An evaluation method is developed for single blow experiments with liquids on heat exchangers. The method is based on the unity Mach number dispersion model. The evaluation of one experiment yields merely one equation for the two unknowns, the number of transfer units and the dispersive Peclet number. Calculations on an example confirm that one single blow test alone cannot provide reliable values of the unknowns. A second test with a liquid of differing heat capacity is required, or a tracer experiment for the measurement of the Peclet number. A modified method is developed for gases. One experiment yields the effective number of transfer units and approximate values of the two unknowns. The numerical evaluation of calculated experiments demonstrates the applicability of the evaluation methods.

The recently developed special unity Mach number dispersion model prescribes the corrections to heat transfer coefficients which are simple functions of the dispersive Peclet numbers. They can be determined through the residence time measurements. An evaluation method is described in which the measured input and response concentration profiles are numerically Laplace transformed and evaluated in the frequency domain. A characteristic mean Peclet number is defined. The method is also applied to the parabolic dispersion model and the cascade model. A calculated example of a tube bundle with maldistribution and backflow demonstrates the suitability of the evaluation method.

An evaluation method is developed for temperature oscillation experiments in heat exchangers. The unity Mach number dispersion model is applied. For the consideration of lateral wall heat conduction an effective wall thickness is introduced together with a wall heat transfer coefficient. The evaluation method may also be applied to single blow experiments with pulse signals. A sensitivity analysis describes and discusses the accuracy of different evaluation procedures.

The paper presents a mature concept of an intelligent monitoring system of air pollution inflow and its realization in the form of a SINZaP system lunched at Institute for Ecology of Industrial Areas (]ETU) in 2006. SINZaP is a real time operating system resembling a neural network. It is designed for modeling of pollutant emissions and air pollutants concentrations, addressed to specialists or decision makers responsible for air quality management. For modeling of emission and air pollutants concentrations in SIZNaP system, a back trajectory model -BackTrack has been used, which is based on YLSTRACK model. The essential feature ofthe BackTrack model is the application of back trajectories in the selection of emission sources influencing a given receptor. For modeling of trajectories BackTrack uses three-dimensional wind fields, friction velocity, MoninObukhov length and mixing layer height. SINZaP consists of four main modules: (I) data module including data scanner for reading public data accessible in the Internet, (2) module for preparation of meteorological data, (3) BackTrack module for simulations of pollutants emissions and simulations of air pollutants concentrations, and (4) Trainer module, the task ofwhich is correction of input parameters for adjusting modeling and observed data.