Engineering topics
Pronskikh, V. S.
Publications and source records attributed to Pronskikh, V. S..
Study of secondary neutron interactions with 232 Th, 129 I, and 127 I nuclei with the uranium assembly “QUINTA” at 2, 4, and 8GeV deuteron beams of the JINR Nuclotron accelerator
The natural uranium assembly, “QUINTA”, was irradiated with 2, 4, and 8 GeV deuterons. The 232 Th, 127 I, and 129 I samples have been exposed to secondary neutrons produced in the assembly at a 20-cm radial distance from the deuteron beam axis. The spectra of gamma rays emitted by the activated 232 Th, 127 I, and 129 I samples have been analyzed and several tens of product nuclei have been identified. For each of those products, neutron-induced reaction rates have been determined. The transmutation power for the 129 I samples is estimated. Furthermore, experimental results were compared to those calculated with well-known stochastic and deterministic codes.
Validation of Monte Carlo simulation of neutron production in a spallation experiment
A renewed interest in experimental research on Accelerator-Driven Systems (ADS) has been initiated by the global attempt to produce energy from thorium as a safe(r), clean(er) and (more) proliferation-resistant alternative to the uranium-fuelled thermal nuclear reactors. The ADS research has been actively pursued at the Joint Institute for Nuclear Research (JINR), Dubna, since decades. Most recently, the emission of fast neutrons was experimentally investigated at the massive (m = 512 kg) natural uranium spallation target QUINTA. The target has been irradiated with the relativistic deuteron beams of energy from 0.5 AGeV up to 4 AGeV at the JINR Nuclotron accelerator in numerous experiments since 2011. Neutron production inside the target was studied through the gamma-ray spectrometry measurement of natural uranium activation detectors. Experimental reaction rates for (n,γ), (n,f) and (n,2n) reactions in uranium have provided valuable information about the neutron distribution over a wide range of energies up to some GeV. The experimental data were compared to the predictions of Monte Carlo simulations using the MCNPX 2.7.0 code. In conclusion, the results are presented and potential sources of partial disagreement are discussed later in this work.
Computer Modeling and Simulation
Verification and validation of computer codes and models used in simulation are two aspects of the scientific practice of high importance and have recently been discussed by philosophers of science. While verification is predominantly associated with the correctness of the way a model is represented by a computer code or algorithm, validation more often refers to model’s relation to the real world and its intended use. It has been argued that because complex simulations are generally not transparent to a practitioner, the Duhem problem can arise for verification and validation due to their entanglement; such an entanglement makes it impossible to distinguish whether a coding error or model’s general inadequacy to its target should be blamed in the case of the model failure. I argue that in order to disentangle verification and validation, a clear distinction between computer modeling (construction of mathematical computer models of elementary processes) and simulation (construction of models of composite objects and processes by means of numerical experimenting with them) needs to be made. Holding on to that distinction, I propose to relate verification (based on theoretical strategies such as inferences) to modeling and validation, which shares the common epistemology with experimentation, to simulation. To explain reasons of their intermittent entanglement I propose a weberian ideal-typical model of modeling and simulation as roles in practice. I suggest an approach to alleviate the Duhem problem for verification and validation generally applicable in practice and based on differences in epistemic strategies and scopes