Iranian Journal of Chemical Engineering (IJChE)

Abstract ewline"> Dispersion of heavy gases is considered to be more hazardous than the passive ones because it takes place more slowly. When the gas is accidentally released at ground level or where there are many obstacles in the area it is considered to be a heavy gas. In this paper, based on the extensive experimental work of McQuid and Hanna, the model was tested against two types of experiments: A simple experiment “Thorney Island” and a complex experiment “Kit Fox” in order to validate CFD code. In order to accomplish this validation the multiphase approach was employed. Also, the vertical temperature gradient in the atmosphere was investigated. The investigation of wind speed was done taking factors such as time, height and direction into consideration. In order to reduce the number of elements in the computational domain, a combination of 2D and 3D geometry was utilized. The results showed that the wind inlet correction, as well as the temperature gradient, had a significant influence on gas concentration records.

Abstract One of the important aspects of distillation control design is the choice of a good control structure since improper choice of manipulated/controlled variable pairing can lead to poor control performance. In this paper, comparison and selection of structures is mainly based on the plant condition number. Columns with reflux ratios less than unity or greater than five show large condition numbers. For the ratio structures, the results show that double-ratio structures such as the (D/(L+D))(V/B)-structure have smaller values of the relative gain array element (λ11 ) compared with single-ratio structures, such as the (D/(L+D))V-structure. In addition, the relative values of λ11 corresponding to the values of minimized condition number (γmin), instead of condition number (γ), provides a better basis for comparison. It has also been shown that the maximum singular value (σmax) of the relative gain array (RGA) is a good criteria between ratio and non-ratio structures, and also between various ratio structures selection. At a constant reflux ratio, columns with smaller values of σmax show small values of γmin . Finally, a frequency-based analysis is performed for the selection of the appropriate structure. The analyses show that although the DV-structure has a relatively small value of condition number with respect to other structures, the value of λ11 is far from unity. In contrast, ratio structures have λ11 values near unity. Frequency-based behavior of ratio structures show small oscillations at higher frequencies (> 1 rad/min), while the conventional LV-structures show large oscillations for smaller values of λ11 at higher frequencies.

Abstract Distillation control is a challenging undertaking given the inherent nonlinearity of the process, severe coupling present for dual-composition control and the sensitivity of disturbances. Among various distillation operations, management of the high-purity column poses a difficult control situation due to a number of characteristics of these systems, including strong directionality, ill-conditioning and strongly nonlinear behaviour. In this paper, a diagonal PID controller is designed and analysed for a high-purity distillation column by computing the structured singular value μ introduced by Doyle (1982). For this purpose, a structured uncertainty model has been developed which describes the dynamics of the column for the entire operating range. The achievable manageable performance is also defined in terms of the H-norm of the weighted sensitivity function.