materials are in the form of solid particles that include surrogate isotopes, namely, short-lived isotopes selected and formed to serve as surrogates for the radioactive materials of a nuclear fallout without including isotopes that are, or that decay to, biologically or environmentally deleterious and persistent isotopes. The surrogate material may be formed using high-purity reactant material and irradiation and separation techniques that enable tailoring of the isotopes and ratios thereof included in the surrogate material, and the surrogate material may be dispersed, e.g., in a training environment, in solid form.

Illustrative embodiments provide for the operation and simulation of the operation of fission reactors, including the movement of materials within reactors. Illustrative embodiments and aspects include, without limitation, nuclear fission reactors and reactor modules, including modular nuclear fission reactors and reactor modules, nuclear fission deflagration wave reactors and reactor modules, modular nuclear fission deflagration wave reactors and modules, methods of operating nuclear reactors and modules including the aforementioned, methods of simulating operating nuclear reactors and modules including the aforementioned, and the like.

A method of determining a core design parameter of a nuclear reactor, includes: calibrating an isolated voltage output from a NIS cabinet associated with the nuclear reactor using a calibrated signal source as an input to the NIS cabinet; recording values of the calibrated signal source used in the calibrating and corresponding values of the output voltage from the calibrating in an as-left cabinet calibration data table; using a computing device connected to the isolated voltage output from the NIS cabinet, converting the voltage output signal to a converted detector signal using at least some of the values in the as-left cabinet calibration data table in an improved signal conversion equation; and using the computing device, employing the converted detector signal to determine the core design parameter.

A method is used to design nuclear reactors using design variables and metric variables. A user specifies ranges for the design variables and target values for the metric variables. A set of design parameter samples are selected. For each sample, the method runs three processes, which compute metric variables to thermal-hydraulics, neutronics, and stress. The method applies a cost function to each sample to compute an aggregate residual of the metric variables compared to the target values. The method trains a machine learning model using the samples and the computed aggregate residuals. The method shrinks the range for each design variable according to correlation between the respective design variable and estimated residuals using the machine learning model. These steps are repeated until a sample having a smallest residual is unchanged for multiple iterations. The method then uses the final machine learning model to assess relative importance of each design variable.

According to an embodiment, a design method for a light-water reactor fuel assembly comprises: accumulating a determined fuel data, showing that each of a combination of p.Math.n/N and e is feasible as the core or not, wherein N is a number of the fuel rods in the fuel assembly, n is a number of the fuel rods containing the burnable poison, p is a ratio wt % of the burnable poison in the fuel, and e is an enrichment wt % of the uranium 235 contained in the fuel assembly; formulating a criterion formula which determines whether a combination of p.Math.n/N and e is feasible as a core or not and is formulated based on the determined fuel data; and determining whether a temporarily set composition of the fuel assembly is approved as a core or not based on the criterion formula.

A method of controlling a nuclear power plant includes obtaining sensor data from one or more sensors of the nuclear power plant, providing the sensor data and a desired plant response to a neural network, wherein the neural network has been previously trained using a simulated nuclear power plant and is structured to determine at least one control system setting to achieve the desired plant response, determining at least one control system setting to achieve the desired plant response with the neural network, and setting or changing at least one control system setting of a control system of the nuclear power plant to the at least one control system setting determined by the neural network.

A covariance data creation apparatus configured to execute assembly calculations on a fuel assembly based on microscopic cross sections, the apparatus executing: a perturbation data generation step of deriving a plurality of perturbation quantities of the microscopic cross sections based on microscopic covariance data that is data regarding uncertainties of the microscopic cross sections, and generating microscopic perturbation data from the derived perturbation quantities of the microscopic cross sections; a macroscopic cross section derivation step of executing the assembly calculations based on the microscopic perturbation data generated at the perturbation data generation step, and deriving a plurality of macroscopic cross sections individually corresponding to the perturbation quantities of the microscopic cross sections; and a macroscopic covariance data generation step of generating macroscopic covariance data that is data regarding uncertainties of the macroscopic cross sections based on the macroscopic cross sections derived at the macroscopic cross section derivation step.

A method designs nuclear reactors using design variables and metric variables. A user specifies ranges for the design variables and threshold values for the metric variables and selects design parameter samples. For each sample, the method runs three processes, which compute metric variables for thermal-hydraulics, neutronics, and stress. The method applies a cost function to compute an aggregate residual of the metric variables compared to the threshold values. The method deploys optimization methods, either training a machine learning model using the samples and computed aggregate residuals, or using genetic algorithms, simulated annealing, or differential evolution. When using Bayesian optimization, the method shrinks the range for each design variable according to correlation between the respective design variable and estimated residuals using the machine learning model. These steps are repeated until a sample having a smallest residual is unchanged for multiple iterations. The final model assesses relative importance of each design variable.