The module is only available as a service. 

The toolbox takes as input the data acquired from the external sensors (e.g. accelerometers) for different Degrees of Freedom (DoFs), and the minimum number of Degree of Freedoms (DoFs) to be retained (selected). Then, thanks to auto- and cross-informatio criteria, the best setup among the possibles is selected. Finally, the code save, in a dedicated external file, the selected (optimal) DoFs number.

The module is only available as a service.

The toolbox takes as input the data acquired from the external sensors (hardware component) for different points and different angle, the number of tested points, and the minimum number of Degree of Freedoms (DoFs) to be retained (candidates) for the Optimal Sensor Placement (OSP) step. Then, for each point, the horizontal and vertical directions that bring more information are estimated. Finally, the code estabilish a ranking of the DoFs sorting them in descending order of information degree. The coordinates of the retained DoFs for the OSP step are saved in a dedicated external file, such as all the results of the toolbox.

The module is only available as a service. 

The toolbox perform the Optimal Sensor Placement in terms of best Degree of Freedoms (DoFs), i.e. in terms of position and direction. The toolbox takes as input the type of objective function and the type of searching method for the minimization porblem. Then, it needs the definition of the maximum number of availabe uniaxial sesnors (e.g. accelerometers - triaxial accelerometers count for 3 sensors), the definition of the retained mode shapes among the total numerical available, a set of numerical mode shapes, the position and directions of the DoFs and the position and directions of the candidate DoFs used during the OSP procedure (searching subspace). The toolbox gives in output the otpimal (selected) DoFs to be instrumentated, i.e. position and direction, among the assumed candidates, and the MAC matrix evaluated on the selected DoFs and the retained mode shapes.

The module is available as a service and as a product with module 6 (software license). 

The toolbox performs Signal Analysis (pre-processing) and linear System Identification (with the Stochastic Subspace Identification - SSI). The toolbox takes a set of parallel acquisitions as input (with ambient vibrations as excitation source) and returns the identified natural frequencies, mode shapes, and damping ratios in output.

The module is only available as a service. 

The toolbox takes as input the identified modes (mode shape, natural frequencies and damping ratio) for different setups. In each setup the number of identified modes can change. However, in the different setups the same mode can appear, with, obviously, different channels and number of coordinates of the eigenvector. The output of the toolbox is represented by the so called composed modes, i.e. a set of modes (mode shape, natural frequencies and damping ratio) with all the instrumented channles of the different initial setups. The several setups must have at least 1 (better 2) coordinates of the eigenvectors that represent the same point (location), direction and verse of acquisition. The toolbox works on each idenitifed mode, indipendently, thus the actions should be repeted for the different modes that the user want to compose.

The module is available as a service and as a product with module 4 (software license). 

The toolbox aims to support the interpretation of dynamic monitoring data using statisics and machine learning tools. From the results of the structural identification over time, a training dataset is selected in which the structure is in an initial (healthy) condition. If available, temperature (or other enviromental and operational variables) data may also be involved in the analysis, as temperature-frequency dependence is demonstrated for many structures. One of the modes is indicated by the user, who should be more sensitive to the damage expected on the structure. The Support Vector Machine (SVM) algorithm is trained on this dataset, to build a model aimed at predicting the modal frequency values of the specific mode. The regression parameters can be meta-optimized or defined by the user. The toolbox provide as output the information on the global heath state of the structure based on three different state leves.

The module is only available as a service. 

The toolbox takes as input the identified modes (mode shape and natural frequencies) and a numerical Finite Element (FE) model of the system. The output of the toolbox is represented by the updated model parameters and the related modal quantities. By comparing the experimental values with the numerical one (in terms of modal quantities), the code performs the minimization of a cost function. The optimal (minimum) value found at the end of the optimization process represents the best solution associated to a FE model able to predict the modal behaviour of the system.