The main goal of the study was to analyze vibration of pumping aggregate. There are fourth position which could be very harmful for the total working operation of the pumping aggregate. The pumping aggregated should have smooth continuous operations without any mistake. Vibration could affect different parts or segments of the pumping aggregate and therefore it is need to analyze the vibrating. Analyzing of the vibration could be highly nonlinear task since many different parameters are involved in the model. To avoid the analytical model in this article soft computing approach was used since the soft computing approach does not require internal knowledge of the vibration model. For the soft computing model there is enough to collect the input output data pairs through experimental measurement procedure. Based on the input/output data pairs the model will be created. The approach should rank the influence of the measuring positions vibration on the pumping aggregate. Finally three different soft computing methods were compared and results were reported.

Introduction

The development of non-invasive methods of monitoring of the machine conditions are important in order to predict maintenance of the machines. There are various indicators for machine condition; however, the method of use vibration monitoring for determination of machine operating conditions was proved the most important. Vibration monitoring is one among many methods of technical diagnostics which has been continuously monitoring a technical state of a device by observing the level of a mechanical oscillation in a real time [1]. The mechanical oscillation is the manifestation of a device during its operation [2]. Certain parts become vibration exciters, others, depending on excitation, react specifically [3]. Therefore the vibration monitoring is one of the most important methods used in technical diagnostics for identifying a technical state [4]. With the use of vibration diagnostics we are able to detect an incipient failure, locate the place of an incipient failure and predict the length of time during which a device is going to work before a failure occurs or a preventive action is performed [5-8]. For example in article [9] the vibration behavior of microscale beams and plates was studied based on micropolar theory (MPT). Presented in paperr [10, 11] was a size-dependent analysis of the surface stress and nonlocal influences on the free vibration characteristics of rectangular and circular nanoplates. Maintenance and repair of pumping units requires a lot of high material costs and time. Therefore, control of vibration and parametric characteristics of pumping units in real time allows to identify defect of pumping units at an early stage [12]. The condition monitoring of a rotating machine is efficient, but often it is difficult and labour intensive task for maintenance crew to troubleshoot the machine and vibration analysis is a method used for condition monitoring of the machine [13]. Measured signals are usually fed into filters or signal decomposers to extract useful features to assist making identification in state monitoring or fault diagnosis. But what is routinely ignored is that an experienced expert can realize what is happening just by watching the signals presented on the oscilloscope even without the analyzing report [14]. Vibration-based condition monitoring is an important approach to ensure the reliability of industrial machines [15].