Parametric Identification of the Models with Specified Quality Characteristics

Introduction. The model of a given structure should be identified based on the results of solving the problem of parametric identification. This model should provide the best possible the database development reproduction of the experimental data. The concept of “best” is not strictly structured. Therefore, the procedure for identifying such a model is subject to natural logic and includes the stages of data a determination of a set of acceptable models and subsequent selection of the best of them. If the set of acceptable models is large, the procedure for determining the best one can be time-consuming. In this regard, the development of methods for parametric identification, which at the stage of creating a set of acceptable models allows taking into account the qualitative aspects of the identified dependence, which are of interest to the researcher, is of particular importance. Materials and Methods. The set of acceptable methods in the problems of parametric identification largely depends on the type of the experimental data. Uncertainty for example, probabilistic and statistical methods are useful if the observed factors are random and subject to any law of probability distribution. If the conditions for the use of such methods are not met, it may be useful to present an approach based on identifying the boundaries of location of the model parameters that ensure the achievement of specified levels of quality characteristics. Results. The procedure of parametric identification of models is formalized. It is based on the use of maximum permissible parameter estimates and allows one to determining the set of parameter values that guarantee the achievement of the required qualitative level of experimental data description, including from the standpoint of analyzing the impact of changes in accord with requirements to the accuracy of their reproduction. The approbation of the developed method on the example of the construction of a one-factor model of chemical kinetics is presented. Discussion and Conclusion. It is shown that the obtained value of the chemical reaction rate constant, in accordance with the introduced criteria, provides acceptable accuracy, adequacy, and stability of the identified kinetic model. At the same time, the results of calculations revealed the information that can form the basis for planning experiments carried out in order to improve the accuracy of the experimental data.

Modeling of the Formation Process of the Coherent Intermetallides in Nickel Alloys during Laser Treatment

A solution for new phase inclusion growth is suggested for the case of the Ni-based alloys. This solution takes into account a chemical composition of the alloy and conditions of the laser treatment process. A method for calculation of the chemical reaction rate constant is also suggested for the Ni3Al formation in the Ni-Al alloy. Modeling is fulfilled for the laser cladding process. Time dependences for the reacting component concentration is obtained as well as for intermetallic inclusion growth and thermal cycle.

Kinetic Model of Spherical Sponge Iron Reduced by Hydrogen

Spherical sponge iron (SSI) with high activity and intension could be prepared through direct reduction by hydrogen. To optimize the reduction technology, kinetic model of SSI reduction was established. The total reaction rate changing with reduction index R was deduced which describing the total reaction rate with effective diffusion coefficient De and chemical reaction rate constant k. According to the weight loss curve of SSI reduction, De and k were calculated. The total reactive rate increased with the increasing of temperature because both De and k increased with the increasing of reaction temperature. Compared De with k, it concluded that SSI reduced by hydrogen was controlled by the chemical reaction, combination of the chemical reaction and the internal diffusion, the internal diffusion when the temperature was lower than T3, from T3 to T4, over T5, respectively.