Abstract

Nowadays compressors are used in almost every industry. Compressor failure can suspend production of the whole system so the importance of preventing from failures is obvious and essential. One of the most dangerous flow instabilities which is capable of destroying machine in few seconds is surge, which occurs in conditions of low mass flow rate. Greitzer model, apart from its long history, is still most common mathematical model describing surge. It is widely used to predict the surge onset and pressure oscillations during it. However, it is based on parameters that are not directly related to real machine and their choice is not always obvious. Therefore, the calculations may be inaccurate which results in wrong surge prediction. The other approach to Greitzer model is presented in this paper, which in some cases can assure that the process of compressor modeling is more accurate. The applicability of Greitzer surge model for real machines has been analyzed. Method of implementation is based on experimental pressure signal gathered during unstable work of compressor. Presented method is based on experimental compressor characteristic and outlet pressure signal from unstable work of compressor. From that data it is possible to determinate the value of Greitzer model’s parameters for selected operational point. Thanks to this method this model could be applied for reliable antisurge protection.