information at strategically important locations will ensure that important information is not lost during actual experimentation. Figure 13: Different sensor locations used to determine the response of the structure using expansion algorithm at all 434 FEA node locations, (a) 14 sensors, (b) 24 sensors. Figures 14 and 15 show comparison between predicted and reference Y normal strain solution for 14 sensor (a) and 24 sensor (b) set-ups at the tip and inside the box-beam, respectively. Observe that the 24 sensor set-up show slight improvement in the strain prediction as compared to the 14 sensor set-up. However, none of the sensor set-ups provide accurate strain prediction because only the first six modes are used in the expansion process. Recall that for accurate strain prediction, more than six modes were shown to be required in case 1. This indicates that use of more number of sensors does not necessarily satisfy the requirement of use of greater number of modes (more than six modes ) in the expansion process. However, in general, use of redundant sensors at strategically important locations could be beneficial from a practical stand-point. Figure 14: Strain response comparison at the tip of the box-beam by reference and predicted solutions using first six modes for expansion, (a) with 14 sensors, (b) with 24 sensors. 198
RkJQdWJsaXNoZXIy MTMzNzEzMQ==