Analysis and prediction of leak detection in the low-pressure heat treatment of metal equipment

Abstract

The low-pressure heat treatment of metals enables the continuous improvement of the mechanical and plastic properties of products, such as hardness, abrasion resistance, etc. A significant problem related to the operation of vacuum furnaces for heat treatment is that they become unsealed during operation, resulting from the degradation of seals or the thermal expansion of the construction materials. Therefore, research was undertaken to develop a prediction model for detecting leaks in vacuum furnaces, the use of which will reduce the risk of degradation in the charge being processed. Unique experimental studies were carried out to detect leakages in a vacuum pit furnace, simulated using the ENV 116 reference slot. As a consequence, a prediction model for the detection of leaks in vacuum furnaces- which are used in the heat treatment of metals- was designed, using an artificial neural network. (93% for MLP 15-10-1) was developed. The model was implemented in a predictive maintenance system, in a real production company, as an element in the monitoring of the operation of vacuum furnaces.

Full text

Ek s p l o a t a c j a i Ni E z a w o d N o s c – Ma i N t E N a N c E a N d REl ia b i li t y Vo l . 24, No. 4, 2022
719
Eksploatacja i Niezawodnosc – Maintenance and Reliability
Volume 24 (2022), Issue 4
journal homepage: http://www.ein.org.pl
Indexed by:
(*) Corresponding author.
E-mail addresses:
The low-pressure heat treatment of metals enables the continuous improvement of the me-
chanical and plastic properties of products, such as hardness, abrasion resistance, etc. A
signicant problem related to the operation of vacuum furnaces for heat treatment is that
they become unsealed during operation, resulting from the degradation of seals or the ther-
mal expansion of the construction materials. Therefore, research was undertaken to develop
a prediction model for detecting leaks in vacuum furnaces, the use of which will reduce the
risk of degradation in the charge being processed. Unique experimental studies were carried
out to detect leakages in a vacuum pit furnace, simulated using the ENV 116 reference slot.
As a consequence, a prediction model for the detection of leaks in vacuum furnaces- which
are used in the heat treatment of metals- was designed, using an articial neural network.
(93% for MLP 15-10-1) was developed. The model was implemented in a predictive mainte-
nance system, in a real production company, as an element in the monitoring of the operation
of vacuum furnaces.
Highlights Abstract
The infiltration method is not effective in predict-•
ing furnace leaks.
The prediction of furnace leaks is possible with •
the help of a Lambda probe.
Artificial neural networks are useful in predicting •
furnace leaks.
The predictive furnace leak detection model can •
be used on-line.
Analysis and prediction of leak detection in the low-pressure
heat treatment of metal equipment
Sławomir Kłos a, Justyna Patalas-Maliszewska a, Michal Bazel b
a University of Zielona Góra, Institute of Mechanical Engineering, ul. prof. Z. Szafrana 4, 65-516 Zielona Góra, Polska
b Seco/Warwick S.A. , ul. Sobieskiego 8, 66-200 Świebodzin, Poland
Kłos S, Patalas-Maliszewska J, Bazel M. Analysis and prediction of leak detection in the low-pressure heat treatment of metal equipment.
Eksploatacja i Niezawodnosc – Maintenance and Reliability 2022; 24 (4): 719–727, http://doi.org/10.17531/ein.2022.4.12
Article citation info:
vacuum heat treatment, vacuum furnace leaks, predictive maintenance, articial neural
networks.
Keywords
This is an open access article under the CC BY license
(https://creativecommons.org/licenses/by/4.0/)
S. Kłos (ORCID: 0000-0001-7110-9052): s.klos@iim.uz.zgora.pl, J. Patalas-Maliszewska (ORCID: 0000-0003-2439-2865):
j.patalas-maliszewska@iim.uz.zgora.pl, M. Bazel (ORCID: 0000-0002-1460-5110): michal.bazel@secowarwick.com
1. Introduction
Metal heat treatment processes are a very important element of the
manufacturing processes of products in industries such as the energy,
automotive, aviation and mining industries etc. The process of hard-
ening, carburising and/or nitriding is often the last process in the pro-
duction cycle and allows the reproducible properties of the processed
details, such as the surface hardness, core hardness or thickness of
the diffusion layer to be achieved. The key to obtaining the assumed
properties of the processed details is to maintain identical conditions
for the implementation of heat treatment processes in the process
chamber of the furnace, such as operating pressure, heating time, tem-
perature, amount of technical gas etc. Even a slight change of param-
eters inside the vacuum furnace, during the heat treatment process,
may result in irreversible damage to the charge and/or to the furnace.
Both in the energy and mining industries, due to large dimensions
and high production costs, the value of the charge to the furnace may
exceed the value of hundreds of thousands of euros. The relevance of
the problem may be proved by the scope of application of the details
processed, for example in control systems, gears, heat exchangers in
aeroplanes and cars, drive shafts for wind turbines, gears with large
modules, elements of drill bits and cutters in mining machines, drive
transmission systems, dies, surgical instruments etc.
Low-pressure carbonisation processes are usually carried out
within temperatures ranging from 950°C to 1020°C at a pressure of
approximately 5x100 mbar. In the tests conducted, the measured pres-
sure values are given in millibars, where 1mbar = 100 Pa due to the
use of this unit of measurement in economic practice in the field of
vacuum technology. In low-pressure carburising processes, it is nec-
essary to maintain the stability of the process, which guarantees ob-
taining a precise and repeatable thickness of the carburised surface
layer within the specified tolerance range. Possible leaks may result
in a change in the parameters of the heat treatment process as well as
damage to the charge, as a result of failing to meet the technological
requirements expected.
Predictive maintenance is an important component of the Industry
4.0 Concept, especially when using production resources where fail-
ures do not occur suddenly and the risk of their occurrence increases
over time. An example of such a progressive failure may be a leak
in a vacuum furnace. Due to changes in the temperature of the con-

Ek s p l o a t a c j a i NiEzawodNosc – Ma i N t E N a N c E a N d RE l ia b i li t y Vo l . 24, No. 4, 2022
720
struction of the furnace chamber and the wear of some elements, such
as the seal in the furnace cover during operation, the furnace may
become unsealed. Since modern vacuum furnaces are equipped with
highly efficient. cascade pumping systems consisting of a mechanical
pump and a Roots’ pump, pressure build-up in the vacuum furnace
due to leaks is often virtually unnoticeable. However, as a result of
leaks, the composition of the atmosphere inside the furnace chamber
changes, on account of a higher oxygen content, which may lead to
damage to the charge or failure to achieve the assumed parameters in
the carburising process and, as a result, not achieving all technological
requirements.
The implementation of predictive maintenance methods into indus-
trial practice is one of the key assumptions of the Industry 4.0 Concept
[13][15]. The progressive automation and digitisation of production
processes requires the implementation of systems that will enable
states of emergency to be diagnosed in advance. This approach makes
it possible to:
reduce losses associated with damaged parts as a result of fail-•
ures,
reduce unplanned downtime,•
reduce the maintenance costs incurred, regarding specialised, •
technical personnel, such as automation engineers and mechan-
ics.
Systems for monitoring the parameters of complex technical sys-
tems are particularly important where these are situated in remote re-
gions of the world and where access to qualified engineering staff is
limited. Currently, data on monitoring the operational parameters of
devices is stored in databases of ERP or MES systems. The Industry
4.0 Concept assumes real-time monitoring and recording of machine
operating parameters and stores them in the form of big data sets in
the Cloud. Multi-criteria data analysis with the use of artificial intel-
ligence methods will enable the construction of efficient algorithms
that will allow failure conditions to be predicted and thus, effectively
prevented due to the inspection and repair of production resources.
Research related to the development of the structure of a predictive
maintenance system, based on the monitoring of real-time data within
the reference model of industry architecture (RAMI 4.0) and DSR
(Design Science Research) to reduce costs and operations, were led
by Sahba et al. [24]. Predictive maintenance models, based on mathe-
matical programming and deep learning allow the technical condition
of individual elements of the production system to be predicted [12].
Machine learning methods based on machine performance analysis
and monitoring of production environment variables are used in many
of the studies carried out, in order to predict the emergency conditions
of production systems [6][5][17][27][29]. A very important aspect of
the implementation of predictive maintenance into industrial practice,
is the analysis of big data sets, based on computationally efficient
algorithms [31]. Another approach involves the construction of math-
ematical models to calculate maintenance rates for any schedule of
time up to failure [20]. An important issue regarding maintenance
management, based on prediction methods, is an approach based on
multi-criteria decision making, integrated with the traditional method
of analysing failure mode, effects, and criticality (FMECA) [1].
Low-pressure carburising is a long-lasting and energy-consuming
heat treatment process, especially considering its requirement to ob-
tain thick carburising layers, i.e., over 2.00 mm. One of the important
parameters that influence the stability and repeatability of the process
is the low oxygen content in the vacuum furnace chamber. The most
common cause of the deterioration of this parameter is a leak in the
furnace, which results in air entering the furnace chamber. The is-
sue of leakage, in devices operating in high vacuum conditions is an
important area of research and the subject of scientific publications
[4] [9]. There are many methods for testing the leakage of vacuum
devices, such as the pressure increase or decrease test, the leak test
with gas-sensitive vacuum gauges, the test by immersion in various
formulations or spraying with foam, the Krypton 85 test, the high fre-
quency vacuum test, test with chemical reactions and dye penetra-
tion [21]. An important method for testing the tightness of techni-
cal objects is the tracer gas method. The most popular tracing gases
are helium, ammonia, hydrogen, nitrogen and semi-precious gases.
A commonly used method of leak detection is the helium method,
so-called [13] [28]. Helium is very well suited to the leak detection
of vacuum devices because it is a non-toxic, non-flammable gas and
does not form explosive mixtures with other gases. It is also neutral
to the environment and does not enter into chemical reactions with
other substances. The use of helium is not limited by the range of
temperatures or pressures as it is a very temperature stable gas, which
facilitates the testing of objects in environments with extremely high
or low temperatures. Helium is the gas- right after hydrogen- with the
smallest unit particle which enables penetration of micro-fractures in
materials, therefore helium tests are very accurate. The relatively low
price of helium compared to other noble gases should also be noted.
The method itself is relatively simple and consists in connecting a
so-called helium detector to the object being studied, while the opera-
tor dispenses a small portion of helium to any possible leak. The test
object must reach a high, negative pressure. The method requires a
detector with high accuracy, viz., the detection of leaks at the level of
5x10-12 mbar l/s. When using a helium detector, it is also possible to
approximate the location of the leak. Another method for detecting
leaks in vacuum furnaces is the so-called infiltration method, which
consists in heating the furnace to a certain temperature at a specific
negative pressure - usually as low a pressure as possible and then, as
it slowly cools, calculate the value of the infiltration, i.e., the decrease
in the value of the vacuum level, within a specific timeframe. The
infiltration testing process allows leaks to be detected, however, it is
long-lasting (several hours to several dozen hours) and does not allow
the leak to be located. In general, the leak test should be preceded by
the identification of possible leaks using the helium method.
Based on an analysis of the results of the literature on the subject,
the infiltration method was selected for detecting leaks in vacuum fur-
naces with experimental tests being carried out to analyse leakages in
pit furnaces for specific settings of the ENV 116 standard slot.
The use of various tools such as the Markov Process, Bays Net-
works, artificial neural networks or simulation methods, based on the
Monte Carlo method for predictive maintenance purposes, has been
the subject of publications by many authors [7][25]. Predictive main-
tenance methods use the Markov Process, Bays Networks, artificial
neural networks and simulation methods based, for example, on the
Monte Carlo method.
Many publications cover the use of machine learning methods for
predictive maintenance, based on an analysis of machine perform-
ance and the variables of the production environment [5][17][27].
Part of the research concerns the acquisition and storage of data in
the Cloud and the construction of a system platform for predictive
maintenance [26]. Artificial neural networks are used, among other
things, in order to analyse and classify data, both current and forecast,
with the help of in-depth learning techniques and will enable images
to be both recognised and reproduced. In the maintenance department,
models are expected for detecting and forecasting future failures, in
real time. [3] Based on an analysis of the literature, the most popular
data-driven decision methods include the Support Vector Machine
(SVM), Principal Component Analysis (PCA), Linear Discriminant
Analysis (LDA), Random Forest, K-Nearest Neighbours and the Hid-
den Markov Model [22]. Currently, many methods, techniques and
procedures, using intelligent production systems for maintenance
workers, are based on deep learning techniques [17][32], however,
there are still good examples of the use of artificial neural networks in
maintenance, in the monitoring of tool wear, in the diagnosis of vibra-
tion in machining systems, in the thermal analysis of machines, in the
analysis of other malfunctions affecting production, as well as in the
diagnostics of finished products [9].

Ek s p l o a t a c j a i Ni E z a w o d N o s c – Ma i N t E N a N c E a N d REl ia b i li t y Vo l . 24, No. 4, 2022
721
In the literature on the subject analysed, no model was found that
could be used to predict the detection of leaks in devices used in the
low-pressure heat treatment of metals. Therefore, research work was
undertaken to design an effective leak detection model in vacuum
furnaces, based on data obtained from the study of a working pit fur-
nace.
The article presents the stages of a research experiment concerning
the analysis of leaks with the use of a standard slot for various pres-
sure and temperature parameters. Based on an analysis of the results
of the experimental studies, a model for detecting leaks predictively,
using artificial neural networks, was designed.
2. Research methods
2.1. Experimental work- testing furnace tightness- based on
an analysis of the inltration
The infiltration test is an effective, but relatively time-consuming
method for obtaining knowledge about leakage in vacuum furnaces.
The procedure for testing the tightness of the pit furnace, using the
infiltration method, is shown in Figure 1.
Fig. 1. The procedure for testing a leaking pit furnace, using the infiltration
method
The longest stage in the infiltration test procedure here presented,
is the cooling down of the furnace to a temperature of 50°C. Depend-
ing on the design of the furnace, that is, regarding the thickness of the
insulation and the size of the heating chamber etc., the furnace cool-
ing down process may take from several hours to several dozen hours.
The control calculations for the infiltration test are performed, based
on formula (1) shown below:
273
0, 5 1 273
1000 60
b
eb
e
T
PP
T
NV
t


+
⋅⋅+ −



+


=⋅⋅ ⋅
(1)
where
N - (leak rate) [mbar* l/s]
V – volume of the furnace heating chamber [l]
Pb – initial pressure [mbar],
Pe – final pressure [mbar],
Tb – initial temperature [°C],
Te – final temperature [°C],
t – time [min].
The criterion for the tightness of the device, vis-à-vis the infiltra-
tion test, was set at 5,0 10-3 [mbar·l/s].
Table 1 shows examples of the results of the pit furnace infiltration
test wherein a leak was detected.
The results presented in Table 1 differ many times from the cri-
terion adopted, therefore the device tested shows increased leakage.
After a review of the structure of the device and replacement of the
valves, the infiltration test was carried out again. The test was per-
formed with a slightly higher final temperature. The results are pre-
sented in Table 2.
As can be seen from the data presented in Table 2, the minimum in-
dicator, adopted and established for the infiltration test, was exceeded
over 100 times, thus indicating a leaking device. After re-analysis of
tightness with the use of a helium detector and the sealing of the struc-
tural elements having been replaced, the infiltration test was carried
out again; this gave satisfactory results. Table 3 presents the results of
an infiltration test for a furnace which meets the criteria for leakages.
According to the analysis of the infiltration tests carried out, the
method is not only effective, but is also long-lasting and may require
several repetitions in the case of unsatisfactory results. In production
conditions, it is also quite expensive, because it requires shutting down
the heat treatment device for several hours. These devices are heavily
loaded with orders in most enterprises and constitute bottlenecks in
the manufacturing process. It is very important that, before starting
an infiltration test, a helium test is carried out, in order to eliminate
possible leakage in the device. Due to the specific features of the infil-
tration test, i.e., its long duration, furnace shutdown and high energy
costs, this method cannot be used for a maintenance predictive sys-
tem. Therefore, research work was undertaken to design a pit furnace
leakage prediction model, using artificial neural networks.
2.2. Data collection and analysis
The Lambda probe is a sensor commonly used in the automotive
industry to analyse the oxygen content in exhaust gas. The voltage
generated by the sensor is lower than the higher oxygen content in
exhaust gas, while the small amount of oxygen ions in the exhaust
gas generates high voltage. To test for leakages in a pit furnace, the
standard EVN116 slot (gas dosing valve) was used, which enables
Table 1. Examples of the results of the first infiltration test for a leaky vacuum furnace.
V t PbPeTbTeN
[l] [s] [mbar] [mbar] [°C] [°C] 10-3 [mbar·l/s]
1308 3600 1.40 4.06 42 42 967.83
1308 7200 1.40 6.87 42 41 995.00
1308 10800 1.40 9.79 42 41 1018.07
1308 14400 1.40 12.79 42 40 1037.80
1308 18000 1.40 25.12 42 39 1082.99
1308 21600 1.40 49.51 42 36 1102,87
1308 25200 1.40 74.52 42 34 1121.14

Ek s p l o a t a c j a i NiEzawodNosc – Ma i N t E N a N c E a N d RE l ia b i li t y Vo l . 24, No. 4, 2022
722
the leak rate to be determined manually. Figure 2 shows the test site
(Figure 2).
As can be seen from the characteristics presented, setting the slot
to 100 results in a horizontal leak at a level of 0.5·10-3 mbar·l/s, but
when the slot is set to 200, the leak is at a level of 0.8·10-3 mbar * l/s.
The test programme included an analysis of the characteristics of
changes in the Lambda probe indications, depending on the size of
the reference gap for different temperatures in the heating of the pit
furnace. The following activities were planned for the purpose of con-
ducting research experiments:
setting a specific leakage for the reference slot, without opening •
the slot,
setting the maximum heating temperature T• max,
creating a vacuum in the furnace at the P• min level,
heating the furnace to temperature T• max with the pump run-
ning,
keeping T• max for 15 min with the pump running,
closing the pump shut-off valve and turning the pump off,•
opening reference slot t• 1,
closing reference slot t• 2,
turning the furnace off. •
The tests were carried out at different heating temperatures:
500°C (loading the charge), 800°C (near temperature for harden-
ing) and 1000°C (vacuum carburisation temperature). Figure 3
shows the temperature diagrams and the pressure and Lambda
probe readings during heating to a temperature of 500°C, with
the vacuum pump working, in order to determine the refer-
ence characteristics of the Lambda probe for a tight furnace.
The furnace was started at 7:20 a.m. at a temperature of 28°C;
the target temperature of 500°C was reached after 60 minutes. The
heating ramp was approximately 8°C/min. The vacuum in the furnace
chamber was 7·10-3 mbar. As can be seen from the diagram of the
Lambda probe readings, taken during pumping out the furnace, the
amount of oxygen ions in the furnace atmosphere decreased, which
resulted in an increase in the voltage to approximately 880 mV.
Experiments at the test site (Figure 2) were then carried out. They
simulated furnace leaks, with the standard slots set to 100, 150, 200
and 250, respectively, according to the characteristics of the EVN 116
Table 2. Examples of the results of the second infiltration test for a leaky vacuum furnace
V t PbPeTbTeN
[l] [s] [mbar] [mbar] [oC] [oC] 10-3 [mbar·l/s]
1308 3600 1.46 2.85 76 75 504.47
1308 7200 1.46 4.35 75 74 525.88
1308 10800 1.46 5.90 74 72 538.53
1308 14400 1.46 7.42 72 71 542.02
1308 18000 1.46 8.91 71 70 542.51
1308 21600 1.46 10.36 70 69 539.76
1308 25200 1.46 11.80 69 68 537.59
1308 28800 1.46 13.17 68 67 532.93
1308 57600 1.46 23.01 67 58 495.93
1308 86400 1.46 31.70 58 52 462.59
Table 3. The results of the infiltration test for a furnace with a tight vacuum
V t PbPeTbTeN
[l] [s] [mbar] [mbar] [oC] [oC] 10-3 [mbar·l/s]
1308 3600 0.238696 0.244000 43 43 2.00
1308 7200 0.238696 0.249304 43 42 1.98
1308 10800 0.238696 0.255091 43 42 2.04
1308 14400 0.238696 0.259913 43 42 1.98
1308 18000 0.238696 0.265700 43 41 2.02
1308 21600 0.238696 0.271004 43 41 2.01
1308 25200 0.238696 0.276308 43 41 2.01
1308 28800 0.238696 0.282095 43 40 2.03
1308 57600 0.238696 0.287399 43 40 2.03
1308 86400 0.238696 0.292704 43 40 2.02
Fig. 2. Test stand for testing vacuum furnace leakage

Ek s p l o a t a c j a i Ni E z a w o d N o s c – Ma i N t E N a N c E a N d REl ia b i li t y Vo l . 24, No. 4, 2022
723
standard slot in which the numerical codes of the manual setting of the
standard slot correspond to the specified leak (the so-called “digital
display”, Figure 4), in the leakage range from approximately 10-3 do
10-2 mbar·l/s.
Figure 4. Characteristics of EVN 116 standard slot [33]
Changes in the furnace pressure for the individual settings of the
reference slot and simulated voltage, as well as for the Lambda sensor
readings are shown in Figure 5. As can be seen from the diagram pre-
sented, differences in the Lambda sensor readings for different sizes
of the device leak are already visible 30 minutes after the opening of
the standard slot. Table 4 shows the values of pressure and indications
of the Lambda probe, which were recorded at equal intervals every 30
minutes, after opening the reference slot. Based on the data presented
in Table 4, it is possible to diagnose any leakage in the device after 30
minutes and determine its size.
Analysis of the data obtained as a result of the
conducted experimental tests (Table 4) shows
that for increasingly larger leaks in the ENV 116
slot, the Lambda probe generates a correspond-
ingly lower voltage.
The data received in the form of 1600
measurements concerning parameter values:
y - Lambda probe voltage, x1 - temperature in
the Furnace chamber, x2 - working pressure in
the furnace chamber (values in bar), x3 - hous-
ing temperature, x4 – water flow in the cooling
system, x5 - insulation temperature 1, x6 - in-
sulation temperature 2, x7 – water flow in the
cooling system of current passages, x8 - insula-
tion temperature 3, x9 - water temperature at the
inlet of the cooling system, x10 - gas pressure
of the pneumatic system, X11 - temperature of
the current bushing, X12 - heater current 1, x13
- heater current 2, X14 - heater current 3, x15 -
partial pressure in the furnace chamber (values
in millibars) constituted the basis for building a
leak detection prediction model in vacuum fur-
naces using artificial neural networks.
3. Test results
The basic element of artificial neural net-
works are neurons, each of which is an autono-
mous processing unit. Each neuron carries out
its own simple calculations and the structure,
consisting of a large number of neurons, facili-
tates the multiplication potential of these calcu-
lations. The task of the neurons is to operate on
the input data and present the results computed
by the function activation. The neuron also de-
scribes a bias, which is an element that models
the threshold above which the neuron sends an impulse, that is, an
adjustment of the activation threshold value. The artificial neural net-
work thus designed has parameters (weights) that must be assigned
initial values. The role of the activation function is to determine the
degree of excitation of the neuron, on the basis of the values reach-
ing it. Based on the function used, the output value of the neurons is
calculated.
The model was built using an artificial neural network due to its
utility in both reactive and preventive maintenance as well as predic-
tive maintenance. The model was designed on the basis of real data
obtained from the operation of a vacuum furnace. An artificial neural
network was chosen because using it facilitates, among other things,
data classification and identification, the forecasting of wear in ma-
chine elements [9], the forecasting of the abrasive wear of cutting
tool blades and the monitoring of machines and devices in action. The
ANN type of unidirectional MLP multilayer neural network was se-
lected (Multi-layered Perceptron) [23]. The network consists of neu-
Table 4. Lambda probe indications [mV] for various ENV 116 slot leaks
digital display
time 100 150 200 250
00:00 835 826 818 813
00:30 840 831 822 809
01:00 844 835 824 806
01:30 849 839 826 802
02:00 852 842 828 799
02:30 855 842 828 795
b)
a)
c)
Fig. 3. Changes in a) temperature in the furnace chamber, b) pressure, c) voltage of the Lambda probe for
a tight furnace

Ek s p l o a t a c j a i NiEzawodNosc – Ma i N t E N a N c E a N d RE l ia b i li t y Vo l . 24, No. 4, 2022
724
rons arranged in layers. Each of the neurons computes the weighted
sum of its inputs; the excitation level thus determined becomes an
argument for the transition function (activation function) that com-
putes the output value of the neuron. For a unidirectional multilayer
network, determining the appropriate number of hidden layers and the
number of neurons in individual layers is not a simple issue. [2].
In order to build the model, an artificial neural network with a lo-
gistic activation function, a sigmoid unipolar function (formula 2) was
used, due to the form of data on the Lambda probe indication for a
tight furnace (Figure 3c).
( ) x
1
x
1e
−
=+
(2)
where:
x
– is the input value of the activation function,
e
- Euler’s
number. The activation function is especially useful in artificial neu-
ral networks with back propagation. The function maps the interval
( )
,∞∞−
to
( )
0 , 1
[19].
The weighting factors were determined in the process of training
neurons through supervised learning. To generate a neural network
Statistica, ver. 13.3 was used. The neural network model used in the
study is shown in Figure 6.
where:
w1…w15 - weights
Y -Lambda probe measurement value
Table 5 compares the MLP network, with the
activation logistics function, in terms of the net-
work quality achieved, for training, testing and
validation, respectively and the error function.
The best model was the MLP network with the
structure 15‒10‒1, where 15-10-1 refers to the
number of inputs (15), the number of neurons in
the hidden layer (10) and the number of output
networks (1).
The leak detection prediction model in vac-
uum furnaces was then verified with the use of
ANN MLP 15-10-1 for the data presented in
Appendix no. 1. Tests were carried out while the
vacuum furnace was working and the actual re-
sults were compared with the forecast obtained
(Table 6).
In Figure 7 a prediction model for the detec-
tion of leaks, in devices for the heat treatment of
metals, has been presented.
4. Discussion
The aim of the study was to formulate a pre-
diction model for detecting leaks in vacuum fur-
naces using artificial neural networks. The re-
search required experiments to be conducted on
a vacuum furnace in actual operation, followed
by work related to the acquisition and analysis
of data and the use of an appropriate ANN struc-
ture. The research was conducted in the R&D
Department of SECO/WARWICK S.A. using a
pit vacuum furnace. Table 7 presents the main characteristics indicat-
ing the originality of the research results obtained and the contribution
made to the research on leakage in vacuum furnaces. In particular,
the approaches to the detection of furnace leakages were described,
taking into account: (1) the methods used, (2) effectiveness, (3) veri-
fication in business practice. According to the present authors’ knowl-
edge, there is no approach, in existing studies, integrating the results
of experimental work on the detection of leaks in a pit furnace using
the ENV 116 reference slot with the use of artificial neural networks.
Technical devices that work in high vacuum conditions and are
subjected to large temperature changes, become unsealed after a cer-
tain period of time. This can be due to various reasons, such as seal
wear, damaged valves, leaks due to expansion and contraction of the
metal parts of the furnace etc. Leaks can occur suddenly, as a result of
a fault, or gradually, as a result of the normal operation of equipment.
It is possible to prevent gradual leakage of the heat treatment device
by continuously monitoring selected parameters. As currently used,
pumping systems for obtaining a high vacuum in furnace chambers
are very efficient, therefore the occurrence of even a large leak may
not be registered by the pressure sensors. Due to the economics of heat
treatment processes and for the avoidance of damage to the charge,
it is important that the prediction of potential leakage occurs before
actual treatment is required, such as carburising or quenching,
so that the charge is not degraded. This can be achieved by us-
ing artificial intelligence methods. The study demonstrated the
usefulness of an artificial neural network and its effectiveness in
supporting the prognosis of furnace leakage (93% for MLP 15-
10-1). The model designed for the detection of leaks in vacuum
furnaces (Figure 6) can be integrated in systems supporting the
operation of the maintenance department. Such adaptation can
even be done online, if all the features / input values taken into
account can be changed and controlled online.
Fig. 5. Changes in the readings of the pressure sensor and the Lambda sensor for different levels
Fig. 6. MLP network structure for leak detection prediction in vacuum furnaces.
b)
a)

Ek s p l o a t a c j a i Ni E z a w o d N o s c – Ma i N t E N a N c E a N d REl ia b i li t y Vo l . 24, No. 4, 2022
725
5. Summary and conclusions
The low-pressure heat treatment of metals is most often carried
out in the last phase of a production cycle. Inadequate heat treatment
parameters, such as too much oxygen in the furnace chamber or too
high a pressure, caused by leaks in the furnace, may lead to produc-
tion shortages or the production of products with lower strength and
reliability. Leakage in heat treatment equipment cannot be completely
eliminated as it is process-specific, vis-à-vis the thermal expansion of
metals, damage to the seal resulting from the operation of the furnace
etc.
The article proposes a model for predictively detecting leaks
in vacuum furnaces with the use of artificial neural networks.
As a result of the experimental tests conducted, consisting in
simulating the leakage, the Lambda sensor readings were deter-
mined for various settings of the standard slot size. The predic-
tion model formulated for the detection of leaks in vacuum fur-
naces, as used in the heat treatment of metals using an artificial
neural network (93% for MLP 15-10-1) will allow leakage in
the heat treatment equipment of metals- already in the heating
phase- to be detected before the start of the carbonisation or
hardening processes, thus protecting the load against damage.
As part of further research and development work on the pre-
dictive maintenance of metal heat treatment devices, work will
be carried out on detecting the conditions preceding burnout of
the furnace heating elements, resulting from an increase in the
level of carbon deposits on the heating elements and current
passages and from low-pressure carburising processes.
Authors’ contributions
SK: developing the concept of the article, planning research experi-
ments, conducting research experiments, developing a predictive
model for a vacuum furnace leak detection system, writing the content
of the article.
JPM: developing the concept of the article, conducting research ex-
periments, developing a predictive model of a vacuum furnace leak
detection system, writing the content of the article.
MB: carrying out research experiments.
Acknowledgements
The article was created as a result of the project co-financed by the
European Union and the National Centre for Research and Develop-
ment POIR.04.01.02-00-0064/17.
Table 5. Artificial neural networks for leak detection in a vacuum furnace
Network name Quality
(learning)
Quality (test-
ing)
Quality (vali-
dation)
Error
(learning)
Error (test-
ing)
Error (vali-
dation)
Learning
algorithm
Error
function
MLP 15-7-1 0.980799 0.751166 0.828486 177.310 158.8133 86.2470 BFGS 40 SOS
MLP 15-10-1 0.983770 0.751710 0.933961 184.058 245.6714 174.4568 BFGS 42 SOS
MLP 15-12-1 0.563121 0.462774 0.514705 3191.967 399.8066 433.7868 BFGS 4 SOS
MLP 15-14-1 0.966649 0.571415 0.575219 308.369 256.8963 249.1537 BFGS 48 SOS
MLP 15-13-1 0.973078 0.568790 0.681975 249.102 304.1500 156.2498 BFGS 41 SOS
Table 6. Comparison of actual values with the values forecast for leakages in a vacuum furnace
Time Lambda probe voltage - actual value
[mV]
Lambda probe voltage – forecast
value [mV] Error
15:45:00 727,2063 793,5139 -66,3076
16:00:00 743,9453 795,6114 -51,6661
16:15:00 755,042 793,7966 -38,7546
16:30:00 763,1294 792,5906 -29,4612
16:45:00 769,8062 793,7554 -23,9492
17:00:00 775,2605 792,3767 -17,1162
17:15:00 779,8684 794,0572 -14,1888
17:30:00 783,9121 794,9163 -11,0042
Fig. 7. Prediction model for detecting leaks in vacuum furnaces

Ek s p l o a t a c j a i NiEzawodNosc – Ma i N t E N a N c E a N d RE l ia b i li t y Vo l . 24, No. 4, 2022
726
References
Ahmed U, Carpitella S, Certa A. An integrated methodological approach for optimising complex systems subjected to predictive maintenance. 1.
Reliability Engineering and System Safety 2021; 216: 108022, https://doi.org/10.1016/j.ress.2021.108022
Bishop C. Training with noise is equivalent to Tikhomov regularisation, Neural Computation 1995, 7 (1): 108-116. 2. https://doi.org/10.1162/
neco.1995.7.1.108
Bousdekis A, Lepenioti K, Apostolou D, Mentzas G. Decision Making in Predictive Maintenance: Literature Review and Research Agenda 3.
for Industry 4.0. IFAC-Papers On Line 2019, 52 (13): 607-612. https://doi.org/10.1016/j.ifacol.2019.11.226
Calcatelli A, Bergoglio M, Mari D. Leak detection, calibrations and reference ows: Practical example. Vacuum 2007; 81(11–12): 1538–4.
1544, https://doi.org/10.1016/j.vacuum.2007.04.019.
Cline B, Niculescu RS, Human D, Deckel B. Predictive maintenance applications for machine learning. Proceedings - Annual Reliability 5.
and Maintainability Symposium 2017. 1-7, https://doi:10.1109/RAM.2017.7889679.
Dalzochio J, Kunst R, Pignaton E, Binotto A, Sanyal S, Favilla J, Barbosa J. Machine learning and reasoning for predictive maintenance in 6.
Industry 4.0: Current status and challenges. Computers in Industry 2020; 123: 103298. https://doi.org/10.1016/j.compind.2020.103298.
Efthymiou K, Papakostas N, Mourtzis D, Chryssolouris G. On a predictive maintenance platform for production systems. Procedia CIRP 7.
2012; 3: 221-226, https://doi:10.1016/j.procir.2012.07.039.
Fradette R J, Jones W R. Vacuum Furnace Leaks and Detection Techniques; https://www.industrialheating.com/articles/95173-vacuum-8.
furnace-leaks-and-detection-techniques, 2019.
Gawlik J, Kiełbus A. Zastosowania metod sztucznej inteligencji w nadzorowaniu urządzeń technologicznych i jakości wyrobów. Praktyka 9.
zarządzania jakością w XXI wieku, 2012.
Gu B, Huang X. Investigation of leak detection method by means of measuring the pressure increment in vacuum. Vacuum 2006; 80(9): 10.
996–1002, https://doi.org/10.1016/j.vacuum.2006.01.005.
Haripriya M, Saravanan S, Rejul M. Iot Enabling of Vacuum Heat Treatment Chambers for Data Acquisition and Analytics. 3rd International 11.
Conference on Computing Methodologies and Communication (ICCMC) 2019; 18958316, https://doi.org/10.1109/ICCMC.2019.8819829
Hesabi H, Nourelfath M, Hajji A. A deep learning predictive model for selective maintenance optimisation. Reliability Engineering & 12.
System Safety 2021; 219: 108191, https://doi.org/10.1016/j.ress.2021.108191.
Li Z, Wang K, He Y. Industry 4.0 - Potentials for Predictive Maintenance. International Workshop of Advanced Manufacturing and Automation 13.
(IWAMA) 2016, hps://doi.org/10.2991/iwama-16.2016.8
Meng D, Sun L, Yan R, Shao R, Yu X, Li X, Zhang H, Zhao Y. Eects of cryopump on vacuum helium leak detection system. Vacuum 2017; 14.
143: 316–319. hps://doi.org/10.1016/j.vacuum.2017.06.036.
Mobley R K. An introduction to predictive maintenance. 2nd edition. Butterworth-Heinemann 2002. 15. https://doi.org/10.1016/B978-
075067531-4/50006-3
Oakes J, Lutz J. Furnace Atmosphere Controls in Heat Treating. Steel Heat Treating Technologies. ASM International 2014; 4B: 16. https://doi.
org/10.31399/asm.hb.v04b.a0005928
Paolanti M, Romeo L, Felicetti A, Mancini A, Frontoni E, Loncarski J. Machine Learning approach for Predictive Maintenance in Industry 17.
4.0. 14th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2018; 1-6, hps://
doi:10.1109/MESA.2018.8449150.
Patalas-Maliszewska J, Halikowski D. A Model for Generating Workplace Procedures Using a CNN-SVM Architecture. Symmetry 2019; 11: 18.
1-14. hps://doi.org/10.3390/sym11091151
Ponti M A, Ribeiro L S F, Nazare T S, Bui T, Collomosse J. Everything you wanted to know about deep learning for computer vision but 19.
were afraid to ask. 30th SIBGRAPI Conference on Graphics, Patterns and Images Tutorials (SIBGRAPI-T) 2017; 17-41, https://doi:10.1109/
SIBGRAPI-T.2017.12.
Raza A, Ulansky V. Modelling of Predictive Maintenance for a Periodically Inspected System. Procedia CIRP 2016; 59 (TESConf 2016): 20.
95–101, https://doi.org/10.1016/j.procir.2016.09.032.
Rottländer H, Umrath W, Voss G. Fundamentals of leak detection. Leybold GMBH (ed) Cat 2016:https://www.leyboldproducts.fr/media/21.
pdf/90/c7/87/Fundamentals_of_Leak_Detection_EN.pdf
Ronao C A, Cho S B. Human activity recognition using smartphone sensors with two-stage continuous hidden markov models. Natural 22.
Computation (ICNC), 10th International Conference on. IEEE 2014; 681-686, https://doi.org/10.1109/ICNC.2014.6975918.
Rumelhart D E, Hinton G E, Williams R J. Learning Internal Representations by Error Propagation in Parallel Distributed Processing. 23.
Explorations in the Microstructure of Cognition, Foundations: MIT Press, 1986; Vol. 1, Cambridge MA. https://doi.org/10.7551/
mitpress/5236.001.0001
Sahba R, Radfar R, Rajabzadeh Ghatari A, Pour Ebrahimi A. Development of Industry 4.0 predictive maintenance architecture for broadcasting 24.
chain. Advanced Engineering Informatics 2021; 49: 101324, https://doi.org/10.1016/j.aei.2021.101324.
Sakib N, Wuest T. Challenges and opportunities of condition-based predictive maintenance: a review. Procedia CIRP 2018; 78: 267–272, 25.
https://doi.org/10.1016/j.procir.2018.08.318
Schmidt B, Wang L. Cloud-enhanced predictive maintenance. Int J Adv Manuf Technol. 2018; 99: 5-13, 26. https://doi:10.1007/s00170-016-
8983-8
Susto G A, Schirru A, Pampuri S, McLoone S, Beghi A. Machine learning for predictive maintenance: A multiple classier approach. IEEE 27.
Trans Ind Informatics 2015; 11(3): 812-820 https://doi:10.1109/TII.2014.2349359.
Takeda H. Helium leak detection method using ambient temperature of canister top. Nuclear Engineering and Design 2019; 352: 110135. 28.
https://doi.org/10.1016/j.nucengdes.2019.05.031
Theissler A, Pérez-Velázquez J, Kettelgerdes M, Elger G. Predictive maintenance enabled by machine learning: Use cases and challenges in 29.
the automotive industry. Reliability Engineering and System Safety 2021; 215: 107864, https://doi.org/10.1016/j.ress.2021.107864.
Vlasov A I, Echeistov V V, Krivoshein A I, Shakhnov V A, Filin S S, Migalin V S. An information system of predictive maintenance analytical 30.
support of industrial equipment. Journal of Applied Engineering Science 2018; 16(4): 515–522. https://doi.org/10.5937/jaes16-18405
Wen Y, Fashiar Rahman M, Xu H, Tseng T L B. Recent advances and trends of predictive maintenance from data-driven machine prognostics 31.
perspective. Measurement: Journal of the International Measurement Confederation 2022; 187: 110276, https://doi.org/10.1016/j.

Ek s p l o a t a c j a i Ni E z a w o d N o s c – Ma i N t E N a N c E a N d REl ia b i li t y Vo l . 24, No. 4, 2022
727
measurement.2021.110276.
Wuest T, Weimer D, Irgens C, Klaus D T. Machine learning in manufacturing: advantages, challenges, and applications. Production & 32.
Manufacturing Research 2016; 4 (1): 23-45, https://doi.org/10.1080/21693277.2016.1192517.
Valve gas dosing, EVN 116. [http://www.pfeier-vacuum.com/productPdfs/PFI32031.en.pdf. EVN 116, Gas dosing valve with separate 33.
shut-o valve, manual].