Thermal Convection Loop Research Articles

Optimal and adaptive control of chaotic convection—Theory and experiments

In theory and experiments, optimal and adaptive control strategies are employed to suppress chaotic convection in a thermal convection loop. The thermal convection loop is a relatively simple experimental paradigm that exhibits complex dynamic behavior and provides a convenient platform for evaluating and comparing various control strategies. The objective of this study is to evaluate the feasibility of employing optimal control and nonlinear estimator to alter naturally occurring flow patterns and to compare the performance of the optimal controller with that of other controllers such as neural network controllers. It is demonstrated that when the system’s model is not known, experimental data alone can be utilized for the construction of a proportional controller.

Self-adjustment of Li in Pb–17Li systems

With temperature gradient capsules and thermal convection loops it is shown, that self-adjustment of 15.8 at.% Li in lead–lithium systems is possible. All which is needed is a diffusion type cold trap with a solid phase. Lithium can be deposited there, it will be released if the concentration is reduced below the eutectic in case of lithium consumption.

Controlling chaotic convection using neural nets—theory and experiments

An exploratory study is conducted to assess the feasibility of using neural networks to control flow patterns and to evaluate the performance of these controllers. Neural networks were used to control (suppress) chaotic convection both in experiments and in a theoretical model of a thermal convection loop. It is demonstrated that the neural network controller can successfully cause the flow to behave in a desired way. The performance of the neural network controllers was compared with that of previously used conventional linear proportional controllers.

Rendering a subcritical Hopf bifurcation supercritical

It is demonstrated experimentally and theoretically that through the use of a nonlinear feedback controller, one can render a subcritical Hopf bifurcation supercritical and thus dramatically modify the nature of the flow in a thermal convection loop heated from below and cooled from above. In particular, we show that the controller can replace the naturally occurring chaotic motion with a stable, periodic limit cycle. The control strategy consists of sensing the deviation of fluid temperatures from desired values at a number of locations inside the loop and then altering the wall heating to counteract such deviations.

Bifurcation control of a chaotic system

The viability of controlling chaos by controlling associated bifurcations is demonstrated in the context of a thermal convection loop model. Washout filter-aided feedback controls are employed to delay and to extinguish chaos in the model, and to target preferred equilibrium points. An important feature of the control laws is that they do not result in any change in the set of equilibria, even in the presence of model uncertainty.

Identification Based Adaptive Control of Chaotic Continuous-Time Systems

The present paper further developes ideas of controlling chaos by the speed gradient method. It is shown how to design parameter identifier for chaotic systems ensuring achievement of the objective of parameter identification. The data obtained by the identifier can be used for adaptive control law design. Two examples are considered: identification of parameters of periodically perturbed Duffing equation and stabilizing the inherent unstable equilibrium point of the thermal convection loop model when parameters of the model are unknown.

Fabricating niobium test loops for the SP-100 space reactor

This article describes the successful fabrication, operation, and evaluation of a series of niobium-alloy (Nb-1Zr and PWC-11) thermal convection loops designed to contain and circulate molten lithium at 1,350 K. These loops were used to establish the fabrication variables of significance for a nuclear power supply for space. Approximately 200 weldments were evaluated for their tendency to be attacked by lithium as a function of varying atmospheric contamination. No attack occurred for any weldment free of contamination, with or without heat treatment, and no welds accidentally deviated from purity. The threshold oxygen content for weldment attack was determined to be 170–200 ppm. Attack varied directly with weldment oxygen and nitrogen contents.

115 Controlling chaos and targeting in a thermal convection loop

115 Controlling chaos and targeting in a thermal convection loop

Controlling Chaos and Targeting in a Thermal Convection Loop

Abstract The stabilizing control of a deterministic chaotic system is investigated. The system under study is a thermal convection loop which exhibits bifurcations, transient chaos, and fully developed chaos. The paper shows how focusing on the control of primary bifurcations in the model can result in the taming of chaos. This is an example of the ‘bifurcation control’ approach. First it is shown that a linear-plus-nonlinear controller, which result in delaying and stabilizing the primary bifurcations (Hopf bifurcations), is effective to suppress both chaos and transient chaos for any parameter range. Then the design of 'targeting' controllers which result the system preferring one equilibrium to another is given. These controllers all employ washout filters, which results in robustness with respect to model uncertainties, and, in particular, uncertainty with respect to the system equilibrium points.

Selective surface preoxidation to inhibit the corrosion of AISI type 316L stainless steel by liquid Pb17Li

This paper describes the formation of a ternary oxide coating on 316L stainless steel from the reaction of Pb17Li with the preoxidized surface of steel specimens. The preoxidized surfaces were prepared by heating 316L stainless steel specimens to 800°C in a controlled H 2/H 2O atmosphere (ratio 1000:1). The oxide layer before reaction with the Pb17Li was characterized as Mn 1.5Cr 1.5P 4. Analysis after reaction with Pb17Li indicated a LiMn 2O 4 structure with some of the manganese sites occupied by chromium. 316L specimens prepared with different oxide layer thicknesses, along with uncoated specimens, were corroded in the isothermal hot legs of two Pb17Li filled thermal convection loops. Post-test analysis of the specimens indicated that the oxide coated specimens had, on average, a thinner ferritic corrosion layer than the uncoated specimens. The coated specimens also showed areas with no ferritic corrosion layer.

Controlling chaos in a thermal convection loop

It is demonstrated experimentally and theoretically that through the use of an active (feedback) controller one can dramatically modify the nature of the flow in a toroidal thermal convection loop heated from below and cooled from above. In particular, we show how a simple control strategy can be used to suppress (laminarize) the naturally occurring chaotic motion or induce chaos in otherwise time-independent flow. The control strategy consists of sensing the deviation of fluid temperatures from desired values at a number of locations inside the loop and then altering the wall heating to either counteract or enhance such deviations.

Thermal convection loop with heating from above

The dynamics of single phase, buoyancy-induced flow in a toroidal loop are investigated theoretically under conditions of time-dependent, periodically varying, wall temperature. The heating and cooling, which are applied to the loop walls, are symmetrical with respect to the loop's axis which is parallel to the gravity vector. On average, the upper half's temperature is maintained at a higher value than that of the lower half. That is, on average, the temperature field inside the loop is stably stratified and the Rayleigh number based on the average temperature difference assumes a negative value. Despite this fact, as the Rayleigh number decreases, the flow in the loop exhibits a rich range of flow structures with a net amount of heat transport from the hot to the cold boundary. As the Rayleigh number decreases, the flow structures change from no-motion to time periodic motion to chaotic flow occasionally interrupted by periodic windows.

Behavior of deuterium and rare gases in thermal convection loops with molten Pb17Li

The behavior of deuterium in thermal convection loops with molten Pb17Li was investigated in the temperature range from 300 to 610°C, and in the range of deuterium partial pressures from 0.05 to 1000 mbar. Dissolution and desorption are controlled by diffusion through a 0.02 mm thick LM boundary layer at the interface, no chemical reactions are involved in the rate determining step. This boundary layer is also effective in case of permeation through membranes, if one side is covered by the LM. The permeation through 0.6 mm iron was reduced by a factor of 100. However in case of a fusion reactor blanket this boundary layer will not be important, because the wall thickness of the components is much larger. For the 2 mm stainless steel of the thermal convection loops with a downstream oxide layer, no effect of the boundary layer could be seen. The amount of oxides in the loop had no influence on the results. Furthermore an excess of H2 at low partial pressures did not change transfer rates of deuterium.The solubility of deuterium in the LM was determined from the kinetics of loading and degassing. The found values are one order of magnitude smaller than the lowest values so far published.The transport behavior of the rare gases He, Ne, Ar, Kr and Xe was investigated. The solubility of helium was found five orders of magnitude lower than that of deuterium, those for Ne, Ar, Kr and Xe even lower than that for helium. Helium-bubble formation has to be considered if the flow rate of the LM in a blanket is small, or in case of static irradiation experiments. On the other hand argon can be used as covergas for a fusion reactor blanket. Because of the low solubility in the LM, the Ar-41 activity will be much smaller than in sodium cooled reactors.

Deposition behavior of ferrous alloys in molten lead-lithium

The mass transfer of type 316 stainless steel in Pb-17Li at% Li was studied using a thermal convection loop operating at a maximum temperature of 500°C to generate mass change and surface composition data as a function of time and loop position. Data analysis indicated that particles suspended in the flowing liquid metal (particularly those containing nickel) probably played a significant role in overall transport and deposition. There was also some evidence of physical detachment of deposits. The deposition of chromium (but not nickel) correlated with the temperature dependence of solubility, as did previous weight change results from a study of ferritic (FeCr) steels in nonisothermal Pb-17 at% Li. Due to the influence of particulate matter in the liquid metal and deposit detachment, mass transfer prediction for austenitic (FeCrNi) steels in Pb-17 at% should be more complicated than that for FeCr steels.

Controlling a chaotic system.

Using both experimental and theoretical results, this Letter describes how low-energy, feedback control signals can be successfully utilized to suppress (laminarize) chaotic flow in a thermal convection loop.

Compatibility problems in tritium breeding blankets

The compatibility between tritium breeding materials (liquid or solid), neutron multiplier and structural steels is a concern for the choice of a tritium breeding blanket for NET. For solid tritium breeding blanket, it seems that the more severe compatibility problem is due to the interaction of beryllium with steel. As for the water-cooled Pb17Li blanket, the first results obtained in experimental conditions closed to the concept have evidenced lower corrosion rates than those measured in thermal convection loops.

Transport of deuterium and rare gases by flowing molten Pb-17Li

In thermal convection loops the transport of deuterium and rare gases by molten Pb-17Li was investigated in the temperature range of 300 to 700 °C. Dissolution and desorption of the gases are controlled by diffusion through a LM boundary layer. No chemical effects could be detected. The solubility of deuterium was found one order of magnitude lower than values from the literature. Transport rates and solubilities of helium are 10 3 to 10 5 times smaller compared to deuterium, for all other rare gases smaller than those of helium.

The Role of Carbides in the Corrosion Behavior of Fe-12Cr-1MoVW Steel in Liquid Lithium

Experimental investigations with lithium and Fe-12Cr-1MoVW alloy in two thermal convection loops (360 to 505/sup 0/C and 530 to 655/sup 0/C) have shown that weight changes in these systems are not simple increasing functions of temperature. It was found that redistribution of carbon from the ..cap alpha..-ferrite phase of the isothermally annealed material by the formation of large (10 - 20 ..mu..m in diameter) surface carbides with the crystallographic structure of M/sub 23/C/sub 6/ (M = Fe or Cr) dominate the weight changes between 450 and 580/sup 0/C. The release of carbon from the surface ..cap alpha..-ferrite is probably due to decreased surface chromium activity by the formation of Li/sub 9/CrN/sub 5/ in the lower temperature experiment. In the higher temperature experiment, the difference in carbon activity between the steel and lithium is probably due to the rapidly increasing solubility of carbon in the lithium. The stoichiometry of the surface carbides varied from Cr/sub 18.8/Fe/sub 4.2/C/sub 6/ to Cr/sub 3.2/Fe/sub 19.8/C/sub 6/ with temperature and loop position, while surface substrate chromium content ranged from 4 to 8% by weight, decreasing with increasing temperature. Controlling initial chromium and carbon activities in the steel by decreasing carbon contents and/or by substituting moremore » stable carbide-forming elements such as titanium or vanadium for chromium could help to mitigate the problem.« less

Experimental and analytical investigations of mass transport processes of 12cr-1 movw steel in thermally-convected lithium systems

Experimental data on corrosion and mass transport in lithium/12Cr-1 MoVW steel were obtained from two thermal convection loops; one operated from 360 to 505 °C for 3040 h and the other from 525 to 655 °C for 2510 h. The experimental effort was supported by analytical investigations of possible mechanisms of corrosion and mass transport. It was found that mass transfer is not a simple function of temperature and alloy component solubility. Above 580 °C mass transfer appears to be dominated by alloy solubility via the temperature gradient. Between 450 and 580 °C, mass transfer appears to be related to surface reactions involving dissolved carbon and nitrogen in lithium with chromium, and carbides on the steel surface. The corrosion rates, as interperted from weight-change as uniform dissolution, from this work, are significantly lower than those adopted in recent blanket design studies.

Corrosion of austenitic and martensitic stainless steels in flowing 17Li-83Pb alloy

With regard to the behaviour of 316 L stainless steel at 400°C in flowing anisothermal 17Li-83Pb the mass transfer suffered by this steel appears to be quite important without noticeable influence of constant or cyclic stress. Evaluation made from solution-annealed specimens leads to a corrosion rate of approximately 30 μm yr−1 at steady state to which a depth of 25 μm has to be added to take into account the initial period phenomena. On the other hand, with semi-stagnant 17Li-83Pb at 400° C, the mass transfer of 316 L steel appears to be lower and more acceptable after a 3000-h exposure; but long-time kinetics data have to be achieved in order to see if that better behaviour is persistent and does not correspond to a longer incubation period. As for the martensitic steels their corrosion rate at 450°C in the thermal convection loop TULIP is constant up to 3000 h and five times lower than that observed for 316 L steel in the same conditions.