Fidelity, model

John Maher, Mark E. Davis, Robert J. Hancock, and Sidney W. Theis, High Fidelity Modeling of Space-Based Radar, 2003 IEEE Radar Conference, Huntsville, AL, pp. 185-191, May 2003. 

Develop high-fidelity models for process design, revamp, or optimization... 

Apply high-fidelity models for equipment design Apply consistent design by leveraging the same validated thermophysical properties knowledge and models used in preliminary design stages... 

Develop a high-fidelity model of the plant for operator training by using the same validated, rigorous thermophysical property models... 

In this paper we present a Model Reduction technique incorporated with multi-parametric programming and control, namely Balanced Truncation (T3T). The use of Balanced Truncation eliminates a number of states of dynamic linear systems, while a bound on the maximum error obtained for the output vector can be established. This then allows for the derivation of (approximate) linear parametric controllers, which can be tested and validated (against the original high-fidelity model) off-line. These theoretical developments are presented next. 

High-fidelity models are a useful tool for predicting flow over and around the superstructure of naval vessels under a variety of meteorological conditions as such, they provide useful guidance for the optimal placement of onboard sensors. Less work has been done on predicting flow at ports and bases. 

Invest in high-fidelity modeling of releases in and near ports and bases to guide sensor placement and to form the basis for simplified operational tools. 

While it is relatively easy to outhne the functionahty and unique aspects of the software discussed above, it is a great deal harder to describe the level of difficulty required to achieve a desired end result and to characterize how weU functions are performed. Online help systems, multimedia components, associated software development tools, improved parameter measurement tools, and higher-fidelity models will no doubt impact these aspects. 

J.A. Kypuros and R.G. Longoria Variable fidelity modeling of vehicle ride dynamics using an element activity metric. ASME 2002 International Mechanictil Engineering Congress and Exposition 71, 525-534 (2002)... 

For the study of complex systems or of transient plant behavior, more simplified models can also be used, provided that proper experimental validation is obtained. For macro-modeling of the system, an ad hoc steady-state code was developed on the basis of LIBrary for Process Flowsheeting (LIBPF) technology [31], capable either of integrating the detailed codes for stack simulation of the MCFC-D3S and SIMFC, or of using an intermediate-fidelity model for the stack to improve the calculation performance when required. 

CGC II High-fidelity modeling for the prediction of corrosion degradation... 

Understanding the nature of protective fiims and scaies, inciuding structure Complete and comprehensive understanding of eiectrochemistry and other interfaces from the electronic to the microscale level Prediction Prediction CGC II High-fidelity modeling for the prediction of corrosion degradation in actual service environments... 

CGCII High-Fidelity Modeling for the Prediction of Corrosion Degradation in Actual Service Environments... 

In the past, for benign clutter conditions, radar detection performance has been limited by the presence of noise generated in the receiver. Low RCS targets flying near the Earth s surface create a problem in detection due to the interference of the clutter signal. So-called clutter limited performance analysis depends on a high-fidelity model of the clutter reflection characteristics. 

High-fidelity models for industrial catalytic reactors can be incorporated into these flowsheet simulators via a number of techniques depending on the characteristics of the flowsheet simulator. In some steady-state flowsheet simulators, there is a built-in facility for the user to add his own module. In such cases, the high-fidelity reactor model can be added as one of these special modules. The reactor model in this case can make full use of the physical properties database available in the simulator. However, in most flowsheet simulators, this facility is not available in such cases, the module can be used with the flowsheet simulator on the basis that the output variables from the last unit preceding the catalytic reactor is used as an input file to the catalytic reactor model, and when the calculations for the reactor are complete, the output file from the reactor model is used as an input file to the unit next to the reactor in the process flowsheet. Clearly, this last technique is less efficient and takes more effort to implement than in the first case. 

As implied in the figure, the outputs of the more detailed fundamental models can be used in lower-order models. This flow of information is, in fact, a critical application for high fidelity models. Recently, much work has been done in the development of algorithms to integrate or embed high-fidelity models into system analysis simulation tools. 

A high fidelity model of a fuel cell may include many dependent variables such as concentration, temperature, velocity, potential, pressure, and other fields described in the model. For this reasOTi, a segregation of the solution procedure is required by solving a smaller set of equations fuUy coupled and then iterating over all possible smaller sets in one nonlinear iteration. To obtain faster convergence, variables such as the potential in the electrodes and in the electrolyte should then be solved in the same set, due to their tight coupling and presence in the... 

Once the numerical model equations are solved, the convergence of the solution can be studied. This can be achieved by solving the problem for different element sizes and also for different element orders. One possible method for developing high fidelity models of the reaction kinetics and mass transport effects in the gas diffusion electrodes, and in the electrolyte, is to start with one-dimensional models. These model equations are relatively quick to solve and it is also possible to run the simulation for many different element sizes, which may give a hint of the accuracy and convergence that may be... 

To show how sensitive the ID model is to the screen flow area term Ac in the FTS pressure drop equation. Figure 9.32 plots the model predictions against the data at a reduced FTS area that minimizes error between data and model (60% flow area for TVS cooled 325 and 40% flow area for 325 standard channel). With the reduced area, about half of the data lies above and below the predictions. A higher fidelity model would be required to accurately explain the FTS pressure distribution along the LAD. Nonetheless, the ID model qualitatively tracks the trends in the data. 

The high fidelity model developed in the previous section can be used to conduct simulation case studies, what-if analysis, decision support, dynamic optimization, and soft-sensing with model predictive control. The off-line dynamic optimization approach is introduced here to illustrate a strategy that can be followed to develop optimum operation trajectories to achieve a specific PSD shape. 

---