Digital filtering of acceleration data acquired during the intervention of a lift safety gears

In this paper two digital filters are applied to elaborate acceleration data acquired from two accelerometers fixed in a lift frame during safety gears tests. The experimental tests consist in free falls of a test lift car and the subsequent gripping phases initiated by the safety gears activation due to over-speed condition. Unfortunately, the reliability of acceleration data is heavily impaired by measurement noise, which needs to be suppressed to the largest extent possible in order to use the data for comfort evaluation. To this end, two classes of digital filters are tested: the Savitzky–Golay and the Butterworth filters. This choice is motivated by the key features of these filters, including their computational simplicity and high suitability to represent time varying features in acceleration measurements. A tuning procedure is proposed for these filters, such that the measurement noise corrupting the experimental data is maximally dumped