Hybrid Modeling Scheme Research Articles

Approximate near-wall treatments based on zonal and hybrid RANS–LES methods for LES at high Reynolds numbers

Two strategies, combining a LES scheme with different near-wall RANS approximations, are investigated by reference to simulations for plane channel flow and two separated flows at moderate and high Reynolds numbers, respectively. One strategy is a hybrid modelling scheme, wherein the subgrid-scale model in the outer LES domain is interfaced with a RANS model in a predefined near-wall layer. The other is a zonal method in which a thin-shear-flow RANS solution in the near-wall layer, embedded within the LES domain which covers the entire flow, is used to provide boundary conditions for the LES computation, the two thus being loosely coupled. Both methods allow the thickness of the near-wall RANS layer to be chosen freely. In the hybrid LES–RANS scheme, the near-wall layer is interfaced to the outer LES region, subject to compatibility conditions for velocity and turbulent viscosity imposed across the interface. These conditions are extracted dynamically as the simulation progresses. In the zonal approach, a mixing-layer model provides the eddy-viscosity field in the near-wall layer, while in the hybrid approach, a two-equations ( k– ϵ) model is used.

SYNERGETIC EFFECT COMBINING UML AND PETRI NETS IN DESIGNING MODERN INFORMATION SYSTEMS

This paper reviews the most archetypal features of UML and Petri nets. Formed modeling structural scheme reflects the trends of designing modern information systems. This paper also overviews the possibility of converting UML net into Petri net. At the end, the paper gives a conclusion, and also a prospective of further research – using methods presented here. Development of organizational information systems (IS) recently becomes an important subject. Lately, the Unified Modeling Language (UML) became the standard in creating software, and in designing relatively simple organizational information systems. Some UML diagrams are also used for organizational sectional modeling, reengineering of business processes, and process analysis. By current view UML is used for specification of task sequences, and in the process of creating software for generation of program code class diagrams are being used. Class diagrams are sourced using activity diagrams, which are based on empirical reasoning and are not described by mathematical methods. UML methodology only allows very limited modeling of simple systems, to say nothing of the possibilities to evaluate process dynamics. That’s why Petri nets are often tried to use for designing of systems. High development of IS is determined by formal mathematical methods. Petri nets are formalism for performing process characterized by parallel links analysis. Petri nets characterize parallel structures, flexible modeling, and highly developed mathematical formalism. However, the results show that it is still failed to implement universal modeling of dynamic processes. Using methodology of Petri nets and its extensions (ex. Object Petri nets) is not absolutely rational because additional efforts are needed for evidence of objective features, besides by modeling systems using objective methodologies, particularly UML, cumulative experience is being lost. It limits system development possibilities, especially in a case of advanced systems, where deeper analysis of activity alternatives is necessary, demanding to evaluate dependences on variable business conditions under the influence of external factors, also continuous modernized data mining systems influencing system’s flexibility and quality requirements, and that is the basis of dynamic IS creation. Therefore this paper proposes comprehensive information system creation model, which combines internals of both, net and object designing methodologies. This model also applies to simplified cases when information system (IS) development stages may be executed by using only UML features, while in a general system creation case hybrid modeling scheme is recommended. For that purpose methods of transformation from UML to Petri net and modeling possibilities’ expansion ways, which allow better implementation of work sequence management, are being improved in the process. The conclusion state that implemented IS using suggested modeling scheme characterizes flexibility, and enables appending new features to the system without blocking its exploitation. Concerning work, particularly mark management methods enable aiming that implementing IS would have modeling structure, possibilities to change system’s functions and features by supplementing system with separate program modules without changing whole system.

ONLINE MODEL-BASED DIAGNOSIS TO SUPPORT AUTONOMOUS OPERATION OF AN ADVANCED LIFE SUPPORT SYSTEM

This article describes methods for online model-based diagnosis of subsystems of the advanced life support system (ALS). The diagnosis methodology is tailored to detect, isolate, and identify faults in components of the system quickly so that fault-adaptive control techniques can be applied to maintain system operation without interruption. We describe the components of our hybrid modeling scheme and the diagnosis methodology, and then demonstrate the effectiveness of this methodology by building a detailed model of the reverse osmosis (RO) system of the water recovery system (WRS) of the ALS. This model is validated with real data collected from an experimental testbed at NASA JSC. A number of diagnosis experiments run on simulated faulty data are presented and the results are discussed.

Two-dimensional prediction of spatial variation in topsoil compaction of a sandy loam field-based on measured horizontal force of compaction sensor, cutting depth and moisture content

The development of a topsoil compaction map, based on real-time measurement of the draught ( D) of a compaction sensor provides a quick view of positions of extremely compacted zones. The measured D of a subsoiler, used as a compaction sensor was utilised to draw the two-dimensional spatial variation in soil compaction of a sandy loam field (Arenic Cambisol). On the basis of a previously developed formula, dry bulk density ( ρ d) indicating soil compaction was estimated as a function of the measured horizontal force, cutting depth ( d) and moisture content ( w). This formula was the output of a numerical–statistical hybrid modelling scheme, which aimed to estimate the variation in sensor D as a function of w, d and ρ d. The ARCVIEW 3.1 GIS software was used to draw the field maps of measurement and model-based ρ d, d, w. The measurement and model-based ρ d ranged from 1343 to 1750 and from 1271 to 1523 kg m −3, respectively. The model-based ρ d was underestimated by a mean error of 14%. A comparison of measurement and model-based ρ d maps indicates a similar tendency of spatial variation in soil compaction, particularly positions of extremely compacted zones. This allows providing the farmer with a compaction map, which illustrates the extreme zones of soil compaction. A correction factor of 14% in ρ d is incorporated into the developed numerical–statistical model, which improved the magnitude of the model-based predicted soil compaction. Furthermore, the real-time measurement of w with better control of d might be helpful to improve the magnitude of ρ d predicted and spatial distribution of soil compaction.

A numerical–statistical hybrid modelling scheme for evaluation of draught requirements of a subsoiler cutting a sandy loam soil, as affected by moisture content, bulk density and depth

The development of simple regression relationships, based on finite element (FEM) analyses of cutting a sandy loam soil by medium–deep subsoiler will provide confidential and quick tools to predict the subsoiler draught requirement for any combination of moisture content ( w), wet bulk density ( ρ w) and tillage depth ( d). Large number of FEM analyses were performed to assess the effect of w, d and ρ w on subsoiler draught. This draught was calculated from the output of 126 FEM modelling analyses. The model variations were selected based on the interaction among the studied variables, namely, six moisture contents ranging from 0.03 to 0.22 m 3 m −3, five wet bulk densities ranging from 1.3 to 2.0 Mg/m 3 and six tillage depths ranging from 0.10 to 0.37 m. The interaction among these three variables was also taken into consideration, and a multiple linear regression analysis was performed aiming at establishing a mathematical relationship for simulating the draught variation as a function of these variables. Further relationship was developed for relating the variation in draught with w, d and dry bulk density ( ρ d). The FEM showed that subsoiler draught increased with d, ρ w and ρ d, whereas it decreased with w. The decrease in draught with w did not extend beyond a w of 0.17 m 3 m −3, since the draught calculated for 0.22 m 3 m −3 w was very close to that calculated for a w of 0.17 m 3 m −3. Regression equations developed to relate the subsoiler draught with w, d, ρ w and ρ d were of quite simple forms, and had high determination coefficients ( R 2) close to 0.95. These equations indicated that the horizontal force varied linearly with w and non-linearly with d, ρ w and ρ d. This non-linear variation of draught was found to be a quadratic function of d and ρ w, and a cubic function of ρ d. A comparison of the calculated and measured draught for cutting a similar sandy loam soil with a similarly designed subsoiler, showed a good approximation. The small divergence between the calculated and measured draught was attributed to the difference in soil texture and subsoiler chisel length. Thus, the regression equation developed proved to be a capable tool of predicting the draught requirements of the selected tillage tool design, when cutting sandy loam soils under any combination of w, d and ρ w.

Mesoscale Model for Morphologic Change at Tidal Inlets

A mesoscale modeling system (time step of hours, space step of 100 m) for bathymetric change at tidal inlets is described. The Mesoscale Inlet Morphology modeling system is composed of a linked two-dimensional (depth-integrated) hydrodynamic module, a wave transformation module, a shoreline change module, and a sediment transport module. A hybrid modeling scheme is employed, featuring a microscale (time steps of minutes) hydrodynamic module within a mesoscale model. Sediment transport is assumed to be as bed load outside of the surf zone, driven by the mean wave- and tide-induced flow. The Mesoscale Inlet Morphology modeling system was applied to Price Inlet, S.C. A two-year simulation of the morphologic changes at Price Inlet was performed using measured wave data and tidal stages predicted from harmonic analysis coefficients, and results compared to measured bathymetric and shoreline changes. Predicted bathymetric changes are similar in sign and location to observations, but the model tends to underpredict the magnitude of the changes. Predicted shoreline changes exhibit similar trends.

Integration of mathematical modelling and knowledge-based systems for simulations of biochemical processes

In this work, a hybrid modelling scheme that integrates available mathematical description with artificial intelligence techniques was developed to simulate complex biochemical processes. The hybrid model features the parallel arrangement of a numeric module and a knowledge-based module. The numeric module was composed of 44 process variables and 145 model parameters, which involves algorithmic calculation of the differential equations for the metabolic behaviour of the biochemical processes. The knowledge-based module was designed to carry out frame-based inference to determine the cause of the modelling discrepancy using the relationship between the derivation of the process variables and the model parameters. Two types of frames (e.g., 8 frames for the variable error information and 14 frames for the parameter adjustment) were built to support the decision-making process. This novel hybrid modelling method was evaluated by comparing the predictions of a hybrid model with conventional mathematical modelling. The result has shown that by using the proposed scheme, the discrepancy between the model predictions and the experimental data can be reduced, and that the predictive strength of the simulation can be enhanced.

L-reps: 2次元ディスプレイ上で3次元ピンポイントを可能にする2次元指向幾何変換モデル<SUP>*</SUP>

In interactive geometric modeling, it is a key operation for a user to specify precisely an arbitrary point of interest in a computer-generated 3D finite work space. We point out that traditional ways of projection transformation have several difficulties in performing the above operation, which is called 3D pinpoint in this paper. In order to alleviate the 3D pinpoint difficulties, we propose L-reps (Lines representations) as a new geometric transformation model for standard 2D displays. L-reps is a 2D drafting-oriented geometric model, and a generalization of conventional lines: a 2D engineering drawing which has been commonly used for the design of ship hulls or automobile bodies. We provide here the data definition and manipulations of L-reps. Also, we present a methodology for utilizing L-reps in a solid modeling environment. Two-way conversion algorithms between B-reps and L-reps are developed as principal components of the hybrid modeling scheme, where precise and aesthetic design tasks can be performed simultaneously. A pilot geometric modeling system is implemented on the basis of the hybrid scheme. The feasibility of our approach is demonstrated with a camera design example on the system.