Instrumentation & Measurement Magazine 23-9 - 22

engineering. In fact, among the many different SHM systems,
the implementation of those devoted to vibration monitoring
is particularly challenging because it is usually based on dense
sensor networks, characterized by high sampling frequencies
and heavy-duty cycles.
Sensor networks built on MEMS accelerometers have recently drawn considerable attention [1] due to their ability to
precisely capture acceleration signals in a cost-effective manner. The structural characterization is then performed by
computing a set of damage-sensitive parameters embedded
in vibration data. However, the data retrieval and processing
tasks are strictly application-dependent; thus, apart from general recommendations [7], no precise standardization has been
formalized yet.
Operational Modal Analysis (OMA) is a widely adopted
strategy to extract meaningful features from vibration data,
and it can be performed when the structures are in operation
and the loading conditions (traffic, wind, seismic events, etc.)
are unknown [8]. OMA procedures are fed with vibration-related signals (e.g., accelerations, rotation) and output the so
called " modal parameters. " These features may comprise natural frequencies (i.e., the frequency components carrying most
of the total structural energy), damping factors and mode
shapes, namely the specific spatial patterns of vibrations exhibited by the monitored structure at the different natural
frequencies.
A schematic overview of a typical OMA-based processing
flow is depicted in Fig. 1, illustratively comprising a monitoring application with Ns = 9 accelerometers and P = 3 natural
frequencies and as many identified mode shapes.
Vibration signals (ai(t), i = 1...Ns) acquired at individual sampling positions (Ai) are the only input required by the system.
As it can be observed, the set f = [f1...fp] of P natural (modal) frequencies is identified from the collection of P dominant peaks

that appear in the Power Spectral Density (PSD) profile of
gathered signals. A global estimation of the cumulative vibration frequencies is commonly obtained as a point-by-point
average of the peak frequency values estimated at each sensor of the network. Alongside, the absolute value of the p-th
N   1
mode shape vector Φ p   s , corresponding to the equallyindexed modal frequency fp, can be trivially reconstructed
by interpolating in spatial domain the previously computed
peak spectral magnitudes. Once all of the P mode shape vectors have been estimated, they can be vertically arranged as
N   P
columns of the mode shape
matrix Φ   Φ1  Φ P    s .
Through conventional spectral analysis tools, just the absolute
value of the mode shape can be extracted; therefore, more advanced techniques have been developed to reconstruct the
actual modal curve such as eigenvector-based algorithms or
Blind Source Separation strategies [8].

Proposed SHM Architecture
Even if there is a growing number of SHM solutions presented in
literature [6], two main difficulties still hamper their wide adoptions, i.e.,: the lack of standard sensing solutions and estimation
methods, and the need for adequate data management tools
to aggregate, process and analyze the possibly big-data volume produced by the sensor devices for fine-grained predictive
maintenance applications. The issues discussed above are tackled within the MAC4PRO project [4], where a reference SHM
architecture which integrates the traditional components of
multi-source structural monitoring with data management and
analysis is proposed. Specifically, three functional requirements
have been considered during the design and deployment of the
HW/SW elements: (1) scalability, i.e., the possibility to cope with
large sensor installations likely producing high data volumes;
(2) heterogeneity, namely the need to support multi-type sensor
devices (e.g., MEMS and piezoelectric transducers) with different data formats, required
estimation procedures and
outputs; (3) extendibility, i.e.,
the seamless support for
the dynamic adding of new
sensors and/or their remote
configuration updating.
The proposed architecture
consists of three main layers, as shown in Fig. 2; in the
following sections, the enabling technologies of each
architectural level are discussed, while an integrated
validation on an SHM usecase is presented.

Data Measuring
Layer

Fig. 1. Time-Frequency-Spatial domain representation of a typical OMA-based processing flow.
22	

IEEE Instrumentation & Measurement Magazine	

The measuring layer is a
sensor network composed
of low-power, light weight
December 2020



Instrumentation & Measurement Magazine 23-9

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 23-9

No label
Instrumentation & Measurement Magazine 23-9 - No label
Instrumentation & Measurement Magazine 23-9 - Cover2
Instrumentation & Measurement Magazine 23-9 - 1
Instrumentation & Measurement Magazine 23-9 - 2
Instrumentation & Measurement Magazine 23-9 - 3
Instrumentation & Measurement Magazine 23-9 - 4
Instrumentation & Measurement Magazine 23-9 - 5
Instrumentation & Measurement Magazine 23-9 - 6
Instrumentation & Measurement Magazine 23-9 - 7
Instrumentation & Measurement Magazine 23-9 - 8
Instrumentation & Measurement Magazine 23-9 - 9
Instrumentation & Measurement Magazine 23-9 - 10
Instrumentation & Measurement Magazine 23-9 - 11
Instrumentation & Measurement Magazine 23-9 - 12
Instrumentation & Measurement Magazine 23-9 - 13
Instrumentation & Measurement Magazine 23-9 - 14
Instrumentation & Measurement Magazine 23-9 - 15
Instrumentation & Measurement Magazine 23-9 - 16
Instrumentation & Measurement Magazine 23-9 - 17
Instrumentation & Measurement Magazine 23-9 - 18
Instrumentation & Measurement Magazine 23-9 - 19
Instrumentation & Measurement Magazine 23-9 - 20
Instrumentation & Measurement Magazine 23-9 - 21
Instrumentation & Measurement Magazine 23-9 - 22
Instrumentation & Measurement Magazine 23-9 - 23
Instrumentation & Measurement Magazine 23-9 - 24
Instrumentation & Measurement Magazine 23-9 - 25
Instrumentation & Measurement Magazine 23-9 - 26
Instrumentation & Measurement Magazine 23-9 - 27
Instrumentation & Measurement Magazine 23-9 - 28
Instrumentation & Measurement Magazine 23-9 - 29
Instrumentation & Measurement Magazine 23-9 - 30
Instrumentation & Measurement Magazine 23-9 - 31
Instrumentation & Measurement Magazine 23-9 - 32
Instrumentation & Measurement Magazine 23-9 - 33
Instrumentation & Measurement Magazine 23-9 - 34
Instrumentation & Measurement Magazine 23-9 - 35
Instrumentation & Measurement Magazine 23-9 - 36
Instrumentation & Measurement Magazine 23-9 - 37
Instrumentation & Measurement Magazine 23-9 - 38
Instrumentation & Measurement Magazine 23-9 - 39
Instrumentation & Measurement Magazine 23-9 - 40
Instrumentation & Measurement Magazine 23-9 - 41
Instrumentation & Measurement Magazine 23-9 - 42
Instrumentation & Measurement Magazine 23-9 - 43
Instrumentation & Measurement Magazine 23-9 - 44
Instrumentation & Measurement Magazine 23-9 - 45
Instrumentation & Measurement Magazine 23-9 - 46
Instrumentation & Measurement Magazine 23-9 - 47
Instrumentation & Measurement Magazine 23-9 - 48
Instrumentation & Measurement Magazine 23-9 - 49
Instrumentation & Measurement Magazine 23-9 - 50
Instrumentation & Measurement Magazine 23-9 - 51
Instrumentation & Measurement Magazine 23-9 - 52
Instrumentation & Measurement Magazine 23-9 - 53
Instrumentation & Measurement Magazine 23-9 - 54
Instrumentation & Measurement Magazine 23-9 - 55
Instrumentation & Measurement Magazine 23-9 - 56
Instrumentation & Measurement Magazine 23-9 - 57
Instrumentation & Measurement Magazine 23-9 - 58
Instrumentation & Measurement Magazine 23-9 - 59
Instrumentation & Measurement Magazine 23-9 - 60
Instrumentation & Measurement Magazine 23-9 - 61
Instrumentation & Measurement Magazine 23-9 - 62
Instrumentation & Measurement Magazine 23-9 - 63
Instrumentation & Measurement Magazine 23-9 - 64
Instrumentation & Measurement Magazine 23-9 - 65
Instrumentation & Measurement Magazine 23-9 - 66
Instrumentation & Measurement Magazine 23-9 - 67
Instrumentation & Measurement Magazine 23-9 - 68
Instrumentation & Measurement Magazine 23-9 - 69
Instrumentation & Measurement Magazine 23-9 - 70
Instrumentation & Measurement Magazine 23-9 - 71
Instrumentation & Measurement Magazine 23-9 - 72
Instrumentation & Measurement Magazine 23-9 - Cover3
Instrumentation & Measurement Magazine 23-9 - Cover4
https://www.nxtbook.com/allen/iamm/25-9
https://www.nxtbook.com/allen/iamm/25-8
https://www.nxtbook.com/allen/iamm/25-7
https://www.nxtbook.com/allen/iamm/25-6
https://www.nxtbook.com/allen/iamm/25-5
https://www.nxtbook.com/allen/iamm/25-4
https://www.nxtbook.com/allen/iamm/25-3
https://www.nxtbook.com/allen/iamm/instrumentation-measurement-magazine-25-2
https://www.nxtbook.com/allen/iamm/25-1
https://www.nxtbook.com/allen/iamm/24-9
https://www.nxtbook.com/allen/iamm/24-7
https://www.nxtbook.com/allen/iamm/24-8
https://www.nxtbook.com/allen/iamm/24-6
https://www.nxtbook.com/allen/iamm/24-5
https://www.nxtbook.com/allen/iamm/24-4
https://www.nxtbook.com/allen/iamm/24-3
https://www.nxtbook.com/allen/iamm/24-2
https://www.nxtbook.com/allen/iamm/24-1
https://www.nxtbook.com/allen/iamm/23-9
https://www.nxtbook.com/allen/iamm/23-8
https://www.nxtbook.com/allen/iamm/23-6
https://www.nxtbook.com/allen/iamm/23-5
https://www.nxtbook.com/allen/iamm/23-2
https://www.nxtbook.com/allen/iamm/23-3
https://www.nxtbook.com/allen/iamm/23-4
https://www.nxtbookmedia.com