The paper presents the practical application of a wireless vibration sensor system for the ongoing monitoring of the technical condition of industrial fans used in gas furnaces to heat metal billets prior to plastic working. During almost a year of research, vibration and temperature data was collected and subjected to time domain analysis. The measurement results were divided into weekly time intervals and statistically processed. In order to facilitate the decoding, segmentation into time-based segments, and statistical analysis of the research data, a software solution was developed. The application enabled the analysis of changes in weekly histograms and statistical parameters of data sets. A critical analysis was made of the statistical parameters obtained from the research material. The average weekly value of rms vibration acceleration and the highest weekly value of the relative frequency of data were selected as the parameters that best reflects changes in industrial fan technical condition. Functions modelling these changes were also developed. During the research, significant changes in the recorded vibration accelerations on fan motor No. 2 were observed. The developed functions proved useful for assessing and predicting changes in the technical condition of the tested fans.
The research data was obtained thanks to The Polish National Centre for Research and Development project no. POIR.01.01.01-00-0304/19-00, titled 'Hardware and software system for the diagnostics of machines and devices based on a wireless network of monitoring sensors and methods of knowledge engineering and computational intelligence'.