Three-step approach for prediction of limit cycle pressure oscillations in combustion chambers of gas turbines

Currently, gas turbine manufacturers frequently face the problem of strong acoustic combustion driven oscillations inside combustion chambers. These combustion instabilities can cause extensive wear and sometimes even catastrophic damages to combustion hardware. This requires prevention of combustion instabilities, which, in turn, requires reliable and fast predictive tools. This work presents a three-step method to find stability margins within which gas turbines can be operated without going into self-excited pressure oscillations. As a first step, a set of unsteady Reynolds-averaged Navier–Stokes simulations with the Flame Speed Closure (FSC) model implemented in the OpenFOAM® environment are performed to obtain the flame describing function of the combustor set-up. The standard FSC model is extended in this work to take into account the combined effect of strain and heat losses on the flame. As a second step, a linear three-time-lag-distributed model for a perfectly premixed swirl-stabilized flame is extended to the nonlinear regime. The factors causing changes in the model parameters when applying high-amplitude velocity perturbations are analysed. As a third step, time-domain simulations employing a low-order network model implemented in Simulink® are performed. In this work, the proposed method is applied to a laboratory test rig. The proposed method permits not only the unsteady frequencies of acoustic oscillations to be computed, but the amplitudes of such oscillations as well. Knowing the amplitudes of unstable pressure oscillations, it is possible to determine how these oscillations are harmful to the combustor equipment. The proposed method has a low cost because it does not require any license for computational fluid dynamics software.     

Rapid chemiluminescence biosensing of microcystin-LR

A rapid immunoassay for sensitive detection of microcystin-LR using a portable chemiluminescence multichannel immunosensor (CL-MADAG) was developed. The sensor device is based on a capillary ELISA technique in combination with a miniaturized fluidics system and uses chemiluminescence as the detection principle. Minimum concentrations of at least 0.2 μg L⁻¹ microcystin-LR could be unambiguously measured in a spiked buffer system as well as in spiked real water samples. A single sample analysis for detection of microcystin-LR could be accomplished in just 13 min on the CL-MADAG. Besides providing a highly reproducible, fast and easy to perform test format, one major advantage of the newly established capillary immunoassay is represented by the feasibility of an internal retrospective quality control mechanism. Finally, simultaneous CL-MADAG measurements employing our inhibition immunoassay and a sandwich ELISA could be successfully demonstrated.     

铜负载湿式氧化催化剂焙烧条件的研究

Cu-O/FSC是一种优化制备的负载型催化剂,实验中以其催化湿式氧化处理模拟印染废水,考察焙烧条件对催化剂的影响.以水样COD去除率和脱色率评价催化剂的活性、以处理出水Cu溶出浓度评价催化剂的稳定性,通过AA、XRD、SEM等研究了焙烧温度和时间对催化剂性能、物相及形貌等的影响.结果表明:随着焙烧温度的提高和焙烧时间的延长,催化剂活性降低、稳定性提高;催化剂的比表面和孔容随着焙烧温度的提高而减少;适宜的催化剂焙烧温度和时间分别为650 ℃和5 h.     

A fully hydrodynamic model for three‐dimensional,free‐surface flows

The hydrostatic pressure assumption has been widely used in studying water movements in rivers, lakes, estuaries, and oceans. While this assumption is valid in many cases and has been successfully used in numerous studies, there are many cases where this assumption is questionable. This paper presents a three‐dimensional, hydrodynamic model for free‐surface flows without using the hydrostatic pressure assumption. The model includes two predictor–corrector steps. In the first predictor–corrector step, the model uses hydrostatic pressure at the previous time step as an initial estimate of the total pressure field at the new time step. Based on the estimated pressure field, an intermediate velocity field is calculated, which is then corrected by adding the non‐hydrostatic component of the pressure to the estimated pressure field. A Poisson equation for non‐hydrostatic pressure is solved before the second intermediate velocity field is calculated. The final velocity field is found after the free surface at the new time step is computed by solving a free‐surface correction equation. The numerical method was validated with several analytical solutions and laboratory experiments. Model results agree reasonably well with analytical solutions and laboratory results. Model simulations suggest that the numerical method presented is suitable for fully hydrodynamic simulations of three‐dimensional, free‐surface flows. Copyright © 2003 John Wiley & Sons, Ltd.