In the master's thesis we consider the problem of detecting inadequately operating compressors in refrigerating and freezing appliances. Compressors of various types and manufacturers are analysed and compared to each other on the basis of statistical features in the time and frequency domain. We extract features based on statistical analysis of vibration signals of the operating compressor, which is measured with an accelerometer. To detect inadequate compressors we introduce the decision thresholds, which are set based on the acoustic evaluation of one type of compressor. Then we analyse the decision thresholds and provide recommendations for their application on the entire compressor population.
|
