Summary
A structure with semi‐active control devices is usually a highly nonlinear system. To investigate the aseismic performance of such a system, real‐time hybrid testing (RTHT) can be a cost‐effective experimental method. However, a substructure with a semi‐active friction device is difficult to be tested by the RTHT, because the dynamic behavior of a friction device, which consists of sliding and sticking phases, is determined by the exerted force of the primary structure, rather than its displacement response. To overcome this problem, a methodology of RTHT with a shake table (RTHT‐ST) is utilized in this study. In the RTHT‐ST, which is an experimental technique combining shaking table test and RTHT, the shake table is employed to mimic the acceleration response of the primary structure that is simulated by a numerical model and imposed to the substructure, which is mounted on the shake table. In order to verify the feasibility of the experimental method, a semi‐active piezoelectric friction controllable mass damper substructure is tested by using the RTHT‐ST. The test results of the RTHT‐ST are compared with those of a full shaking table test, in which the integrated primary structure and piezoelectric friction controllable mass damper system have been physically tested. Moreover, to evaluate the accuracy of the RTHT‐ST result, one category of indicators called root‐mean‐square energy error index is also proposed. Unlike previously existing hybrid‐testing indices, the root‐mean‐square energy error index is able to distinguish modeling error from control system error.