Author:
Mousmoulis G,Yiakopoulos C,Kassanos I,Antoniadis I,Anagnostopoulos J
Abstract
Abstract
Centrifugal pumps are widely used in industry and cover a significant percentage of the total energy consumption of a power plant. This fact makes the application of efficient maintenance tools to be of crucial importance and obliges researchers and engineers to develop reliable detection methods for special types of malfunction, such as cavitation. Cavitation is a hydrodynamic mechanism that affects the steady and dynamic operation of a pump and is usually responsible for its head and efficiency reduction, as also for premature wear and destruction of the impeller. The stochastic nature of the phenomenon and the complex flow characteristics, as well as the noise from the mechanical components of the pump complicates the detection of cavitation. For this reason the use of several sensors in combination with advanced signal processing techniques is often proposed for the successful identification of two phase flow in the signal obtained. In this study, a radial centrifugal pump is tested under various flowrates in order to derive its cavitation characteristic curves. At the same time, the signal obtained from various vibration and acoustic emission sensors is recorded. The sensors are located at the rolling bearings that support the overhung impeller, and close to the suction of the pump, where the static pressure takes its minimum value. Subsequently, by processing the digitized signals of all noise and vibration measurements both in time and frequency domains, it is possible to identify the appearance of cavitation. Finally, the results show that the position of the sensor and the acquisition characteristics undoubtedly affect the detection of cavitation.
Cited by
4 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献