FIS-based Domestic Milling Machine PHM System Considering Multi-speed Frequency Variation

Abstract

The purpose of this study is to use a real-time measurement system to collect the operation signals of the milling machine under multi-speed frequency variation, and to conduct prognostics and health management research. First, we installed a three-axis accelerometer sensor on a domestic milling machine with different health conditions. Then in the experimental project, we completed the data acquisition of the 4-step speed, and used the waterfall map to achieve the purpose of data visualization. In signal analysis, we perform data mining for the speed, frequency and amplitude of the frequency domain. Finally, the fuzzy inference system is used to construct the decision mechanism, and the representativeness of the input parameters is weighed by the weight of the rules. From the experimental results, the FIS-based PHM method can effectively identify the spindle motor state and predict the trend of health deterioration from the frequency characteristics.