Dataset of measurements for the experimental CEA-beam benchmark structure subjected to one stochastic broadband excitation.

This data article comprises data to investigate the non-linear dynamic behavior of the CEA-beam benchmark structure subjected to one stochastic broadband excitation. Experiments have been performed on the CEA-CESTA laboratory. The data provided include the input Power Spectral Density for four levels of excitation and the associated output nonlinear dynamic behavior of the CEA-beam benchmark structure. All the results from this data will help researchers and engineers in proper analysis of hardening effect and the enlargement of the response peak due to one stochastic broadband excitation, as well as the presence of harmonics. One of the main original contributions is to share the data sets to give the opportunity to researchers for testing and validating analytical or numerical models of a nonlinear beam with non-ideal boundary conditions and subjected to one stochastic broadband excitation. This Data in Brief article is an additional item directly alongside the following paper published in the Communications in Nonlinear Science and Numerical Simulation (CNSNS) journal: T. Roncen, J-P. Lambelin and J-J. Sinou, Nonlinear vibrations of a beam with non-ideal boundary conditions and stochastic excitations - experiments, modeling and simulations, Communications in Nonlinear Science and Numerical Simulation,74 (2019) 14-29. doi.org/10.1016/j.cnsns.2019.03.006


a b s t r a c t
This data article comprises data to investigate the nonlinear dynamic behavior of the CEA-beam benchmark structure subjected to one stochastic broadband excitation. Experiments have been performed on the CEA-CESTA laboratory. The data provided include the input Power Spectral Density for four levels of excitation and the associated output nonlinear dynamic behavior of the CEA-beam benchmark structure. All the results from this data will help researchers and engineers in proper analysis of hardening effect and the enlargement of the response peak due to one stochastic broadband excitation, as well as the presence of harmonics. One of the main original contributions is to share the data sets to give the opportunity to researchers for testing and validating analytical or numerical models of a nonlinear beam with non-ideal boundary conditions and subjected to one stochastic broadband excitation. This Data in Brief article is an additional item directly alongside the following paper published in the Communications in Nonlinear Science and Numerical Simulation (CNSNS) journal: T. Roncen

Value of the Data
• The database provides the nonlinear response measurements at the center of the beam for the CEA-beam benchmark structure subjected to one stochastic excitation for four levels of excitation: 0.4 m.s −2 , 2 m.s −2 , 4 m.s −2 and 8 m.s −2 . • The data could be useful for researchers and industrial in understanding of the nonlinear bending behavior of a beam subjected to one stochastic excitation: the enlargement of the response peak in the vicinity of the primary resonance, as well as the appearance of secondary peaks resulting from the harmonics generated by the primary resonance.
Each provided data corresponds to the vertical displacement at the center of the beam (see the position A1 in Fig. 1 ). Acquisition of the experimental outputs are performed by using a miniature triaxial piezoelectric accelerometer with integral hybrid electronics (Ref. Endevco model 66M5).

Experimental Design, Materials and Methods
The CEA-beam benchmark structure is a clamped-clamped steel beam with non-ideal boundary conditions. It consists of a beam and two blocks made from a single piece of steel as shown in Fig. 1 . The two blocks are bolted onto a large circular aluminum plate, itself bolted onto the vibrating pot. The technical drawing of the CEA-beam benchmark structure is given in [2] . All the physical parameters are also given in [3] .
The following paragraph briefly describes the experimental protocol. As described in [3] , the shaker is piloted with a Power Spectral Density (PSD) of various levels of excitation (i.e. four RMS levels of 0.4 m.s −2 , 2 m.s −2 , 4 m.s −2 or 8 m.s −2 ). The sampling frequency and the resolution frequency are 12800 Hz and 0.39 Hz, respectively. The output signal is decomposed into 100 temporal blocks of 2.62 s each. A periodogram estimate computes an approximation of the PSD for each of the 100 blocks. Then, the PSD of the output signal is calculated by averaging all the estimates. The input PSD is constant over time to ensure that the response is stationary. The  input PSD is centered around the frequency of the primary resonance (i.e [115;135] Hz as seen in [3] ). It is constant over time to ensure that the output response is stationary. It can be noted that the dynamic behavior of the reference experimental structure of the CEA beam subjected to harmonic excitations has been previously performed in [1][2] which allows us to identify the value of the first resonance frequency.
Despite on the fact that the previous study scientific analysis [3] was carried out only for the frequency range [50;500] Hz, the frequency range of interest has been extended to [50;10 0 0] Hz for the open data provided. This gives the opportunity to researchers for conducting additional analysis compared to [3] and the development of extended analytical and numerical models for predicting the nonlinear dynamic behavior within a frequency range of interest [50;10 0 0] Hz.
Input and output experimental results for the four levels of excitation are plotted in Fig.  2 and Fig. 3 , respectively. All results on the frequency range [50;500] Hz have been previously analyzed and discussed in [3] for characterization of the nonlinear behavior of the CEA-beam structure beam subjected to a broadband random excitation. As previously explained in [3] , the presence of noise in the bandwidth at low and medium excitation levels (around

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships which have or could be perceived to have influenced the work reported in this article.