Directions of Development of Diagnostic Methods of Vacuum Medium-Voltage Switchgear

Abstract

The development of power grid infrastructure and increasingly stringent environmental regulations have intensified work, carried out by researchers and electrical equipment manufacturers, to develop innovative gas-insulated, environmentally neutral devices. The emergence of new designs of circuit breakers and disconnectors, in which the resulting electric arc is extinguished in a vacuum environment, requires the development of appropriate techniques for diagnosing the chambers responsible for the dielectric parameters of the device. The following article presents an overview of the directions of development of diagnostic methods for medium-voltage vacuum switching equipment, which can potentially be used to develop a real-time pressure monitoring system that can be applied to vacuum switching equipment used in electrical infrastructure.

1. Introduction

Electricity is one of the basic utilities used in today’s world. It is essential in almost every area of our lives, from powering large manufacturing plants to protecting human health and life. The ever-increasing amount of electricity consumed by consumers and the increased demands placed on networks in terms of continuity and quality of energy supply have led to the development of energy infrastructures [1,2,3], which is constantly being improved to meet the challenges posed. Innovative technologies are being applied to modern networks and have prompted the development of appropriate control and measurement techniques to verify the condition of the power network in operation [4]. The development of the electricity network has led to an increase in the importance of gas-insulated apparatus, particularly the importance of vacuum interrupter technology (VI—Vacuum Interrupters), whose main aspect of suitability for operation is the pressure inside the extinguishing chamber. Given the paramount importance of the pressure level in a vacuum chamber, the power industry faced another challenge of developing a real-time pressure measurement system to enable chamber diagnostics and thus predict potential damage and thus power outages. To the best knowledge of the article’s authors, there is currently no technology solution for continuous pressure measurement designed for equipment operating in continuous mode. In connection with the work carried out, by the research group, on the development of the aforementioned system for monitoring the parameters of a vacuum chamber in on-line mode, the state of the art of selected methods of measuring pressure in vacuum equipment was analyzed and compared, which could potentially be used to develop a system for monitoring the pressure of vacuum quench chambers in real time.

2. Switching Devices in Medium-Voltage Networks

The VI technology, especially modern vacuum circuit breakers (VCBs—Vacuum Circuit Breakers) are of enormous importance in the power system, enabling switching processes to be carried out autonomously or with negligible dispatcher intervention. Vacuum circuit breaker technology is widely known and used in network infrastructure, mainly through the VD 4 series from ABB [5] and the SION series from SIEMENS [6]. Vacuum circuit-breakers practically dominate the switchgear in transformer substations due to their small size and excellent mechanical durability of up to 30,000 switching cycles, as well as the fact that no maintenance is required during normal operation.

Over the decades, the market for arc extinguishing agents has been dominated by vacuum, this was not always the case, as shown in Figure 1 below.

As can be seen from the diagram above, oil, air, sulfur hexafluoride (SF₆) and vacuum have played the role of insulating medium in switchgear over more than four decades [8].

In the 1980s, oil-insulated circuit breakers were the most trusted. The entire circuit breaker unit was immersed in a tank filled with oil, whose function was to provide insulation and an arc extinguishing mechanism. These were usually located outside [9]. However, the great days of oil-insulated circuit breakers are already long gone and now they have been virtually eliminated from industrial applications in favour of vacuum circuit breakers.

Pneumatic air circuit breakers [10,11,12] have been used extensively in various types of circuits, especially conditions. They are equipped with blast coils, applicable for arc extinguishing by magnetic blast, and blast tubes that operate by forcing air down the tube from each nozzle of the contact assembly. They were completely displaced from the market in favour of vacuum units more than two decades ago.

Another type of gas-insulated apparatus encountered in networks are arc extinguishing switches in compressed SF₆ [13,14]. In industrial environments where the use of oil circuit breakers was not possible due to fire hazards. The cool air created by convection enables them to operate at the correct temperature for the operating.

Sulfur hexafluoride and vacuum circuit breakers are similar in design. The main difference between these two apparatuses is the insulation, which is provided by different dielectric medium. The highly compressed gas is expelled from the arcs in the circuit breakers while the mechanism is operating and is then collected in a low-pressure container before being transported back to the high-pressure container for reuse. The effect of extinguishing the arc in SF₆ is an extremely harmful residue from arc combustion. The fact that SF₆ is a powerful greenhouse gas is also not insignificant. The pollutants released into the environment can have a very negative impact on the atmosphere. The technology used reduces the possibility of leakage from the circuit breaker during operation.

The most common adaptation of circuit breakers for more than four decades has been vacuum circuit-breakers [15,16]. This solution is nowadays widely used in various types of power systems. They are famous for their long service life and functionality and, above all, for their lack of negative environmental impact, which is extremely important in the context of an ongoing environmental policy. They provide a higher degree of reliability compared to air circuit breakers. Due to the fact that there is no ionizing agent and only the contact material is present, vacuum circuit breakers are characterized by low arc energy. The arc is extinguished already at a distance between contacts of 2–3 mm (typically 10 mm for MV networks). Near the point where the current passes zero, the discharge can no longer be maintained by the arc energy and the arc is extinguished.

As Figure 1 shows, for almost forty years there has been a constant increase in the amount of equipment in which vacuum technology is used, which has been reflected in the work of scientists and in the innovative designs of vacuum-insulated apparatus offered by manufacturers. in the work of researchers and in the innovative designs of vacuum-insulated apparatus offered by manufacturers. The dynamic development of the grid and stricter regulations on power outages have necessitated the development of new designs of vacuum equipment.

In countries with a low level of network cabling, such as Poland, medium-voltage overhead networks play a decisive role in shaping reliability factors. Inherent in these networks are disconnectors, which are typically installed in overhead lines, are responsible for connecting operating currents and are most often of open, less often of closed construction, but based on SF₆ technology. It is only in the last decade that research work has been undertaken to develop a vacuum disconnector with a closed design, free of SF₆, using vacuum technology.

The result of the work of researchers and manufacturers of switchgear intended for medium-voltage networks is EKTOS—an innovative medium-voltage overhead vacuum disconnector in a closed enclosure, dedicated to Smart Grids, which was developed in cooperation with researchers from Lublin University of Technology and a company EKTO from Białystok. The innovativeness of EKTOS lies primarily in the innovative, developed from scratch motor storage drive integrated with the contact system which is based on a DC motor, which has allowed the system’s dimensions and power requirements to be reduced to 90W. The place of contactors and limit switches has been taken by a starting system equipped with reed switches. In order to eliminate tie rods and thus simplify the design, the entire drive unit was located inside the disconnector structure. This form of drive system, to the authors’ knowledge, is not found in other devices of this type [17]. Bearing in mind that EKTOS is a device intended for use on overhead lines, its weight is not without significance. Components with reduced weight and size were used in the construction of the apparatus, which has resulted in the final weight of the device being approximately 1⁄5 that of known construction solutions. It is perfectly in line with global preferences in terms of the direction of network infrastructure development. It has based the operation on an emission-free medium such as vacuum, but there are still improvements in the dielectric resistance of the extinguishing medium through the use of noble gases or mixtures of these gases as an arc-extinguishing medium [18]. Another challenge faced by EKTOS is the need to provide real-time measurement of the gas pressure in the vacuum chamber so that potential damage resulting from unsealing of the extinguishing chambers can be predicted.

2.1. Gas-Insulated Switchgear

The constantly increasing participation of clean, low-carbon, renewable energy sources are extremely important in terms of reducing the amount of greenhouse gases that are supplied to the atmosphere. Currently, the most common switching devices used in electrical networks are circuit breakers, the huge majority of which operate based on gas insulation [19]. These are mainly disconnectors based on sulphurhexafluoride (SF₆) or the vacuum. The necessity to reduce the supply of greenhouse gases to the atmosphere, among others used in electrical appliances, was identified through the Kyoto agreement [20], or during the 2015 Paris climate conference [21]. The Kyoto Protocol identified SF₆ gas as one of the most potent greenhouse gases, with a global warming potential (GWP) of 22,800 times that of what is referred to as harmful, carbon dioxide (CO₂) [22,23,24]. Despite this, SF₆ is still widely used on a large scale in grid infrastructure, as an insulating medium in arc extinguishing apparatus, but the latest European Parliament regulation of 5 April 2022 [25] mandating that newly installed or replaced medium-voltage switchgear up to 24 kV, and up to 52 kV being placed on the market after 1 January 2026 and 1 January 2030, respectively, base their insulation on media having a greenhouse potential (GWP₁₀₀) of no more than 10. The regulation also applies to high-voltage equipment operating at 52 kV to 145 kV and above 145 kV. Appliances placed on the market after 1 January 2028 and after 1 January 2031 may not have an insulating medium with a GWP₁₀₀ above 10. The enactment of this regulation can be considered as the end date for appliances with hexa-fluoride sulphur insulation. Many studies have been carried out to identify an alternative insulating medium that would successfully replace sulphur hexafluoride.

Table 1. Summary of alternatives to SF₆ gas [26,27,28,29,30,31,32].

No.	Gas	Global Warming Potential (GWP100)	Atmospheric Lifetime/(Years)
1	Vacuum	0	-
2	CO2	1	200
3	CF3I	5	0.05
4	CH4	27	12
5	C2H4F2	140	1.5
6	N2O	310	120
7	CF4	6500	50,000
8	NF2	8000	50–740
9	C4F10	8700	2600
10	C4F8	8700	10,300
11	C3F8	8830	2600
12	CHF3	11,700	390
13	C2F6	12,200	10,000
14	SF6	22,800	3200
15	CO2	0	-

shows the greenhouse potential and atmospheric lifetimes of selected gases that have been the subject of these studies.

Finding an environmentally friendly gaseous medium with insulating and quenching properties not inferior to those of sulphur hexafluoride, while being economically affordable other than vacuum, has proven to be a challenge [33]. Vacuum does not allow for the conduction of current in the steady state of the insulation or in the transient state during the shorting or opening of the operating contacts [34]. The change in dielectric properties in vacuum chambers is closely related to the shape and material of the contacts used, the distance between contacts, but it is the pressure inside the apparatus [35] that determines the suitability of the apparatus for operation. VI technology fits well with the prevailing trend towards miniaturization of devices, due to the low volume of the quenching medium, high dielectric resistance for small inter-contact gaps and the lack of maintenance during operation. In addition to a number of advantages, vacuum devices have a number of significant disadvantages, the most significant of which is the lack of an industrially suitable method for real-time monitoring of the pressure inside the device [36] and the generation of switching overvoltage caused by current shear before passing through zero. A comparison of the most significant advantages and disadvantages of vacuum-based and sulphur hexafluoride-based devices is included in

Table 2. Overview of the properties of vacuum-insulated and SF₆-insulated equipment [7,19,35,37].

Vacuum Technology (Advantages)	SF6Technology (Disadvantages)
No need for conservation work during the operating period	The need for frequent maintenance work during the operating period
No influence of outside temperature	Vulnerable to external temperature
No risk of explosion	Risk of explosion at high current levels
No negative impact on the environment	Greenhouse gas with extremely unfavorable decomposition products, specialized disposal equipment required
High breakdown voltage for small contact gaps with large deviations	Lower breakdown voltage for analogous contact breaks but with smaller deviations
The contact system is arc resistant	Requirement for special designs to increase contact life
Low drive energy required for operating mechanism—smaller opening and closing travel	Significant drive energy required for operating mechanism—greater opening and closing travel
Seamless transition to steep transient voltage increase	Problems when switching on/off rapid rise in transient voltage
Opening and closing times with slight deviation—easy controlled disconnection of fault and load currents	Opening and closing times with large deviations—complicated control of disconnection of capacitive and inductive currents
Ability to interrupt large fault currents regardless of the position of the moving contact	Lack of ability to interrupt large fault currents when the moving contact approaches the open position
Arc duration of 5–7 ms	Arc duration of 10–15 ms
Vacuum Technology (Disadvantages)	SF6Technology (Advantages)
Difficulties in carrying out switching operations for rated currents above 3150 A	Seamless implementation of switching operations for rated currents above 3150 A
No real-time monitoring of pressure status during operation	Easy real-time monitoring of quality and pressure during operation
X-ray emission of 5.0 μSv∙h-1 under normal operating conditions	No X-ray emission under normal operating conditions
High number of arc re-ignitions when combining inductive currents due to high frequency current interrupting capability	Low number of re-ignitions when combining inductive currents
Generating switching overvoltage’s due to current clipping before natural zero crossing	No switching overvoltage generation

.

Higher switching capacity and short contact changeover operation times are essential for vacuum disconnectors (VCBs) to improve the efficiency of today’s power systems. Therefore, many studies have been undertaken to optimize the performance of vacuum disconnectors. Progress in the mechanism that transfers energy to the mechanical drive train to move the contacts in the VCB has resulted in a significant reduction in contact opening (current disconnection) time, which has dropped from eight to five cycles of power frequency current, and now averages three cycles [38,39,40,41]. Over the past four decades, many different types of tripping mechanisms have been developed for drive systems, including those based on pneumatic, spring, permanent magnet, digitally controlled motor, electromagnetic repulsion and other principles [42,43,44]. The vacuum method can be used to create a fast current circuit-breaker that is controlled by the AC short-circuit current in a short time and realized by a fast vacuum circuit breaker (FVCB—Fast Vacuum Circuit Breaker), also known as fast vacuum breaker technology, which can reduce the breaker time to half a cycle of the short-circuit current [38,39,40,41,42,43,44,45]. The reliable arrangement of the drive mechanism significantly reduces the time of the contact closing and opening operations, which means that the fault condition in which the arc burns is reduced to a minimum.

Although the drive mechanism is of great importance in the quality of the operation of vacuum apparatuses, the vacuum quench chambers are the basic element of these devices.

Vacuum Chambers

A fundamental component of vacuum insulated switchgear are vacuum extinguishing chambers. They are responsible for providing a safe insulation gap in the contact space. The current path is located inside the ceramic or glass enclosure from which the chamber is made. For minimizing contact pad wear during switching operations, the material and shape of the contacts are selected already during the design phase. The use of a resilient bellows, which guarantees the tightness of the system, makes it possible for one of the contacts to move. The chamber contains a condensation screen on which conductive particles are deposited from arcing discharges occurring between contacts. This element is required because conductive particles that have settled on the chamber housing can weaken the dielectric. Vacuum chambers are made with advanced technology which requires the use of unique materials that can withstand pressure differences and mechanical shocks. The pressure inside the extinguishing chamber is typically 10⁻⁴ Pa [46]. Figure 2 shows a cross-sectional view a sample extinguishing chamber [47,48], while Figure 3 shows selected models of vacuum chambers made by the Tele and Radio Research Network’s Łukasiewicz Institute.

The development of vacuum apparatus technology, as well as the research carried out, has not overlooked one of the most important aspects of development work on vacuum technology—the aspect of pressure control in vacuum chambers.

3. Pressure Control in Switchgear

The control of the pressure level in gas-insulated apparatus is extremely important, as the vacuum retains its excellent insulating properties only up to a pressure level of 1 Pa, once the pressure increases beyond this value the breakdown voltage of the insulating gap drops dramatically [50,51,52]. This situation is shown in Figure 4 [18,53].

This is an extremely unfavorable situation, particularly given that the current switchgear used in the power industry is of enclosed construction, which makes it less susceptible to environmental factors and vandalism, but this is coupled with the inability to observe a safe insulation gap. The inability to determine what state the device is currently in poses a number of dangers, the greatest of which is the possibility of potential being transferred to the side that may be de-energized. This poses a serious risk to human health and even life, as maintenance crews carrying out maintenance work on a theoretically de-energized line can be electrocuted. This illustrates the magnitude of the problem of pressure control on gas-insulated equipment. The dielectric strength of the vacuum for different contact distances is shown in Figure 4 [18,53].

The problem of measuring the pressure in vacuum chambers and thus determining the technical condition of these apparatuses affects all equipment that uses vacuum as an extinguishing medium, but it is particularly acute for disconnectors which, due to the specific nature of the switching operations carried out, require knowledge of the pressure prevailing in the extinguishing chamber and thus the ability to determine the presence of a safe insulation gap.

3.1. Chosen Diagnostic Methods for Vacuum Equipment [54]

In answer to a commercial need arising from the lack of a method to detect the degree of vacuum in a device, methods have been developed to determine the serviceability of the device. A summary of the selected methods is shown in

Table 3. Diagnostic methods for vacuum equipment [54].

Method	Detection Range [Pa]	Pressure Indication	Operational Applicability
Pening method	10−5–10−2	+	-
the Magnetron method	10−5–10−2	+	-
50 Hz ignition voltages	>1	+	+/−
DC ignition voltages	>1	+	+
AC bonding capacity	>10−2	−	+/−
Emission currents	>10−1	−	+/−
Emission currents + HF current	10−5–1	+	+
X-ray	>10−1	−	+
DC arc voltage	10−5–1	+	+
VD/VE	10−3–1	+	+
Screen potential	10−3 + 10−1	+	−
Induction current switch-off	>0.3	+	+

.

Most of the presented methods have only found application in laboratory conditions. However, the Pening and Magnetron methods have found favour with vacuum chamber manufacturers as the primary methods for product quality control. The DC static ignition voltage method, on the other hand, is widely used in commercially available vacuum chamber testers. The great potential of the X-ray method has unfortunately been severely limited due to its significant measurement error.

3.1.1. The Pening and Magnetron methods

The main methods that are now being used by vacuum chamber manufacturers are the Pening and Magnetron methods [55]. The principle of their operation consists in subjecting the chamber to a strong axial magnetic field, followed by the application of a DC voltage of the value of 10–20 kV to the open contact system. Under the application of this voltage, an electron emission phenomenon occurs, which, combined with the influence of the axial magnetic field, forces the electrons to spiral towards the cathode, making the path between the contacts much longer, and thus causing many more collisions with particles of the residual gas. The resulting collisional ionization is reflected in the measured current, the value of which enables the magnitude of the prevailing pressure in the chamber to be determined. For the measurements to be possible, it is necessary to first determine, for each design of apparatus under test, the characteristics of the current-pressure relationship. Moreover, due to the necessity of placing the test object in a strong magnetic field, they are only used in laboratory measurements [56].

3.1.2. DC Static Voltage Ignition Method

Producers of test and diagnostic equipment for vacuum devices consider the static DC ignition voltage method to be the best method for determining instrument pressure [36]. This method involves applying a voltage of a certain value to the device and measuring a current, the value of which should not exceed 30 mA. From this, it can only be determined whether the device is operational or not. Unfortunately, the static DC ignition voltage method can be skewed by conditioning discharges, and the measurement accuracy itself is also unsatisfactory [54]. A device using this method is the Vidar vacuum chamber tester presented in Figure 5.

3.1.3. X-ray Measurement Method

The negative fact of X-ray emission has found a positive application in the method of pressure determination. The method is based on the fact that the intensity of the radiation is directly proportional to the emission currents generated, and its undoubted advantage is the possibility of non-contact measurement and the lack of need to change the position of the contacts or interfere with the operation of the apparatus. An impediment to the widespread use of this method is the radiation interference caused by natural terrestrial radiation, which in the case of tests of such small values results in significant measurement distortions [33,54,58].

3.2. Determination of Pressure Status in Real Time

The methods shown in Figure 5, even though capable of determining the pressure during the operation of the device, do not allow real-time measurements to be made. A measurement taken at any time does not guarantee correct operation in the future. In addition, depending on the method used for inspection, it may be required to take a section of the line out of service in order to remove the apparatus and transport it to the laboratory for voltage testing and to shut down the line again in the event of a fault, thereby increasing reliability factors. A major concern is the safety of the crews working on the line, as a chamber tested by traditional methods does not guarantee correct operation in the near term. Typical methods used in industry to determine pressure levels in vacuum chambers, together with a breakdown of the pressure ranges for which they can be used, are summarized in Figure 6.

Considering the fact that vacuum technology is constantly evolving and that there is no suitable method to determine the pressure level in extinguishing chambers in real time, the topic has received considerable interest from researchers around the world, resulting in the development of many configurations of measurement systems to detect the pressure level in vacuum apparatus.

3.3. Piezoresistive Sensors for Pressure Measurement

The low-pressure range can be measured using a microelectromechanical systems (MEMS) pressure sensor, which has the advantages of small size, low power requirements, high performance and the possibility of mass production by micromachining. One of the most popular MEMS devices on the industry is pressure sensors. Indeed, it is worth noting that there is a wide range of applications for these sensors, including microscale mechatronic systems, vehicles, aerodynamics, process control and biomedical apparatus [60]. Piezoresistive, capacitive, optical and resonant sensors are just some of the subtypes of MEMS pressure sensors that can be classified according to sensing principles. Due to their excellent sensitivity, simple design and durability, piezoresistive sensors stand out among them and have been the subject of many studies [61]. In their study, P. Martins et al. tested piezoresistive pressure sensors made by printing technology on ink-based textile substrates and polyethylene terephthalate (PET) [50]. In their work, they demonstrated that, unlike sensors known in industry, the sensors they tested give a linear response. This requires additional verification to confirm the linearity of large-scale operation. S. Kumar and co-workers presented a sensor design on a square diaphragm, which better transmits mechanical stresses, and an amplifier circuit for the above element [51]. Many more piezoelectric sensor designs have been developed in a variety of materials with significant improvements in sensitivity, as shown in

Table 4. Chosen piezoresistive sensor designs [52,62,63,64,65,66,67].

Structure	Sensitivity [kPa−1]
Micro copula	15.10
Micro pyramid	10.30
Micro pillar	2.00
Hierarchical	19.80
Empty crutch	133.10
Textile	14.40

.

3.4. Fiber Optic Technology in Pressure Measurement

Fiber-optic sensors, such as the Fabry–Perot interferometer (FPI) sensor, Bragg grating (FBG) sensor, polymer FBG sensor and other fiber-optic configurations [68,69,70,71,72,73,74,75,76,77] for detecting deformation due to applied pressure, have gained considerable attention from researchers due to their small size, low weight, resistance to electromagnetic interference and high sensitivity. In addition to a number of advantages, fiber-optic technology has the significant disadvantage of being susceptible to temperature effects, so compensation systems are required, and sensors and measurement systems need to be augmented with additional components to measure the reference temperature.

3.4.1. Sensors with a Fabry-Perot Interferometer

The Fabry–Perot interferometer relies on the use of an optical cavity created by placing two optical reflectors M1 and M2 with reflection coefficients R1 and R2 in a specific position on two sides of an optically transparent medium of length h, as shown in Figure 7. The incident light beam produces multiple reflected beams between which interference occurs, giving an interference wave pattern in the spectral domain. The use of IFP in measurements exploits the susceptibility of the pattern to modulation induced by external interference causing changes in refractive index and optical cavity length [78].

Technology progress has allowed the use of fiber optics as a material for the manufacture of FP sensors. The design of FP interferometers differs depending on the place of use, i.e., for indoor sensors a technology is used where the fiber itself fills the space of the transparent medium (a), while for outdoor designs an air gap or other materials different from the fiber are used (b). The optical gap design for both cases is shown in Figure 8.

Many variants of fiber-optic sensors have been developed to measure, among other things, pressure, which were collected in a book by Y.-J. Rao and others [78]. Other design versions of interferometers are also known. Y.-J. Rao and co-workers presented an FPI made from a hollow photonic core of a bandgap optical fiber [79], J. Villatoro and others developed a version made from a solid photonic core of a crystalline optical fiber [80]. The problem of the former is the need to coat the fiber face, which is complex and expensive, while the latter has questionable stability due to air holes through which the system is even more vulnerable [81]. C. Wu and co-workers developed a high-pressure and high-temperature Fabry–Perot interferometer [82], proving its correct operation up to pressures of the order of 4 × 10³ kPa, with an accuracy of 5.8 × 10⁻³ pm/kPa. W. Wang et al. developed a pressure sensor design capable of measuring from 0 to 50 kPa with an accuracy of 5.3 × 10³ pm/kPa [83]. Recent studies show the use of the Mach–Zehnder interferometer [84] and the Sagnac interferometer [85]; however, due to the need for precise positioning of the interferometer components and the relatively complex structure, they have disadvantages in practical application in operation. A summary of selected fiber-optic methods designed for pressure detection is shown in

Table 5. Overview of different fiber optic gas pressure sensors [61,83,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103].

Sensor Type	Sensitivity [pm/kPa]	Temperature Crossover [kPa/°C].	Range of Linear Response [kPa]	Production Type
Anti-resonance sensor with open side channel	4.24	-	0–2 × 103	Famtosecond laser
Anti-resonance guiding mechanism with hollow fiber reflection (HCF)	3.59	7.5	0–2 × 103	Famtosecond laser
Single FP interferometer with silica membrane	1.036	0.96	0–2 × 103	Shell
FP double cavities with composite membrane	30.2	-	0–0.4 × 103	Shell
MZ cascade interferometer with myctro-machined air cavity	82.131	0.647	0–0.7 × 103	Famtosecond laser
Cascade FP interferometer in a glass capillary tube	86.64	5.18	0–0.6 × 103	Famtosecond laser
Parallel connection of the FP interferometer to the gas well	47.76	5.1	0–0.45 × 103	Famtosecond laser
FP and Sagnac interferometer separated by silver foil	31.73	-	0–1.6 × 103	Shell
Parallel connection of FP interferometers with thin UV adhesive layer	38.3	-	0.1–0.7 × 103	Shell
Equivalent combination of FP interferometer with HCF	45.76	0.097	0–3 × 103	Arc discharge

.

The table shows that, of the presented sensors, the best method is the cascaded FP interferometer in a glass capillary tube, characterized by the highest sensitivity and the lowest temperature influence on the measurement, but the production of the above sensor using a femtosecond laser is technologically complicated and expensive, which poses a serious problem for the commercialization of this technology. A consensus between price and ease of fabrication appears to be the parallel combination of an FP interferometer with an HCF, but this technology requires further research and verification of the sensor’s performance under operational conditions.

3.4.2. Sensors with Bragg Gratings

Fiber optic Bragg gratings have gained widespread use in pressure sensing systems, as reflected in many scientific papers. An optical Bragg grating is a periodic structure with a variable refractive index formed in the core of an optical fiber by treating it with UV light. The operation of the grating can be likened to that of a band-stop filter in the transmission operating state [96]. The structure, with its variable refractive parameter, reflects specific wavelengths that do not differ from the wavelengths for the Bragg resonance, and transmits the others without interfering with the passing signal. A schematic of a fiber-optic Bragg grating is shown in Figure 9.

An unframed FBG pressure sensor with a sensitivity of 3.04 × 10⁻³ pm/kPa was first described by M. Xu and co-workers [97], which is a low value for practical pressure measurement. Pressure sensitivity has been improved over the years by experimenting with different housing designs and fabrication technologies. An improved FBG pressure sensor with a sensitivity of 1.414 pm/kPa constructed by J. Huang and co-workers [98] was introduced. It uses a Bourdon tube as the pressure-treated element. M. Liang et al. [93] developed a circular membrane–cantilever pressure sensor based on FBG, which uses two grids glued to the surface of the beam. However, the sensitivity of 0.34 pm/kPa is far below the preferred range, in practical measurements. Developed by J. Huang et al. [99], a fiber-optic sensor had a pressure sensitivity of 1.57 pm/kPa. The inhomogeneity of the Bragg grating stress due to its adhesion to the surface of the circular membrane remains a problematic issue, which negatively affects the accuracy and reliability of the measurement. The FBG pressure sensor based on a circular membrane was designed by V. Pachava and co-workers [100]. The measured sensitivity of the sensor was quite high at 32.02 pm/kPa. Although, in general, the sensor design is robust the core used is brittle and fails. In addition, the effect of temperature on the pressure measurement was not eliminated by the single Bragg grating in the actual applications. Due to the FBG’s simultaneous sensitivity to strain and external temperature, temperature compensation methods that isolate the wavelength shift caused by temperature in FBG pressure sensors are crucial. The pressure sensitivity coefficient of two FBG pressure sensors bonded to the same production materials is comparable, as reported by F. Gu et al. [101]. Both pressure and temperature affect FBG1, but only temperature affects FBG2, which is the reference point. Both FBGs are in the same temperature field and have generally equal center wavelengths. As a result, they are able to produce the same temperature response or the same temperature-induced wavelength shift. By directly comparing the two FBG wavelength shifts, the recorded pressure levels can be determined. Similar to the work of V. Neeharik [102]. Another concept is implemented using twin FBGs whose wavelength shift responds to both temperature and pressure fluctuations as in the paper by E. Vorathin’s [103]. A structurally similar variant of this approach described by M. Liang [93] is a reference FBG2 that responds to temperature fluctuations. The two FBGs attached to the structure show the same response to temperature, according to M. Lian and co-workers, the difference being that the external pressure loads in FBG1 were converted to positive strain, while in FBG2 they were converted to negative strain. The difference in wave drift helps remove the effect of temperature from the pressure measurements. The pressure sensor, based on the Leal–Junior et al. membrane, was able to have a reduced transverse sensitivity due to a compensation approach involving the modelling of the sensor [104]. A flexible membrane made of martensitic stainless steel was used by G. Hegde and co-workers [105] to create a sensor with a temperature-compensated fiber Bragg grating. The temperature-compensated FBG mesh was then bonded to a stress-free area of the same flexible membrane. The experimental results show that the pressure sensitivity of the sensor was 3.64 pm/bar. A temperature sensor with a circular membrane was developed by L. Wang with his team [106] using metallised optical fibers. The team focused on the resolution and precision of the sensor, with the temperature accuracy reaching 0.1%, although no temperature compensation methods were used. The square diaphragms, steel trusses and vertical beams were used in the design developed by Q. Fan et al. [107] to investigate and propose a pressure sensor. The pressure sensitivity of the sensor was 622.7 × 10⁻³ pm/kPa and its hysteresis was minimal. Q. Fan and co-workers [108], developed a sensor consisting of a combination of a single square membrane and FBG characterized by high temperature compensation, but its sensitivity was too low. Although high sensitivity was guaranteed by the sensing structures used in the above research work, the small design made it difficult to eliminate hysteresis or maintain high sensitivity for better temperature compensation effects. Z. Liu et al. [109] developed a square membrane sensor coupled to a lever structure with a sensitivity of 3.35 pm/kPa. A comparison of selected fiber optic sensors is given in

Table 6. Comparison of selected sensors with Bragg gratings [93,99,100,101,102,103,104,105,106,107,108,109].

Parameters	Construction by Q. Fan [108]	Construction by J. J. Huang [99]	Construction by M. Liang [93]	Construction by V. Pachav [100]	Construction by F. Gu [101]	Construction by Q. Fan [108]	Construction by Vorathin [103]	Construction by G. Hegde [105]	Construction by L. Wang [106]
Pressure range	0–200 kPa	0–1 × 103 kPa	0–10 × 103 kPa	0–207 kPa	0–16 × 103 kPa	0–2 × 103 kPa	0–10 kPa	0.4–70 × 103 kPa	0–1000 kPa
Temperature range	20–55 °C	20–90 °C	5–70 °C	-	25–65 °C	-	27.5–52.5 °C	−40–90 °C	1–35 °C
Pressure sensitivity	3.402 pm/kPa	1.57	0.34	31.9	6.94 × 10–2 pm/kPa	6.227 × 10−1 pm/kPa	1.007 × 102 pm/kPa	3.64 × 10−2 pm/kPa	1.611 pm/kPa
Resolution	6.26 kPa	-	-	-	2.88 × 10−4 kPa	-	-	-	-
Hysteresis	1.09%	-	-	-	-	0.66%	-	0.75%	-
Repeatability	98.80%	99.40%	-	99.91%	99.71%	99.72%	-	99.98%	99.98%
Pressure measurement error	5.53 kPa	17.2 kPa	92 kPa	-	-	1.46 kPa	-	12.6 kPa	-

.

Analyzing the table above, it can be seen that by far the best pressure sensitivity was the sensor developed by E. Vorathin and co-workers. This sensor achieved a pressure sensitivity of 100.7 pm/kPa, with a mean square error of 0.1976 kPa. It should be borne in mind that, apart from the pressure and temperature ranges and pressure sensitivity, E. Vorathin does not specify the parameters of the developed sensor, such as repeatability or measurement hysteresis, so it needs to be analyzed in more detail, but by classifying the presented sensor in terms of pressure sensitivity, it provides a good development basis for the design of a pressure measurement sensor for electrical apparatus.

3.5. Measurement of Partial Discharge

A variety of detection methods have been investigated to track partial discharge (WNZ) activity, including electrical charge pulses, chemical changes, acoustic wave emissions and electromagnetic wave radiation. There are a variety of sensors and signal processing techniques that can be used to monitor and investigate corona discharges. International Electrotechnical Commission (IEC) 60270 standards and other unconventional methods have been developed and used to test for partial discharges in a number of ways to ensure adequate monitoring performance. Depending on the method of assessment, these techniques can be classified as offline or online detection techniques. Online monitoring, as opposed to offline monitoring, is becoming more common as it causes less system disruption and power outages [110].

Given the need to develop a reliable real-time pressure monitoring system, the attention of researchers has been directed to the measurement of incomplete discharges as a solution that shows great potential toward determining the pressure in gas-insulated switchgear. The idea of the measurement is based on the incomplete discharge occurs between the floating shield in the closed state and between the contacts in the open state, and the measurement of the amount of these discharges depending on the given pressure, through a coupling capacitor connected to the shield inside the chamber [53]. Research conducted by the researchers, proved that there is a relationship between the number of discharges and the pressure level in the apparatus, which is a promising method for indirect pressure measurement in gas-insulated devices. Typical methods of detection of incomplete discharges are shown in Figure 10.

3.5.1. IEC 60720 Method

IEC 60720 standards are well known for their accuracy and ability to monitor the level of partial discharge in unplugged environments. The basic elements of the test element circuit (C_x), the coupling capacitors (C_b) and the measurement impedance (Z) are shown in Figure 11 as a test circuit for this technique.

A coupling capacitor can be used to record current pulses below 1 megahertz (MHz), which are caused by partial discharge activity in the test element and connect them to the measurement impedance. The signal can be presented in both time domain and phase domain to highlight the characteristics of the occurrence of partial discharges. There are two different patterns of partial discharge activity that show the relationship between discharge amplitude (q) and number of cycles (n). In terms of phase position (φ), the first are phase-separated partial discharges (PRPD) and phase-separated pulse sequences (PRPS). These were proposed by Fruth in 1990 and are mainly based on statistical considerations to illustrate the characteristic style of several types of damage caused by partial discharges. The three types of partial discharge patterns are surface discharge, hollow discharge and corona discharge. These patterns can be characterized on the basis of prior contributions thanks to data from q-n-φ representations [111]; however, a significant drawback of PRPD and PRPS is their inability to distinguish between source types when several defect types are present and overlapping information in the phase domain can seriously affect the performance of partial discharge classification. To address the challenges of implementing monitoring remotely using IEC 60270 methodologies, researchers have looked at different ways of continuously monitoring detection methods in high-voltage equipment [112]. An actual pole insulator with a microcrack defect was used in an experiment that was conducted by J. Cheng [113]. For a comprehensive study, the radio frequency signal and important parameters based on IEC 60270 standards were measured simultaneously. The results showed that sub-millimeter cracks could occur in the GIS insulator.

3.5.2. Acoustic Method

Acoustic emission is often observed in discharge processes due to the large vibrations of the electrons, and the sampling frequency for acoustic emission detection is 20 kHz–1 MHz. Due to its advantages, such as low sensitivity to electrical noise and the ability to detect discharge places, the acoustic approach is used in partial discharge monitoring systems [114]. Piezoelectric sensors can be installed externally for online monitoring of the appearance of partial discharges without losing power. Solid and liquid insulation, internal insulation and external design of operating equipment tend—in a good way—to influence the propagation characteristics. The solid, liquid and internal insulation of the operating equipment can have a positive effect on thermal propagation. Additionally, depending on the length between the sensors and the source of the partial discharge, the acoustic attenuation can be quite significant [110]. The acoustic way of estimating partial discharges is related to the conventional electrical way, in which partial discharges are measured under different conditions using AET. H. D. Ilkhechi and M. H. Samimi in their work [115] analyzed acoustic approaches and acoustic-electrically coordinated detectors for partial discharges.

3.5.3. Radio Frequency Method

In the case of partial discharges, the radio frequency (RF) method uses matched sensor equipment to distinguish and collect the electromagnetic signal created. On the other hand, this can reveal the occurrence of partial discharges rather than their exact position. In the meantime, due to its noise immunity and localization accuracy, the UHF procedure is commonly used to inspect sources of partial discharges [110]. Since the measurement frequency is between 300 MHz and 3 GHz, which is above the electronic value, the UHF method has an excellent signal-to-noise ratio. UHF detectors (radio wires) are placed in equipment using oil valves or dielectric windows to collect UHF signals. An interesting concept was presented by Y. Wang and his team [116]. They developed a fiber-optic ultrasonic system for the detection of partial discharges. In order to overcome the problem of the influence of electromagnetic interference on the measurement results, they proposed to use a fiber-optic Sagnac interferometer for ultrasonic detection of partial discharges. The design of the system is not complicated, but detection of partial discharges may not be sufficient for industrial applications mainly due to the accumulation of live components, for example in a vacuum circuit breaker. This can lead to detection of discharges from the wrong current path, or even to detection of a partial discharge in the device, but without being able to determine the specific location of the discharge.

3.5.4. Other Innovative Measurement and Detection Methods

Y. Nakano and co-authors [117] investigated the characteristics of partial discharges in the medium vacuum range in air and SF₆ gas as a leakage gas for a vacuum stage diagnosis method. A detection resistor attached to a grounded flat electrode outside the device can detect the partial discharge. From the acquired charges, the team determined the incidence and phase characteristics of the discharges for the applied AC voltage, as well as the internal pressure of the vacuum. The internal pressure of the device can be determined by statistical analysis of the discharge occurrence histogram for its phase, but these results can only be applied to a device with the same geometry and voltage rating.

C.-J. Chou and C.-H. Chen [118] in their research have introduced a measurement method using a high-frequency current transformer (HFCT), and then modelled various partial discharges, including internal discharges, discharges along the surface and discharges at the ends of conductors. In addition, they analyzed the characteristics of the discharge waveform and phase distribution to identify the type of partial discharge. The measured and analyzed results for the real case are compared with standard discharge pattern data from the laboratory. This can form the basis for a database of partial discharges in medium-voltage power equipment, which is also required for the development of an automatic partial discharge identification system.

H.-B. Su and others [119] have presented a technique for monitoring the degree of vacuum remotely in high voltage vacuum circuit breakers (HVVCBs) together with a data processing circuit. In order to receive the electromagnetic signal produced by the pulse discharge, a ring antenna is placed on the vicinity of the vacuum breaker. The electromagnetic signal must then be converted into an electrical signal and sent to the circuit for processing and treatment. The system, after filtering the signal, pre-amplifies it by a certain multiple and keeps track of the number of pulse signals processed and the average amplitude of the signal over 30 s for each vacuum level. A relay is then triggered according to the selected threshold amplitude values and number of signals to signal that the threshold parameter has been exceeded. Measurement by this method is possible without interfering with the design of the device, but requires the use of an antenna, the location of which can be problematic, particularly in operational applications, and the pressure values obtained by this method are in the order of 1 Pa, so there is the possibility of detecting system leakage.

So Yeon Kim and a research group [120] have compared the partial discharge characteristics of SF₆ and dry air, showing that there are no significant differences in the propagation of partial discharges which allows the interchangeable use of identical corona discharge sensors for both sulphur hexafluoride and dry air.

Nam-Hoon Kim et al. [121] studied the characteristics of partial discharges in g3 gas and dry air and compared them to SF₆. They showed that hexafluoride sulphur had better insulating properties, with g3 showing about 80% of the insulating capacity of SF₆.

S. Bang and co-workers [55] have described a method for pressure determination based on partial discharges achieving pressure detection of 10⁻² Pa. The measurement was performed using an LDS-6 partial discharge analyzer from Doble Lemke Co. The system is promising; however, the number of components required is problematic during operational applications.

In another paper, S. Bang and others [122] have compared partial discharges taking place at AC and DC voltages for positive and negative polarity. They used an LDS-6 analyzer for the detection of partial discharges. They proposed maintenance standards depending on the voltage type, monitored the vacuum level in the distribution class and compared AC and DC partial discharges. According to the voltage type, the vacuum level at which the dielectric resistance drops rapidly was validated for proper device operation. In addition, a coupling capacitor was mounted directly on the floating cover of the apparatus to improve the precision of the partial discharge signal detection. The PRPD for the AC partial discharge and the PSA for the DC partial discharge were used to measure the pattern and apparent charge of the partial discharge with respect to the degree of vacuum. The work carried out by S. Bang and co-workers is well suited to testing vacuum chambers for transmission voltages but can be problematic to use in operation.

W. I. Wang, with co-workers [123], has developed a method to determine the degree of vacuum in the apparatus based on continuous wave terahertz imaging. They observed that the signal intensity of the echo spectrum is inversely proportional to the degree of vacuum in the chamber. Which makes it possible to conclude that as the degree of vacuum increases, the amplitude attenuation effect of the detected terahertz signal is gradually weakened, and the time delay is reduced. It shows that the spectrum and absorption spectrum obtained by continuous wave terahertz imaging technology can sensitively distinguish the pressure change in the vacuum environment.

A. Caprara and others [124] have implemented measurement techniques from high-voltage systems into medium-voltage equipment. The use of the Rogowski coil together with the RFCT sensor allowed them to show that the partial discharge detector, thanks to its ability to track individual phenomena, is not affected by the false positive warnings typical of current technologies available for online partial discharge monitoring of distribution networks.

I.-H. Chung [125], in his research, has described the application of a neuro-fuzzy inference system to identify defects in gas-insulated equipment based on partial discharges. He subjected 17 parameters characteristic of corona discharges to analysis. He achieved the system’s ability to detect device defects at a level of 90%; however, a comprehensive database is required for the verification accuracy to be at an adequate level, for the system to give more accurate defect detection results in practical applications, and for the stability of system operation to improve.

4. Summary

The article analyses and compares the degree of knowledge in the field of monitoring and diagnosing the condition of gas-insulated electrical equipment in the context of determining the pressure values prevailing in vacuum extinguishing chambers in switching apparatus. The significant technological advances that have taken place in recent years have resulted in a plethora of technological solutions to improve the speed and accuracy of pressure measurements in gas-insulated switchgear, but despite all efforts, there are still a number of issues that need to be clarified and that require further experimental work. One of these is real-time pressure measurement, as most measurement methods are geared towards measuring overpressure, which is not applicable to switchgear that is based on high vacuum. To the knowledge of the research group, which includes the authors of the above document, there is no technological solution for continuous pressure measurement designed for equipment that is continuously in operation. This leaves a lot of space for research and development work to enable the development of an online vacuum chamber diagnostic system. The most promising methods are associated with the use of fiber optic techniques due to their small size, low weight, resistance to electromagnetic interference and high sensitivity. Current research makes it possible to claim that the development of measurement and diagnostic methods for both vacuum chambers and the target vacuum apparatus will be closely linked to information technology, making it possible to create an autonomous system for monitoring the parameters of electrical apparatus with little operator intervention.