Experimental validation of the sensing model

Three IPMC samples, with dimensions as specified in Table 4.3, were used in the experiments to validate the geometric scalability of the model. All samples were cut from the same IPMC sheet. The relevant physical parameters were first identified using the Slim-2 sample, and then used to construct the models for the Big-2 and Short-2 samples. Fig. 4.14 shows the Bode plots of the frequency responses for the models of the Slim-2 and Big-2 samples, in comparison with their measured frequency responses. Good agreement was seen between the model predictions and the experimental measurements for both samples. In addition, since the two samples differed only in the width, the model would predict that they have the same phase responses and a 6 dB difference in the magnitude response, which was indeed supported by the experimental results shown in Fig. 4.14. 

Experiments with two other types of mechanical stimuli were further conducted to validate the modeling and model reduction approaches. In the first experiment, the Slim-2 sample was allowed to vibrate in air freely upon an initial perturbation on the tip (Fig. 4.16(a)), and Fig. 4.16(b) shows the predicted and measured sensor outputs in response to free damp ed oscillations. In the second experiment, a step change was applied to the tip displacement of the Big-2 sample, as recorded in Fig. 4.17(a). The corresponding predicted and measured sensor outputs are shown in Fig. 4.17(b). In Figs. 4.16 and 4.17, reduced fourth-order models were used for the prediction of sensor output, and the same parameters for the full models were adopted for the reduced ones. The match between the model predictions and the actual sensor measurements in both figures indicates that the modeling approach is effective. 

---