A strategy to control industrial plants in the spirit of Industry 4.0 tested on a fluidic system


ACTA IMEKO 
ISSN: 2221-870X 
June 2022, Volume 11, Number 2, 1 - 7 

 

ACTA IMEKO | www.imeko.org June 2022 | Volume 11 | Number 2 | 1 

A strategy to control industrial plants in the spirit of Industry 
4.0 tested on a fluidic system 

Laura Fabbiano1, Paolo Oresta1, Rosario Morello2, Gaetano Vacca1 

1 DMMM, Politecnico di Bari University, Bari, Italy 
2 DIIES, University Mediterranea of Reggio Calabria, Italy 

 

 

Section: RESEARCH PAPER  

Keywords: Industry 4.0; predictive maintenance; prognostic approach; plant operation simulator; fluidic thrust plant   

Citation: Laura Fabbiano, Paolo Oresta, Rosario Morello, Gaetano Vacca, A strategy to control industrial plants in the spirit of Industry 4.0 tested on a fluidic 
system, Acta IMEKO, vol. 11, no. 2, article 31, June 2022, identifier: IMEKO-ACTA-11 (2022)-02-31 

Section Editor: Francesco Lamonaca, University of Calabria, Italy 

Received November 9, 2021; In final form February 21, 2022; Published June 2022 

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 License, which permits unrestricted use, 
distribution, and reproduction in any medium, provided the original author and source are credited. 

Corresponding author: Laura Fabbiano, e-mail: laura.fabbiano@poliba.it  

 

1. INTRODUCTION 

The costs of maintenance are a consistent amount of the total 
operating costs in industrial production. Moreover, several extra 
costs come from the profit loss due to the undesired failure of 
the plant. They can be dropped down with a failure prediction 
that preserves the devices through just on-time maintenance. 

These two classes of problems can be solved by the 
implementation of preventive maintenance in the perspective of 
the smart industry, [1]-[3], for which all the elements of a factory 
work in a completely, collaborative and integrated way and deal 
in real-time with the timely changes of the workflow. 

Concerning costs reduction in force of the just on-time failure 
detection, the goal of the smart industry is to implement all the 
fundamental measurement procedures for continuous real-time 
monitoring (real-time condition monitoring) and coordinated 
monitoring of all plant elements, [4]-[6]. 

The keywords of the fourth industrial revolution are, 
therefore "preventive maintenance", "intelligent and real-time 

orchestration" and "synchronization of physical and digital 
processes". 

In literature, there are crucial new insights about the 
continuous screening of the devices that made possible the 
detection of incipient failure, [7]-[9]. Among others, the 
prognostic approach is mandatory to obtain an accurate and 
coding of incipient machine failures, [10]-[12]. The detection of 
the degradation and damage of a plant component is the goal of 
the predictive approach. The condition-based maintenance 
(CBM) specifications [13], [14] inspired by the prognostic 
approach has been applied by the authors to the case of a simple 
fluidic thrust system by using a mathematical approach. It 
consists of a pump-motor block with inverter and a control 
valve, which represents the load acting on the P-M block and 
attributable to the network supplied by it. 

The approach uses a numerical simulator to manage and 
analyse in real-time all the characteristic parameters (sensed or 
mathematically predicted) of each system components to get and 
to advise concerning likely incipient anomalies of the monitored 
components. 

ABSTRACT 
The goal of the paper is to propose a strategy of automating the control of wide spectrum industrial processes plants in the spirit of 
Industry 4.0. The strategy is based on the creation of a virtual simulator of the operation of the plants involved in the process. Through 
the digitization of the operational data sheets of the various components, the simulator can provide the reference values of the process 
control parameters to be compared with their actual values, to decide the direct inspection and/or the operational intervention on 
critical components before a possible failure. As example, a simple fluidic thrust plant has been considered, which a mathematical model 
(simulator) for its optimal operating conditions has been formulated for, by using the digitalized real operational data sheets of its 
components. The simple thrust system considered consists of a centrifugal pump driven by a three-phase electric motor, an inverter to 
regulate the rotation of the motor and a proportional valve that simulates the external load acting on the pump. 
As results, the operational data sheets and principal characteristics of the pump have been reproduced by means of the simulator here 
developed, showing a very good agreement. 

mailto:laura.fabbiano@poliba.it


ACTA IMEKO | www.imeko.org June 2022 | Volume 11 | Number 2 | 2 

The virtual simulator is based on the digital reconstruction of 
the data sheets of all the more sensitive components of the 
system to allow the digital prediction of the optimal point of 
system working conditions. These data provided by model are 
compared with the real-time instantaneous acquisitions of the 
correspondent main operating parameters. The eventual 
discrepancy between virtual and actual values of the monitored 
parameter identifies which of the components of the system may 
be in critical conditions, thus dropping down its start and stop 
time responsible of the maintenance and service interruption 
costs. The proposed procedure, based on the digitization of the 
technical data sheets of the plant devices can be easily extended 
to most of the systems operating in industrial processes, thus 
allowing to control the entire system in real-time and detecting 
the dangerous failure scenarios. 

2. PLANT SPECIFICATIONS 

The system here simulated is a simple fluidic thrust plant and 
consists of a centrifugal pump (CALPEDA NM4 25/12A/A, 
[15]) driven by an electric asynchronous motor whose rotation 
speed is regulated by an inverter, and a control valve simulating 
the supplied network. The operational point of such system is 
determined by the coupling of the pump with the valve. The 
technical and geometric characteristics of each component are 
known. Those of the pump are here reported: 

𝑛 =  1450 rpm  rotational speed of reference 
𝑃 =  0.25 kW  absorbed power at 1450 rpm 
𝑄 =  1 ÷  6 m3/h  flow rate working range 
𝐻𝑢  =  6.1 ÷  3.3 m  head range 
𝐷2  =  131.5 mm  external diameter of the impeller 
𝛽2  =  157.5 °   angle of blade exit 
𝑙2 =  4 mm  blade height 
𝑧 =  7  number of blades 
𝜁 =  1 −

𝑧𝑠

π 𝐷2
= 0.95  blade bulk coefficient 

A user interface has been set up to manage the procedure of 
analysis and acquisition of data. Specifically, that interface has 
been created on LabVIEW platform and is divided into several 
sections; each section manages and analyses the different parts 
of the plant, Figure 1. 

In the control section you find the commands to change the 
opening degree of the valve and the inverter frequency. 

The target section enforces the rotation number of the engine. 
Once the frequency is set by the control panel, it automatically 
varies until the system reaches a stable point of operation where 
the number of revolutions of the pump coincides with the value 
imposed in the target panel. 

The transient panel shows the indicators of instant 
monitoring of the plant. In particular, the quantities acquired by 
the transducers (number of revolutions, mechanical torque, flow 
rate, head) and some derived quantities such as instantaneous 
absorbed power are represented. 

The results panel collects in graphic form (three different 
plots) the trends of the pump head (characteristic curve), 
efficiency and absorbed power as a function of the flow rate. 

3. VIRTUAL SIMULATOR 

The mathematical model of the plant consists of:  
1. valve model; 
2. pump - motor block model. 
Most of the geometric and operating data have been inferred 

from the technical specifications of the components. 

3.1. Valve model 

The control valve (here a proportional solenoid valve) allows 
the modulation of the circuit external characteristic (valve 
characteristic), [16]. That is, by acting on the closing/opening 
control devise, the relationship between the pressure drops and 
the flow rate flowing through the valve changes, so determining 
a new operating point as intersection with the internal 
characteristic (pump operating curve). 

From Bernoulli's equation the relationship to the basis of the 
valve operation comes out as: 

𝑄 = [𝑐𝑣max ℎ + 𝑐𝑣min (1 − ℎ)]√
𝐻𝑣
𝐻𝑣

∗
 , (1) 

where:  

• 0 < ℎ < 1 degree of opening of the valve (or shutter 
stroke); 

• 𝑐𝑣max = 0.005 m
3/h and 𝑐𝑣min = 10

−6 m3/h flow rates (or 

efflux coefficients) minimum and maximum estimated by 
the technical data sheet of the pump; 

• 𝐻𝑣 → static pressure drop through the valve (m); 

• 𝐻𝑣
∗ → 1 𝑝𝑠𝑖 static pressure drop through the valve (m). It 

represents the linear operational limit of the valve itself. 
From the previous relationship we get: 

𝐻𝑣 = 𝑄
2

𝐻𝑣
∗

[𝑐𝑣max ℎ + 𝑐𝑣min (1 − ℎ)]
2
 (2) 

If a linear valve is considered, its characteristic coefficient 

𝐾𝑣 (ℎ) reads as: 

𝐾𝑣 (ℎ) =
𝐻𝑢

∗

[𝑐𝑣𝑚𝑎𝑥 ℎ + 𝑐𝑣𝑚𝑖𝑛 (1 − ℎ)]
2

 , (3) 

so, making possible to rewrite the previous relation as: 

𝐻𝑣 = 𝑄
2𝐾𝑣 (ℎ) (4) 

The efflux coefficient of the valve is an increasing 

monotonous function of the shutter stroke, 𝑐𝑣 (ℎ), and it can be 
expressed in dimensionless form if the following quantities are 
considered: 

• the relative efflux coefficient: 𝜑 =  𝑐𝑣 /𝑐max 

• the intrinsic rangeability, ratio between the maximum 
and the minimum values of the efflux coefficient:  

𝑟 =  𝑐𝑣max /𝑐𝑣min   

 

Figure 1. User interface in LabVIEW platform.  



ACTA IMEKO | www.imeko.org June 2022 | Volume 11 | Number 2 | 3 

Examples of valve characteristics expressed in terms of ℎ are 
shown in Figure 2 for linear, equal percentage, fast-opening and 
quadratic valves: 

For the linear characteristic, it holds the relation: 

𝜑(ℎ) = ℎ +
1

𝑟
 (1 − ℎ) (5) 

In the plant operating model (simulator), the characteristic 

coefficient of the valve can be rewritten as, from (4): 

𝐾𝑣 (ℎ) =
𝐻𝑣

∗

𝑐𝑣 (ℎ)
2

=
𝐻𝑣

∗

𝑐𝑣 max
2 𝜑(ℎ)2

 (6) 

this relation for 𝐾𝑣 (ℎ) keeps holding even for other kind of 
valves, if the right formula for 𝜑(ℎ) is introduced in it. 

3.2. Motor pump block 

The 'Pump-Motor' block, Figure 3, represents the virtual 
simulator of a centrifugal pump and its electric control motor. 

This block has 2 inputs and 4 outputs. The 2 inputs are, 
respectively: 

1. The coefficient of the valve, whose operating 
characteristic is enclosed in the 'Valve' model and is 
provided by it. Its regulation will result in the variation of 

the number of revolutions of the pump, 𝑛 
2. The magnetic field frequency, parameter recalled by the 

control unit, which allows to act in feedback on it, in 
order to restore the number of revolutions to a target and 
constant value during the process of digital 
reconstruction of the characteristic curves of the pump. 

The 4 outputs represent the virtual sensors of the P-M block 
suitable for the detection of: 

1. Q flow rate (m³/h) 

2. 𝐻𝑢   head provided by the pump (m) 
3.  𝑛 number of engine revolutions (rpm) 
4. 𝐶𝑚 drive torque, (N m). 
They are recalled by the supervisor function (see Figure 1) and 

made visible in the transient panel, where the instant monitoring 
indicators of the system are reported. In particular, the quantities 

acquired by the transducers (for a number of revolutions, 
mechanical torque, flow rate, head) and some derived quantities 
such as the instantaneous absorbed power are shown. 

In the model there are some geometric parameters directly 
available from the technical characteristics of the pump; other 
necessary parameters are inferred from the characteristic curves 
and the power absorbed by the pump, for a fixed pump 
revolution. 

The internal characteristic of the pump at the revolution 

𝑛0 =  450 rpm can be expressed in polynomial form as follows: 

𝐻𝑢 = −0.0893 𝑄
2 + 0.0706 𝑄 + 6.104 (7) 

obtained by best square polynomial regression of the pump, 
reported in the Table 1 and shown in Figure 4. 

It is possible to generalize this relation and make it valid for 
any number of revolutions, as: 

𝐻𝑢 = 𝑎 𝑄
2 + 𝑏

𝑛

𝑛0
𝑄 + 𝑐 (

𝑛

𝑛0
)

2

 (8) 

where 𝑎, 𝑏 and 𝑐 are the coefficients of the specific previous 
relation. 

Substituting the expression of the valve characteristic (4) in 
(8), in the steady state operating condition of the system, we 
obtain: 

[𝑎 − 𝐾𝑣 (ℎ)]𝑄
2 + 𝑏

𝑛

𝑛0
𝑄 + 𝑐 (

𝑛

𝑛0
)

2

= 0.  (9) 

By expressing then 𝑄 as a function of the geometrical and 
operating parameters of the pump as: 

𝑄 =  𝐾𝑄  𝜑 𝑛 , (10) 

where 𝐾𝑄 = 𝜂𝜈 𝜁 π
2 𝑙2 𝐷2

2, and substituting it in the equation 

that regulates the operation of the plant (9) we get the new 
relation: 

𝐾𝜙0
2 [𝑎 − 𝐾𝑣 (ℎ)]𝜙

2 + 𝑏𝐾𝜙1 𝜙 + 𝐾𝜙 2
= 0 , (11) 

where 𝐾𝜙0 = 𝐾𝑄
2 𝑛0

2[𝑎 − 𝐾𝑣 (ℎ)], 𝐾𝜙1 = 𝑏 𝐾𝑄  𝑛0, 𝐾𝜙2 = 𝑐, 

and 𝜙 the flow parameter. 
The equality of the head provided by the pump with the 

pressure drop introduced by the control valve is not sufficient to 
describe the operation of the experimental system. We need the 

 

Figure 2. Valve characteristic curves in terms of the flow rate coefficient, 𝜙 
as function of the opening position, h. The curves show the cases of linear, 
parabolic, equal percentage and fast opening design.  

 

Figure 3. Inputs and outputs of the pump-motor block.  

 

Figure 4. Pump performance curve at 𝑛 = 1450 rpm. Comparison between 
simulation and actual operational point.  

Table 1. Pump data. 

Q in m³/h 1 1.2 1.5 1.8 2.4 3 3.6 4.2 4.8 5.4 6 

H in m 6 6.05 6 5.9 5.8 5.5 5.2 4.8 4.4 3.9 3.3 

 0

 2

 4

 6

 8

 0  2  4  6  8

H
u
 [

m
]

Q [m
3
 / h]

best-fit
data sheet



ACTA IMEKO | www.imeko.org June 2022 | Volume 11 | Number 2 | 4 

power conservation law, for which the power supplied by the 
motor equals the power absorbed by the pump. 

The last one is expressed by: 

𝑃𝑎𝑝 =
𝛾 𝑄 𝐻𝑢

𝜂𝑦  𝜂𝑣  𝜂𝑚
=

𝜌 𝑄 𝐿𝑖
𝜂𝑣  𝜂𝑚

=
𝜌 𝐾𝑄  𝜙 𝜓 π 𝐷

2 𝑛2

2 𝜂𝑣 𝜂𝑚
 (12) 

in which 𝜓 = 2(1 − 𝜙 cotg𝛽2) − (2 π sin 𝛽2)/𝑧 and 
𝜂𝑣  𝑒𝑑 𝜂𝑚 are the volumetric and mechanical efficiencies to be 
discussed in the paragraph of the results, where the methods to 
evaluate their values during the actual operation of the plant for 
its monitoring will be described. Both could result indices of 
possible anomalies. 

With reference to the expression of the power supplied by the 

motor, 𝑃𝑚, the mechanical torque, 𝐶𝑚, can be expressed.  
The mechanical torque 𝐶𝑚 of the electric motor can be 

expressed as a function of the slip, 𝑠, according to the following 
relation: 

𝐶𝑚  =  2𝐶𝑚 max  
𝑠max ∙ 𝑠

(𝑠max
2  +  𝑠2) 

 (13) 

in which 𝐶𝑚max  is the maximum torque available for the slip 

value 𝑠 =  𝑠max, available from the relative data sheet and equal 
to 𝐶𝑚 = 50 N m and 𝑠 = 0.2. 

The slip is representative of the difference between the speed 
of rotation of the rotor, i.e., the shaft and the speed of rotation 

of the magnetic field, 𝑛𝑠,that is: 

𝑠 =
𝑛𝑠 − 𝑛

𝑛𝑠
 , (14) 

where: 

𝑛 =
60𝑓

𝑝
 (1 −  𝑠) (15) 

with 𝑝 = 2, motor polar couples. 
The mechanical characteristic of the motor, 𝐶𝑚, is 

represented as function of the revolution of the shaft in Figure 5 
for 3 different values of frequency. 

This characteristic shows that when the slip is roughly zero 
the torque reaches its maximum value, and the motor speed is 

close to the synchronism speed 𝑛𝑠. 
The efficiency 𝜂 of the three-phase asynchronous motor can 

be calculated with the well-known formula: 

𝜂 =
𝑃𝑟
𝑃𝑎

 , (16) 

where 𝑃𝑟  is the mechanical power supplied to the rotor and 𝑃𝑎  is 
the electrical power provided by the stator. 

Equating the power absorbed by the pump to the power 
supplied by the motor, we obtain: 

𝜌 𝐾𝑄  𝜙 𝜓 π 𝐷
2 𝑛2

2 𝜂𝑣  𝜂𝑚
= (2𝐶𝑚 max  

𝑠max ∙ 𝑠

𝑠max
2 +  𝑠2

) 2 π 𝑛 (17) 

and plugging into it the expression of the slip 𝑠, we get: 

𝜌 𝐾𝑄  𝜙 𝜓 π 𝐷2
2 (

𝑝

𝑓
)

2

𝑛4 −
𝜌 𝐾𝑄  𝜙 𝜓 π 𝐷2

2 2 𝑝

𝑓
𝑛3

+ 𝜌 𝐾𝑄  𝜙 𝜓 π 𝐷2
2(𝑠max + 1) 𝑛

2

+ 8 𝐶𝑚 max 𝑠max  𝜂𝑣  𝜂𝑚
𝑝

𝑓
𝑛

− 8 𝐶𝑚 max 𝑠max  𝜂𝑣  𝜂𝑚 = 0. 

(18) 

The previous equation can be rewritten in the easier-to-read 
following way: 

𝐾𝑛0 𝑛
4 + 𝐾𝑛1 𝑛

3 + 𝐾𝑛2 𝑛
2 + 𝐾 𝑛3 𝑛 + 𝐾𝑛4 = 0, (19) 

with obvious meaning of the symbols. 
The former equation represents the main equation of the 

plant operating model. Such an equation can be solved iteratively 
by the Newton-Raphson method, for example, to provide the 
revolution speed of the system for a fixed working condition. 
The iterations, in the present case, have been interrupted when 

the percentage error on 𝑛 was less than 10−5. 
The described set of model equations of the thrust system 

components constitute the simulator of its working operation. 

4. RESULTS 

The volumetric efficiency of the pump changes according to 
the operating regime. It is negligible for null flow rates in the case 
of a control valve fully closed. As the flow rate increases, the 
volumetric efficiency increases up to the plateau value that is 
almost constant for a wide range of the pump flow rate, in steady 
state operation. The data sheet values of it are shown in Figure 6 
here below. 

In the case of high flow rate, the following procedure aims to 
calculate the values of the mechanical and the volumetric 

efficiencies, 𝜂𝑚, 𝜂𝑣∞ (asymptotic value), respectively. The 
pressure parameter, 𝜓, can be computed, in the following two 
ways: 

𝜓 =
2 𝑔 𝐻𝑢

𝑢2
2 𝜂𝑝

𝜂𝑚𝜂𝑦  (20) 

or 

𝜓 = 2(1 + 𝜙 cotg 𝛽2) −
2 π sin𝛽2

𝑧
 . (21) 

Equating the above equations and rewriting the flow 

parameter as a function of the flow rate, 𝜙 =  𝜙(𝑄), we get: 

 

Figure 5. Characteristics of the asynchronous motor in terms of torque as 
function of revolution, for different frequency values.  

 

Figure 6. Pump volumetric efficiency experimental data.  



ACTA IMEKO | www.imeko.org June 2022 | Volume 11 | Number 2 | 5 

2 𝑔 𝐻𝑢

𝑢2
2 𝜂𝑝

=
2 cotg 𝛽2

𝜂𝑣  𝜉 π
2 𝑙2 𝐷2

2  𝑛
 

1

𝜂𝑚  𝜂𝑣
𝑄

+
2

𝜂𝑚 𝜂𝑣
(1 −

π sin 𝛽2
𝑧

) . 

(22) 

From the experimental data of the pump CALPEDA NM4 
25/12A/A, it is observed that for flow rates greater than a 

threshold value (0.001 m3/s), the term 2 𝑔 𝐻𝑢 /(𝑢2
2 𝜂𝑝) is a 

linear function of 𝑄, Figure 7. 
In the linearity range, the coefficients of the previous equation 

can be calculated by best fitting the pump data sheet discrete 
values, Figure 8, and therefore it is possible to estimate the 

unknown values of the mechanical efficiency 𝜂𝑚 and the 
volumetric efficiency 𝜂𝑣 . 

Specifically, the mechanical efficiency is equal to 𝜂𝑚 = 0.955 
and it is assumed constant for the whole operating regime, 
whereas the volumetric efficiency obtained from this procedure 
represents the asymptotic limit only in the range of high flow 
rates and it is equal to 𝜂𝑣∞ = 0.719. 

Known the value of the mechanical efficiency, from the (22), 

𝜂𝑣  can be made explicit as function of the flow rate only, such as 
𝜂𝑣 =  𝜂𝑣(𝑄). 

The relationship 𝜂𝑣 =  𝜂𝑣 (𝑄) is well described by the 
following relation: 

𝜂𝑣 = 𝜂𝑣∞ (1 − e
−

𝑄
𝜏 ) , (23) 

where the value of 𝜏 = 3.83 ∙ 10−4𝑚3/𝑠 is obtained from the 
best-fit polynomial of the available data operation data of the 
pump with 𝜂𝑣∞ = 0.719. The following figure shows the good 
agreement between the theoretical prediction and pump data 
over the whole range of the flow rates. 

The simulator function uses this model of the volumetric 
efficiency and the prediction of the mechanical efficiency. It 
provides a good agreement of the control parameters with the 

characteristic data of the pump, in terms of absorbed power 𝑃𝑎 , 
efficiency 𝜂𝑝 and pressure head 𝐻𝑢 . 

Through the simulator, described in the previous section, it is 
possible to digitally reconstruct the pump characteristic curves, 
Figure 10, Figure 11 and Figure 12. 

By using the digitalized technical and operating data of the 
system components and setting a value of the frequency of the 
inverter regulating the number of revolutions of the pump-motor 
block, this one can be iteratively determined from (19), and then 
the outputs of the pump-motor are calculated from its model, i.e. 

the flow rate, the head, the torque 𝐶𝑚  and the efficiency of the 
pump 𝜂𝑝, which constitute the virtual values with which to 
compare the physical ones acquired (and calculated, such as 
mechanical and volumetric efficiencies) by the sensors prepared 
for that purpose in the system. 

So, the entire performance curves of the pump can be 
reconstructed and digitalized just varying the opening degree of 
the valve for each of the desired discrete values of the frequency 
(revolution speed of the pump), and by reiterating the use of the 
simulator equations as reported in the sections. 

 

Figure 7. experimental values for the left-hand side term in (22) as a function 
of the flow rate.  

 

Figure 8. Linear regression of experimental data for the left-hand side term 
of (22) as function of the flow rate.  

 

Figure 9. Least square regression of 𝜂𝑣  experimental data.  

 

Figure 10.Expected pump efficiency from the simulator compared with actual 
operational points.  

 0

 1

 2

 3

 0  0.0005  0.001  0.0015  0.002  0.0025

2
g
H

u
/(
h

p
 u

2
)

Q [m
3
 / s]

data sheet



ACTA IMEKO | www.imeko.org June 2022 | Volume 11 | Number 2 | 6 

The Figure 13 shows the pump performance curves evaluated 
by the simulator for 3 revolution speed values, as example of its 
application.  

The operating parameter values predicted by the models, 
including the mechanical and volumetric efficiency, are the 
reference targets for eventual anomalies detection when 
compared with the relative acquired values. The eventual 
significative discrepancy between the predicted and sensed 

values of each characteristic quantity can indicate an incipient 
criticality for the component whom it refers to. 

In these conditions it is possible to report and provide for the 
irregularity of the operation with an active control on the system. 
In addition, it is even possible to understand the origin of the 
disagreement between the real and expected values of mechanical 
or volumetric efficiency. For example, with reference to their 
possible discrepancy it is possible to suspect a critical condition 
of the bearings or seals of the pump-motor block. 

5. DISCUSSION AND CONCLUSION 

The intent of the authors was to propose a strategy for the 
automation of the monitoring of wide-ranging industrial 
processes, in the spirit of the "Smart Industry", in order to reduce 
the risks of sudden default of the activities caused by the possible 
criticality of the most delicate components, to reduce the time to 
identify the criticality, to reduce the maintenance time necessary 
to restore normal operating conditions, thus reducing the 
associated costs. The strategy is based on the creation of a virtual 
simulator of the operation of the various plants involved in the 
process which, through the digitization of the data sheets of the 
various components, can provide the reference values of the 
process control parameters. Those values are so compared with 
the values acquired by the measuring chains predisposed for the 
purpose, to allow the operator the reporting of any suspected 
anomaly in place and to quickly intervene to restore optimal 
operating conditions through targeted maintenance. To this end 
and by way of an example, we have proposed the formulation of 
a simulator of a simple fluidic thrust system, consisting of a 
pump-motor block with inverter and a regulation valve, which 
represents the load acting on the PM block and attributable to 
the network it supplies. The simulator allows to manage and 
analyze in real time all the characteristic parameters, acquired 
and/or calculated, of each of the monitored system components, 
during normal operation, to determine if there are conditions of 
possible incipient anomalies on the components under 
observation. This operation allows to compare, through real-
time acquisitions, the instantaneous values of the characteristic 
operating parameters with those provided by the simulator, 
corresponding to those that should be in the optimal operating 
conditions of the system. So, it is possible to identify which 
characteristic parameter of the various monitored components 
of the system reveals a more discrepant value from the optimal 
one, thus denouncing a possible critical condition of the 
component to which it refers to. In this way, the operator can 
decide to intervene on the component in time, thus minimizing 
the intervention times and therefore the maintenance and 
restoration costs of the normal operation of the system. Is 
opinion of the authors that the identified procedure, based on 
the digitization of the technical data sheets of the plant 
components, is extendable to most of the operating plants of an 
industrial process, thus allowing the entire process to be 
controlled in real time. 

REFERENCES 

[1] R. A. Luchian, G. Stamatescu, I. Stamatescu, I. Fagarasan, D. 
Popescu, IIoT Decentralized System Monitoring for Smart 
Industry Applications, 2021 29th Mediterranean Conference on 
Control and Automation (MED), 2021, pp. 1161-1166.   
DOI 10.1109/MED51440.2021.9480341  

[2] L. Fabbiano, G. Vacca, G. Dinardo, Smart water grid: A smart 
methodology to detect leaks in water distribution networks, 

 

Figure 11. Expected pump manometric head from the simulator compared 
with actual operational points.  

 

Figure 12. Expected pump absorbed power from the simulator compared 
with actual operational points.  

 

Figure 13. Expected pump performance curves from the simulator for three 
revolution speed (red lines) and four 𝜂𝑝 iso-lines.  

https://doi.org/10.1109/MED51440.2021.9480341


ACTA IMEKO | www.imeko.org June 2022 | Volume 11 | Number 2 | 7 

Measurement, Vol. 151, (2020).  
DOI: 10.1016/j.measurement.2019.107260 

[3] L. Ardito, A. Messeni Petruzzelli, U. Panniello, A. Garavelli, 
Towards Industry 4.0: Mapping digital technologies for supply 
chain management-marketing integration, Business Process 
Management Journal, Vol. 25, (2019), No. 2, pp. 323-346.  
DOI: 10.1108/BPMJ-04-2017-0088 

[4] A. J. Isaksson, I. Harjunkoski, G. Sand, The impact of 
digitalization on the future of control and operations, Comput. 
and Chem. Eng., 114, (2018), pp. 122–129.   
DOI: 10.1016/j.compchemeng.2017.10.037  

[5] G. Dinardo, L. Fabbiano, G. Vacca, A smart and intuitive machine 
condition monitoring in the Industry 4.0 scenario. Measurement, 
126, (2018), pp. 1–12.  
DOI: 10.1016/j.measurement.2018.05.041 

[6] M. Short, J. Twiddle, An Industrial Digitalization Platform for 
Condition Monitoring and Predictive Maintenance of Pumping 
Equipment. Sensors, 19, (2019), 3781.  
DOI: 10.3390/s19173781 

[7] P. Girdhar, C. Scheffer, Predictive maintenance techniques: Part 1 
predictive maintenance basics, in Practical Machinery Vibration 
Analysis and Predictive Maintenance, P. Girdhar and C. Scheffer, 
Eds., Oxford, Newnes, (2004), pp. 1-10. 

[8] P. Girdhar and C. Scheffer, Predictive maintenance techniques: 
Part 2 vibration basics, in Practical Machinery Vibration Analysis 
and Predictive Maintenance, P. Girdhar and C. Scheffer, Eds., 
Oxford, Newnes, (2004), pp. 11-28. 

[9] M. Caciotta, V. Cerqua, F. Leccese, S. Giarnetti, E. De Francesco, 
E. De Francesco, N. Scaldarella, A first study on prognostic 
system for electric engines based on Envelope Analysis, In IEEE 
Metrology for Aerospace (2014), Benevento, Italy, 29-30 May 
2014, pp. 362-366.  
DOI: 10.1109/MetroAeroSpace.2014.6865950  

[10] T. Van Tung, Y. Bo-Suk, Machine fault diagnosis and prognosis: 
The state of the art, International Journal of Fluid Machinery and 
Systems 2.1, (2009), pp. 61-71.  
DOI: 10.5293/IJFMS.2009.2.1.061  

[11] Li Zhe, Yi Wang, Ke-Sheng Wang, Intelligent predictive 
maintenance for fault diagnosis and prognosis in machine centers: 
Industry 4.0 scenario, Advances in Manufacturing 5.4, (2017), pp. 
377-387.  
DOI: 10.1007/s40436-017-0203-8  

[12] E. Petritoli, F. Leccese, G. Schirripa Spagnolo, New reliability for 
industry 4.0: A caste study in COTS-based equipment, IEEE 
International Workshop on Metrology for Industry 4.0 & IoT 
(MetroInd4.0&IoT), Rome, Italy, 7-9 June 2021, pp. 27-31. 
DOI: 10.1109/MetroInd4.0IoT51437.2021.9488555  

[13] E. Quatrini, F. Costantino, G. Di Gravio, R. Patriarca, Condition-
Based Maintenance - An Extensive Literature Review, Machines, 
(2020), 8, 31, pp. 1-28.   
DOI: 10.3390/machines8020031  

[14] A. K. S. Jardine, D. Lin, D. Banjevic, A review on machinery 
diagnostics and prognostics implementing condition-based 
maintenance, Mechanical Systems and Signal Processing, vol. 20, 
(2006), pp. 1483-1510.  
DOI: 10.1016/j.ymssp.2005.09.012  

[15] Caldepa website. Online [Accessed 03 November 2021]   
https://pump-selector.calpeda.com/pump/23  

[16] F. Fornarelli, A. Lippolis, P. Oresta, A. Posa, Computational 
Investigation of a pressure compensated vane pump, Energy 
Procedia, Volume 148, 73rd Conference of the Italian Thermal 
Machines Engineering Association, Pisa, Italy, 12 September 2018, 
pp 194-201.  
DOI: 10.1016/j.egypro.2018.08.068  

 

 

https://doi.org/10.1016/j.measurement.2019.107260
https://doi.org/10.1108/BPMJ-04-2017-0088
https://doi.org/10.1016/j.compchemeng.2017.10.037
https://doi.org/10.1016/j.measurement.2018.05.041
https://doi.org/10.3390/s19173781
https://doi.org/10.1109/MetroAeroSpace.2014.6865950
http://dx.doi.org/10.5293/IJFMS.2009.2.1.061
https://doi.org/10.1007/s40436-017-0203-8
https://doi.org/10.1109/MetroInd4.0IoT51437.2021.9488555
https://doi.org/10.3390/machines8020031
https://doi.org/10.1016/j.ymssp.2005.09.012
https://pump-selector.calpeda.com/pump/23
https://doi.org/10.1016/j.egypro.2018.08.068