Precision Design Modelling of HF2V Devices. (June 2018)

• Record Type: Journal Article
• Authors: Vishnupriya, ; Rodgers, G.W.; Mander, J.B.; Chase, J.G.
• Is Part Of: Structures. Volume 14(2018)
• Page Start: 243
• Page End: 250
• Publication Date: 2018-06
• Subjects: HF2V lead-extrusion damper -- Regression analysis -- Precision modelling -- Leave k validation -- Energy-dissipation device; Structural engineering -- Periodicals; 624.1
• DOI: 10.1016/j.istruc.2018.03.007
• Languages: English
• ISSNs: 2352-0124

Abstract

Abstract: High-force-to-volume (HF2V) lead-extrusion energy dissipation devices have been demonstrated to be efficient, low-cost energy dissipators for managing structural seismic response energy to mitigate damage. They have been employed in full-scale structures, as well as demonstrated in a range of large-scale experiments and numerical structural analysis studies. However, they have proven difficult to precisely design to match a specific force capacity. This study develops a precision HF2V design model based on the sum of friction and extrusion forces modelled as a function of device dimensions. Specifically, the Area Ratio (AR), shaft Surface Area (SA), and Bulge Area (AB). Multiplicative coefficients for these terms in 14 linear and linear-quadratic models are calculated using regression analysis on data from 18 experimental devices with and without bulges. Pearson correlation coefficient values (R 2 ) summarise model completeness and the error between experimental and model predicted force. Leave k = 3 out validation for random and specific groups of devices assesses model robustness to the device data used to identify the model. Model 2 (F = α0 AR. Dcyl + α1 SA) and Model 4 (F = α0 AR. Dcyl + α1 SA + α2 AB) have the best fit explaining all but 5% of the experimentally measured device force (R 2 = 0.95). Overall, Model 4 predicts device forces better for a wide range of devices. Mean model prediction errors are 11%, and attributed to differences in device pre-stress, test velocity and experimental error. On average, the terms for AR and AB, capturing device extrusion force contributions to device force, contribute 65% and 13% of the total device force, with the remaining 22% due to friction, captured by the SA term. The overall results provide a simple generalisable model capturing all relevant mechanics for precise design of HF2V devices to a specific quasi-static force capacity, as well as a good starting point for more specific and detailed mechanics models.