HTGR Research Articles

DESIGN OF HELICAL TYPE STEAM GENERATOR FOR EXPERIMENTAL POWER REACTOR

Reaktor Daya Eksperimental (RDE) is a high-temperature gas-cooled reactor (HTGR) for electricity generation, heat generation, and hydrogen production by Batan. Empirical and numerical calculations are needed to strengthen the existing design. The numerical method by computational fluid dynamic (CFD) analyzes temperature distribution and pressure drop along the pipe. The Batan RDE steam generator design has a seven-layer helical pipe model, while this research uses a one-layer helix pipe. In empirical calculations, the heat transfer region has three sections; single-phase liquid, two-phase, and single-phase vapor heat transfer. In numerical calculations, apply the assumption of constant heat flux and constant working fluid properties. The results of empiric calculations data showed that the helical pipe height was 3.98 m, shorter than the Batan design, which is 4.97 m. This considerable difference due to empirical calculations did not cover the safety factor. The results of numerical calculations show that in the single-phase, empiric calculation data were acceptable since the different values of numerical calculations for empiric calculations data were below 10%. Meanwhile, the case of the two-phase numerical calculations is not satisfactory and needs further research to obtain optimal results.

Numerical simulation of the transient graphite oxidation process based on a dynamic mesh method

The high-temperature gas-cooled reactor uses graphite-matrix spherical fuel elements to coat the tri-structural isotropic fuel particles (TRISO). In accident conditions, graphite oxidation may cause fuel particle exposure, which further leads to severe radioactive contamination. Therefore, it is very important to study the oxidation behavior of graphite during air-ingress accidents. This work aimed to develop a reaction kinetic model of graphite oxidation based on the Institute of Nuclear and New Energy Technology (INET) experiment correlations and to study the transient process of graphite oxidation using computational fluid dynamics and a dynamic mesh method. The results show that convective diffusion and mass diffusion dominate the distribution of reaction products at different flow rates. The reaction rate in the early stage of oxidation was high and then decreased rapidly until the oxygen consumption rate and supplementation rate reached equilibrium. After that, the oxidation rate decreased slowly due to the reduction of graphite surface area. The fluctuating flow around the graphite leads to anisotropic oxygen concentrations at different inflow angles on the graphite surface. The fluctuations caused a non-uniform ellipsoidal graphite structure, in which graphite has a smaller radius of curvature and pits on the windward side and a larger radius of curvature on the leeward side. The oxidation mass of graphite is in good agreement with the INET experimental data, which shows that this study can provide a method for the estimation of the mass and shape of graphite oxidation.

Comparison of unirradiated and irradiated AGR-2 TRISO fuel particle oxidation response

The silicon carbide (SiC) coating in a tristructural isotropic (TRISO) particle acts as a barrier to fission product release during reactor operation and accident scenarios. Oxidation and subsequent failure of the SiC layer during a rare air ingress event is a proposed mechanism for fission product release in a high-temperature gas-cooled reactor (HTGR). Although previous oxidation studies have analyzed unirradiated TRISO particle response, this study compared the oxidation behavior of irradiated and unirradiated TRISO particles from the second Advanced Gas Reactor Fuel Development and Qualification Program irradiation experiment (AGR-2). Particles with exposed SiC were subjected to six varying oxidizing tests in the Furnace for Irradiated TRISO Testing (FITT), examined for failure fraction with the Irradiated Microsphere Gamma Analyzer (IMGA) and characterized with focused ion beam and scanning/transmission electron microscopy techniques to analyze the oxide layer. Uncorrelated unirradiated particle failures throughout the series of exposures suggests that external factors inherent to the experiment increased particle failure sensitivity. However, irradiated particle observations indicated an increased failure response at 400 h 1400 °C in both 2% and 21% O2 atmospheres above failure associated with external factors. Oxide thickness measurements after 400 h at 1400 °C revealed a greater oxidation rate than predicted by parabolic growth, which was attributed to the increased complexity of the oxide structure at longer exposure times. Altering the atmosphere from 21% to 2% O2 reduced the average oxide thickness by approximately 12%–14% in both irradiated and unirradiated particles at 400 h 1400 °C. Overall, the minor variations observed between irradiated and unirradiated particles in this study led to the conclusion that unirradiated TRISO particles can be used to approximate irradiated TRISO oxidation kinetics.

Design analysis of a printed circuit heat exchanger for HTGR using a 3D finite elements model

The intermediate heat exchanger (IHX) is a critical component in most of the High Temperature Gas-cooled Reactor plant layouts. The printed circuit heat exchanger (PCHE) has been studied as an advantageous candidate for this application last years. The extreme operating conditions of the IHX and the higher structural integrity of the PCHE over other heat exchanger technologies have motivated these studies. In this work, the structural assessment of a zigzag channel PCHE under IHX conditions was carried out by means of computational fluid dynamics (CFD) and Finite Element Analysis (FEA). The ASME Boiling and Pressure Vessel Code, Sections III and VIII, were applied to evaluate the safety operation of the PCHE. The temperature field in steady state was obtained using a 3D CFD model. Then it was used as thermal load to calculate thermal stress using a FEA model. The influence of the zigzag angle and the bend radius on the thermal and mechanical stresses was analyzed. Thus, the values studied for the zigzag angle were 0° (straight), 15°, 30 and 45°, while for the bend radius were 0.5 mm, 2 mm, 4 mm and 6 mm. The importance of the 3D temperature field to calculate thermal stress in a zigzag channel was evidenced due to the appearance of local temperature differences in the channel cross-sections. The change in flow direction created areas of higher local temperatures where the flow lines change their directions impacting the channel wall. A significant thermal stress concentration was found in these areas, and this effect intensified with increasing zigzag angle. The bend radius was decisive to reduce the thermal stress concentration when the zigzag angle was higher than 15°.

Critical Evaluation of a Novel Analysis Technique for Assessment of Printed Circuit Heat Exchangers in High-Temperature Nuclear Service

AbstractApplication of printed circuit heat exchangers (PCHEs) to very high-temperature reactors (VHTRs) requires mechanical performance assessment methodologies. The PCHE morphology consists of thousands of millimeter-scale channels, for enhanced thermal efficiency, enclosed in a meter-scale PCHE core. PCHE geometry under thermomechanical creep-fatigue transients results in multi-axial interactions between its different segments, such as channeled core, walls, and headers. These global-level interactions influence the local channel-level responses. Hence, developing a PCHE performance assessment methodology, following the ASME Code, Section III, Division 5 provisions, is a critical gap to be filled. There is no analysis or design methodology available in ASME Code to assess a PCHE for its global and local level performances under high temperature and pressure loadings. This paper critically evaluates a recently proposed two-step analysis technique to estimate global interactions and local channel-level responses of PCHEs. In this novel analysis technique, the channeled PCHE core is replaced with orthotropic solid blocks of representative stiffness properties for the global thermomechanical analysis. Subsequent channel scale submodel analysis with detailed channel geometry, loading, and elastic-perfectly plastic (EPP) material model estimates the local responses for PCHE performance assessment. This paper critically evaluates this novel technique for its effectiveness in PCHE performance assessment. Finite element (FE) models imitating various analysis issues are developed, and FE analysis results are scrutinized. An important outcome of this study is the validation of the novel two-step PCHE analysis technique for application to the performance assessment of PCHEs in VHTRs.

Printed Circuit Heat Exchangers (PCHEs): A Brief Review

Heat exchangers and other heat transfer devices/systems play vital roles of heat transfer in thermal fluid flow systems for industrial application. Sodium cooled fast reactors are normally designed to have two loops of sodium coolants and one loop of water coolant which generates steam for power production. The two loops of sodium coolants consist of primary cooling system of sodium which cools the fuel rods of the reactor core and secondary cooling system of sodium transferring heat from the sodium primary cooling system. The water-cooling system transfers heat from the secondary cooling system of sodium for steam generation. Lead cooled fast reactors on the other hand are designed to have primary cooling system of lead cooling the fuel rods in the reactor core and secondary cooling system of water transferring heat from the lead cooling primary system for steam generation. Water cooled Nuclear Power Plants used water to cool the reactor core in the primary system and the heat removed from the core is used for steam generation directly as in BWRs and SCWRs or in the secondary system of heat exchanger as in PWRs. Other reactor systems such as Gas-cooled fast reactor (GFR), Molten-salt reactor (MSR), High-temperature gas-cooled reactor (HTGR), and Small Modular Reactors (SMRs) also have various types of heat exchangers in their designs to support power/electricity generation. Appropriate heat exchangers are therefore needed for various stages of heat transfer in power generation systems. Thus, Heat exchangers and other heat transfer devices/systems play vital roles of heat transfer in thermal fluid flow systems for industrial applications. This study presents brief review of PCHEs which have comparable advantages over other types of heat exchangers. Recent studies on PCHEs and other heat exchanger types have been reviewed. Design and optimization of PCHEs, optimization of Brayton and Rankine circles, and fluid flow and heat transfer devices/systems have been discussed briefly. The review findings show that the design and optimization of PCHEs depends on the intended industrial application of the heat exchanger. The various channel types and channel cross-section types available for design and optimisation as well as the design and optimised system being able to withstand high pressure and temperature conditions in addition to its compact size for the intended industrial application make PCHEs unique among other types of heat exchangers.

Fusion-fission hybrid reactor facility: power profiling

The current state of research in the field of nuclear and thermonuclear power aimed at creating power generation plants makes it possible to predict the further development of modern power industry in the direction hybrid reactor power plants. Such hybrid systems include a tokamak with reactor technologies, worked out in detail in Russia, and systems with an additional source of neutrons. Power generation plants using tokamaks and accelerators with the required level of proton energy will be of exceptionally large size and power, which will postpone their construction on an industrial scale to the distant future. The ongoing research is aimed at the development of small generation and has the prospect of entering the field of energy use in a shorter period. The hybrid reactor facility under study consists of an axisymmetric assembly of fuel blocks of a high-temperature gas-cooled reactor and a linear plasma source of additional neutrons. The paper demonstrates the results of optimization plasma-physical, thermophysical and gas-dynamic studies, the purpose of which is to level the distortions of the power density field, which are formed in the volume of the multiplicating part of the facility due to the pulsed operation of the plasma source of D-T-neutrons. The studies on increasing the “brightness” of the source and modeling its operating modes were carried out using the DOL and PRIZMA programs. The thermophysical optimization and gas-dynamic calculations were performed using the verified SERPENT and FloEFD software codes. The calculations were made on a high-performance cluster of the Tomsk Polytechnic University.

A GPU-based DEM model for the pebble flow study in packed bed: Simulation scheme and validation

This study developed a GPU-based DEM (GPU-DEM) model for the pebble flow in the packed bed. The details of the GPU-DEM scheme have been illustrated. The uniform mesh neighbor-search method, unified memory addressing (UMA), and special boundary conditions were incorporated in the GPU-DEM model. Cases of the direct impact of two identical particles, the angle of repose of particles drained from a lifted hopper, and the circulating hopper to mimic a pebble flow in High-Temperature Gas-cooled Reactor (HTGR) reactors were carried out to systematically verify the applicability, computational efficiency, and accuracy of the GPU-DEM model. Related results, like the unloading speed, porosity, and velocity distribution, obtained by the GPU-DEM model have been validated too. In addition, the effects of single-precision floating-point number computation on GPU were discussed. The results demonstrated that the GPU-DEM can achieve 18– 20 times acceleration while maintaining accuracy, even when single-precision floating-point numbers were used for calculation. Therefore, it is a powerful tool to explore the particle flows in chemical engineering applications.

Microstructural analysis of tristructural isotropic particles in high-temperature steam mixed gas atmospheres

High-temperature gas-cooled reactors (HTGRs) use tristructural isotropic (TRISO) particles embedded in a graphitic matrix material to form the integral fuel element. Potential off-normal reactor conditions for HTGRs include steam ingress with temperatures above 1,000 °C. Fuel element exposure to steam can cause the graphitic matrix material to evolve, forming an atmosphere composed of oxidants and oxidation products and potentially exposing the TRISO particles to these conditions. Investigating the oxidation response of TRISO particles exposed to a mixed gas atmosphere will provide insight into the stability under off-normal conditions. In this study, surrogate TRISO particles were exposed to high temperatures (T = 1,200 °C) in flowing steam (3% < pH2O < 21%) and CO (pCO < 1%) to determine the oxidation behavior of the SiC layer when exposed to various mixed gas atmospheres. Scanning electron microscopy, x-ray diffraction, and focused ion beam milling was used to determine the impact of CO and steam on the oxidation behavior of the SiC layer. The data presented demonstrates how the SiC layer showed strong oxidation resistance due to limited SiO2 growth and maintained its structural integrity under these off-normal conditions.

Improved Methods for Calculating the Neutron Diffusion Coefficient of General Pebble Bed

There exists a neutron streaming effect in the high-temperature gas-cooled pebble-bed reactor (HTGR) pebble-bed core caused by the spatial heterogeneity of the neutron’s free path, which has a remarkable impact on neutron leakage. It is necessary to take into consideration the streaming effect in evaluating the homogenized diffusion coefficient of the pebble bed, prior to the whole-core diffusion calculation. In this paper, two methods are proposed for calculating the homogenized multigroup diffusion coefficient of the pebble bed based on migration area conservation theory and Benoist’s theory, respectively. Compared with existing methods, the newly proposed methods are adaptable to a general pebble bed consisting of multitype pebbles and nonvacuum atmosphere. Numerical results demonstrate the proposed methods’ effectiveness and consistency in evaluation of the pebble-bed homogenized diffusion coefficient.

Design of a laser welding plugging system for the steam generators of high-temperature gas-cooled reactor

As fourth-generation nuclear power equipment, high-temperature gas-cooled reactor (HTGR) is attracting more and more attention both in industry and academia. The core outlet temperature of the steam generator (SG) in HTGR is much higher than that of pressurized water reactor steam generators, which is more likely to cause damage. Till now, welding plugging is still one of the important means to prevent the leakage of damaged heat transfer tubes. However, how to improve the automation of welding plugging to reduce the short-range operation time in a radioactive environment is still difficult. This work proposes the design of a laser welding plugging system for the HTGR. Firstly, the design requirements are analyzed and the overall design scheme is given. The whole system consists of a laser generator, a multi-axis robot, and supporting devices. Secondly, the key supporting devices including a laser welding head, a pneumatic clamping jaw, and a single-axis motion platform are explained in detail. The structure simulation based on SolidWorks is adopted, which provides the basis for the design of the single-axis motion platform. Thirdly, the control system which takes the PLC control system as the main control unit and the industrial computer as the monitoring data processing unit is described. Finally, an experimental system is built and the laser welding experiment is carried out to test the feasibility of the designed laser welding plugging system.

Inventories of Short-Lived Fission Gas Nuclides in Nuclear Reactors

Taking inventories in reactor cores is critical for understanding their radioactive source terms and establishing the relationship between the activity concentration in the primary loop and the status of the reactor core’s fuel. However, there is a niche in which a simple but accurate relationship between reactor conditions and nuclide inventories can reliably predict the fission gas nuclide activities of the reactor core in the primary loop. In this study, a simple and efficient model called “Inventories of a Point Reactor for Fission Gas Nuclides” (IPRFGN) was proposed to calculate and interpret such inventories, in which a 10 MW high-temperature gas-cooled experimental reactor (HTR-10) was used as the test case. The present study findings were consistent with those of a general point–depletion burnup code such as the KORIGEN code. Here, the relative error was &lt;1%. Based on the application of the IPRFGN model in HTR-10, the results indicate that the proposed IPRFGN model has provided the relationship between the inventories of fission gas nuclides in the core and the reactor conditions in all types of nuclear fission reactors. In the future, the IPRFGN model will be used for calculating fission gas nuclide inventories in various reactors.

Surface and in-depth structural changes in nuclear graphite irradiated with noble gases described with Raman imaging

4th Generation high-temperature gas-cooled nuclear reactors (HTGR) are regarded as possible sources of industrial heat in Poland and Europe, allowing for a substantial reduction of the dependency on gas and coal import. It is mainly due to their safety of use, reliability and economy in a current energetic crisis. In this work, graphite, as a primary construction material and neutron moderator in HTGR, was evaluated before and after ion irradiation since its properties depend on the material’s structure and purity. Commercial graphite materials (IG-110, NBG-17) and the laboratory’s in-home material were chosen for the exemplary samples. The structural damage in HTGR was simulated with energetic Ar+ and He+ ions with fluencies from 1E12 to 2E17 ion/cm2. Raman imaging was chosen to assess radiation damage build-up: the crystallites’ evolution, occurrence and types of defects. The recorded evolution showed stronger disordering of the material with heavier Ar+ ions than with He+.

Research on non-LWRs operational intervention level based on HTR

Nuclear emergency is the final line of defense for humanity and the environment, and operational intervention level (OIL) becomes a key technical element pre-specified in emergency preparedness and guides emergency response actions. The International Atomic Energy Agency (IAEA) issued a new version of OIL system based on light water reactors (LWRs) in 2017. To explore its applicability for non-LWRs, this paper presents a set of specified OIL determination procedures for high temperature gas-cooled reactors (HTRs). Considering a representative OIL based on ground monitoring (OIL1) as the case study, it concludes that the default OIL1 is 500 μSv/h for HTR-PM, half of the IAEA recommended value. In addition, through sensitivity analysis, we find that the slow release of radionuclides does not affect the determination of the default OIL1 value. From the perspective of OIL, this research further confirms that HTR reactor types can technically simplify off-site emergency requirements.

Development of a hybrid model for fuel performance analysis of spherical fuel elements under normal conditions

The spherical fuel element (SFE) dispersed with many tri-isotropic fuel (TRISO) particles has been primarily employed in the high-temperature gas-cooled reactor. Owing to the complex structure of SFE, behaviors will be evolved for both TRISO particles and the fuel matrix with the increasing burnup. To evaluate the fuel performance of SFE, the homogenization method has been conventionally adopted for SFE scale modeling, and dispersed TRISO particles are analyzed subsequently with fixed temperature boundary conditions. However, owing to the varying properties of TRISO particles under reactor operation, the homogenization models dedicated to certain circumstances can hardly reflect the properties evolution of TRISO with high fidelity, which may lead to uncertainties in analysis results. To this end, a hybrid model, namely the conventional homogenization model informed by the lower scale model online, has been developed in this paper, which may increase modeling fidelity for both SFE and particles. Specifically, property models for the TRISO particle and graphite matrix have been implemented in COMSOL. Subsequently, based on the linking of COMSOL and MATLAB, the hybrid model for SFE analysis was established. Through the hybrid model, the temperature field and stress distribution of TRISO, the evolution of effective properties, and the thermal and mechanical state of the SFE were obtained and analyzed. The results show that within the scope of this paper, the central temperature of TRISO in the innermost region rises with the increasing burnup for the decreasing thermal conductivity of materials, and the peak value is about 1324 K, which is about 60 K higher than that in the outermost region. The difference in the maximum plenum pressure is more than 5 MPa among TRISO particles, which is mainly caused by the variations in CO production. Consequently, the hoop stress of SiC in the outermost region is slightly elevated. The distribution of effective thermal conductivity (ETC) for TRISO is nearly 4.3 W/(m·K) at the beginning of life (BOL), while it was reduced to 3.6 W/(m·K) at the end of life (EOL). There are marginal variations along the radial direction for the ETC of TRISO. Meanwhile, the ETC of SFE decreases from about 21 W/(m·K) for fresh fuel to 13 W/(m·K) for spent fuel. The predicted effective elastic modulus and coefficient of thermal expansion of SFE increase gradually versus time. The former is caused by the densification of pyrolytic carbon, and the latter is mainly determined by the high volume fraction of matrix and the changes of CTE for pyrolytic carbon. In addition, the comparison in the temperature field between the proposed model and the Differential Effective Material Theory (D-EMT) model may further demonstrate the feasibility and effectiveness of the hybrid model.

Numerical study on heat transfer characteris tics of supercritical water in straight and helical tubes

The subcritical steam generators in high-temperature gas-cooled reactor pebble-bed modules can be replaced by supercritical steam generators to improve thermal efficiency. The axial and circumferential nonuniform flow and heat transfer characteristics of supercritical water in straight and helical tubes with varying inclinations (0–90°) are simulated to investigate the combined effects of tube structure, flow direction, thermophysical properties, property variations, buoyancy, and centrifugal forces. With heat fluxes of 233–930 kW/m2 and inlet mass fluxes of 225 or 1260 kg/(m2 s) under a working pressure of 24.5 MPa, the friction coefficients, turbulent kinetic energy, secondary flow, temperature, and heat transfer coefficients are studied in the cross section of variable structure tubes. The wall temperature and heat transfer coefficients of four regions are compared in straight and helical tubes; the maximum heat transfer coefficient appeared before the pseudo-critical point. The heat transfer coefficients are the greatest at the bottom of the straight tubes, where fluidity was high, and at the outer of the helical tubes, where the fluid velocity was high under centrifugal force. The secondary flow in the helical tubes causes a nonuniformity in the wall temperature, causing the circumferential maximum wall temperature to move.

Heat transfer of supercritical pressure water in helical tubes

Subcritical steam generators in the current high-temperature gas-cooled reactor pebble-bed module in China can be replaced by supercritical steam generators to better coordinate the reactors and supercritical steam turbine unit to improve the thermal efficiency with no fluid phase change at supercritical pressures. The heat transfer of supercritical-pressure water flow in helical tubes with various helix diameters was simulated using the re-normalization group k–ε model to investigate the effects of finite-thickness walls, inlet mass flux, and helix diameter. The finite-thickness wall barely affected local heat transfer in the straight tube with inlet mass flux of 1,260 kg/(m2 s) whereas the local heat transfer in the helical tube was very different with relative difference in q/qav between the inside and bottom, caused by the centrifugal force, is 36%. In helical tubes, the heat transfer coefficients in the inner bar were the lowest, with a low fluid velocity under the effect of the centrifugal force. The buoyancy and centrifugal force in the helical tubes caused non-uniformity in the wall temperature; however, when the fluid temperature increased below the pseudo-critical temperature, the temperature distributions showed an increasing trend. The non-uniformities in the wall temperature distributions decreased as the centrifugal force increased, and they were suppressed by the finite-thickness wall.

Simultaneous Solution of Helical Coiled Once-Through Steam Generator with High-Speed Water Property Library

Efficient simulation of the helical coiled once-through steam generator (H-OTSG) is crucial in the design and safety analysis of the high-temperature gas-cooled reactor (HTGR). The physical property and phase transformation of water in the steam generator brings great challenges during simulation. The water properties calculation routine occupies a large part of the computational time in the steam generator solution process. Thus, a thermohydraulic property library is developed based on the IAPWS-IF97 formulation in this work to reduce the computational cost. Here the formulation adopts the backward equation method to avoid iterations in thermodynamic property calculation. Moreover, two Newton-method-based simultaneous solutions are implemented as implicitly nonlinear solvers, including Jacobian-Free Newton–Krylov (JFNK) and Newton–Krylov (NK) methods, due to its excellent computational performance. These simultaneous solution algorithms are combined with the developed water property library to simulate the H-OTSG efficiently. The numerical analysis is performed based on the transient and steady-state cases of the HTR-10 steam generator. Successful simulations of HTR-10 steam generator cases demonstrate the capability of the newly developed method.

Indonesia's experimental power reactor program (RDE)

The experimental power reactor program (RDE) was created to introduce Indonesia to nuclear power plants (NPP). This was done by building a prototype of a non-commercial 10 MWth power reactor. The RDE program aims to show the public that Indonesian engineers can design reactors that can supply enough electricity and steam to support industrial needs. The technology used is a high-temperature gas-cooled reactor. The RDE program is a strategic and intermediate target for the country’s energy security and national sovereignty. In the long term, the development of this RDE-based nuclear power plant is expected to have implications for the use of clean energy and the reduction of the country’s dependence on fossil fuels. In addition, the implementation and development of the RDE program will increase national industrial capacity and global competitiveness. Moreover, the RDE program can be used as a reference for commercial nuclear power plants called PeLUIt (Power and Steam Generators for Industry) to meet the demand for industrial electricity and heat in various regions in Indonesia. However, many challenges arise in the implementation of this program, including the estimated significant development costs and the possibility of a long development period. To date, the RDE program has had a positive impact on strengthening the capacity of human resources in Indonesia in the field of power reactor design by being able to map the capabilities of the national industry to support nuclear power plant development programs. This is in addition to the formation of stakeholders from the Government, academics, and entrepreneurs.

Deep Learning for Multigroup Cross-Section Representation in Two-Step Core Calculations

We investigate using deep learning, a type of machine-learning algorithm employing multiple layers of artificial neurons, for the mathematical representation of multigroup cross sections for use in the Griffin reactor multiphysics code for two-step deterministic neutronics calculations. A three-dimensional fuel element typical of a high-temperature gas reactor as well as a two-dimensional sodium-cooled fast reactor lattice are modeled using the Serpent Monte Carlo code, and multigroup macroscopic cross sections are generated for various state parameters to produce a training data set and a separate validation data set. A fully connected, feedforward neural network is trained using the open-source PyTorch machine-learning framework, and its accuracy is compared against the standard piecewise linear interpolation model. Additionally, we provide in this work a generic technique for propagating the cross-section model errors up to the keff using sensitivity coefficients with the first-order uncertainty propagation rule. Quantifying the eigenvalue error due to the cross-section regression errors is especially practical for appropriately selecting the mathematical representation of the cross sections. We demonstrate that the artificial neural network model produces lower errors and therefore enables better accuracy relative to the piecewise linear model when the cross sections exhibit nonlinear dependencies; especially when a coarse grid is employed, where the errors can be halved by the artificial neural network. However, for linearly dependent multigroup cross sections as found for the sodium-cooled fast reactor case, a simpler linear regression outperforms deeper networks.