Power Quality in the Context of Aircraft Operational Safety

Abstract

The article presents the use of ground power sources for aircraft ground support. Both military and civil aircraft (A/C) require high-performance and reliable Ground Power Units (GPUs) to ensure safe operation in diverse environmental conditions. The power quality provided by these GPUs plays a crucial role in determining the reliability, cost efficiency and operational safety of the aircraft. The main objective of the article is to signal and propose a solution to the problems associated with the operation of GPUs in the Polish Armed Forces (PAF), resulting from the diversity of the equipment used (type, duration of operation, defects occurrence, etc.). Currently, the PAF utilize various types of GPUs to provide electrical power to aircraft while they are on the ground. Many of those devices have already been in service for many years. The presented statistics of defects registered in the airworthiness management system showed several dozen A/C failures or defects related to the operation of GPU. The authors highlight the importance and feasibility of diagnosing these ground-based power sources. The presented sample test results confirm that, following the methodology outlined in the article, it is possible to conduct comprehensive diagnostic assessment of the GPU systems currently in use by the PAF, as well as evaluate the quality of the electrical power they deliver in both steady and transient states.

1. Introduction

One of the key factors directly influencing the efficiency of the operational process of any technical object, including an aircraft (A/C), is its readiness. This factor is not solely related to the properties and characteristics of a specific type of aircraft but also to its operating system, the systems cooperating with it, and the surrounding environment. When aircraft are on the ground, not using internal power sources, external ground power units (GPUs) are used to provide electrical power to their systems. Both military and civilian aircraft require a high-performance and reliable GPU to ensure safe operation under various environmental conditions. The quality of the power supplied by the GPU has a significant impact on the reliability, as well as the costs and safety of aircraft operations. The technological advancement of aircraft has made their operation increasingly resource-intensive, demanding a wide array of support to ensure they are used with optimal efficiency and safety. This safety is defined by the aircraft’s continuous airworthiness management and the assurance of high operational reliability, both of which are closely linked to effective ground maintenance with GPU utilization.

The presented analysis is the first of its kind prepared in the Polish Armed Forces (PAF), and indicates that currently, they employ a variety of ground power supply sources for aircraft maintenance and operations. Many of these power sources have been in use for several years, and statistics from the airworthiness management system [1] have recorded numerous aircraft defects or failures related to the operation of GPU. Therefore, ongoing assessment of the quality of power supplied by GPU is crucial for ensuring proper aircraft operation [2,3]. Poor quality of electrical power from GPU, especially during transient states (such as overvoltage pulses, peak pulses, undervoltage pulses or instability), can pose a serious threat to modern onboard systems, which rely on sensitive electronic and microprocessor systems. Additionally, incorrect power parameters during stable states can severely affect the reliability of checks and adjustments performed during aircraft maintenance, potentially leading to erroneous or incomplete execution, posing a serious risk to operational safety, flight security, and mission execution. Therefore, aircraft manufacturers often recommend the use of GPU from authorized suppliers, which increases operational costs and prevents the use of existing ground power sources.

2. Problem Basis

Supervision over the aircraft operation encompasses all aspects related to safety, including adherence to aircraft operation and airworthiness, as well as the maintenance of other aviation equipment regulations, the use of airfields and ground aviation facilities, aircraft ground support equipment, maintaining the qualifications and certifications of aviation personnel, and adherence to air traffic rules and regulations. In accordance with Polish Military regulations, operation refers to a set of purposeful organizational, technical, and economic activities carried out by personnel in relation to the aviation technology equipment, as well as the mutual relationships between them, starting from the moment the aircraft is put into service according to its intended purpose until its disposal [4,5]. Aircraft operation is conducted through the primary process of in-flight use, airworthiness management, as well as readiness. The course of aircraft operation involves planned activities resulting from the need to perform tasks designated for a particular type of aircraft, which is necessary to maintain its readiness while taking into account the physical infrastructure (tools, facilities, personnel, supplies), which are key factors determining operational readiness. The accomplishment of tasks in aviation systems is achieved through compatibility and synchronization of various technical resources interacting with the aircraft in a specified order and within the designated time frame [5].

The operational system significantly influences the reliability and readiness of the aircraft, the safety of flights, and the completion of missions within the designated time frame. While individual components of aviation operation support systems generally do not directly affect the technical readiness of aircraft, in specific cases, they can lead to a reduction in readiness.

This occurs when these components cause failures or defects to the aircraft, often due to errors in the operation of personnel or improper functioning of the technical equipment. Regardless of working conditions, the electrical power supplied by GPU should not negatively affect the aircraft’s electrical system or cause discrepancies [2,3]. One contributing factor to an aircraft becoming unairworthy is the poor quality of electrical power supplied by GPUs, which can have a significant impact on the safety of its operation.

3. Research Object

GPUs are part of an aircraft’s Ground Support Equipment (GSE). Within the inventory of the PAF, there are 28 types of GPUs, including fixed or portable units, which comprise 53% of the total, and vehicle mounted or towable units, which constitute 47%.

The PAF utilize 16 types of fixed or portable GPUs. A detailed list of these GPUs in service is presented in

Table 1. List of fixed or portable ground power units (GPU) operated by the Polish Armed Forces (PAF).

No.	GPU NAME (Type)	Number of Vehicles Operated in the PAF
1.	LUZES-II/M Ground Power Unit	86
2.	GPU-6M Ground Power Unit	77
3.	B1 LUZES II/M 102-15.00-S container	9
4.	GC20 Ground Power Unit	8
5.	LUZES-III/M Ground Power Unit	4
6.	GPU28V DC 400/600A Ground Power Unit	4
7.	GPU 6/M Series III Ground Power Unit	4
8.	GB60/20V7 Ground Power Unit	4
9.	GPU-600 Solid State Ground Power Unit	4
10.	GPU-6 Ground Power Unit	2
11.	LUZES II/MVI Ground Power Unit	1
12.	APOJET AJR 120 50 Hz/400 Hz Ground Power Unit	1
13.	AP. AJS90 COM.90 kVA/400 Hz/GPU28V Ground Power Unit	1
14.	B1 LUZES II/M 102-00-00-01 container	1
15.	SYS. LUZES electro-energetic device set	1
16.	UNITRON UFC-45M Ground Power Unit	1

.

The LUZES-II/M and GPU-6M GPUs are the most numerous groups. Single units of other GPU types, other than those mentioned in Table 1, are also operated in the PAF. The percentage distribution of each GPU type is shown in Figure 1.

Table 2. A list of vehicle-mounted or towable GPUs operated by the PAF.

No.	GPU NAME (Type)	Number of Vehicles Operated in the PAF
1.	APA-5D Ground Power Unit	44
2.	LUZES V/D on the JELCZ442.32 Ground Power Unit	44
3.	LUZES V/D Ground Power Unit on the STAR-944 vehicle	35
4.	LUZES-V/D Ground Power Unit on the STAR-266M2 vehicle	21
5.	LUZES-V/N Ground Power Unit	17
6.	LUZES-V/D Ground Power Unit on the STAR-744 vehicle	14
7.	P70TA000A03-801 Ground Power Unit	6
8.	CUMMINS QSB DIESEL 90CU420 generator	2
9.	HOBART 4400 Ground Power Unit	2
10.	LUZES V/N S. 3 Ground Power Unit	1
11.	EINSA Ground Power Unit	1
12.	LUZES V/D on the STAR 742 Ground Power Unit	1

presents a list of 12 types of vehicle-mounted or towable GPU units operated by the PAF.

The most common in this group are the APA-5D and LUZES V/D GPUs mounted on a JELCZ 442.32 and STAR-944 chassis. Among the 12 types of currently operated vehicle-mounted or towable GPUs, these three models make up 65% of the PAF equipment. There are also single units, other than those mentioned in Table 2, of these types of GPUs operated within the PAF. The percentage distribution of different types of vehicle-mounted or towable GPUs used by the PAF is shown in Figure 2.

Considering only the type of vehicle-mounted or towable GPUs (regardless of its transport platform), their numerical distribution is shown in

Table 3. A list of vehicle-mounted or towable GPUs, excluding transport type.

No.	GPU NAME (Type)	Number of Vehicles Operated in the PAF
1.	LUZES V Ground Power Unit in D and N Series	133
2.	APA-5D Ground Power Unit	44
3.	P70TA000A03-801 Ground Power Unit	6
4.	HOBART 4400 Ground Power Unit	2
5.	CUMMINS QSB DIESEL 90CU420 generator	2
6.	EINSA Ground Power Unit	1

, with the percentage breakdown illustrated in Figure 3. It can be observed that LUZES V comprises a substantial 71% of the inventory of the PAF, followed by APA 5D units at 23.5%, with the remaining 5.5% consisting of other GPU types.

A significant factor for GPUs in use is its operational age. The aging processes that occur during the lifecycle can substantially contribute to the deterioration of the technical condition of the GPUs in use, thereby affecting the power quality they produce. The age of fixed and portable GPUs in the PAF is illustrated in Figure 4. The oldest operational GPU for this group is 37 years (one LUZES-II/M), while eight GPU-6M units have been in service for 33 years. The LUZES-II/M and GPU-6M type units also belong to the youngest group of fixed and portable GPUs used in the PAF. Additionally, in recent years, other types of GPUs have been introduced into service, including: GPU28V DC 400/600A, APOJET AJR 120 50 Hz/400 Hz, AP. AJS90 COM.90 kVA/400 Hz/GPU28V, GC20, GB60/20V7, UNITRON UFC-45M, and GPU-600.

The age of vehicle-mounted and towable GPUs is shown in Figure 5. It should be observed that the older operational GPUs reach as much as 42 years.

Three groups of GPUs can be identified based on their age:

• First Group—ranges from 1 year to 13 years
• Second Group—ranges from 19 to 29 years
• Third Group—ranges from 34 to 42 years

The last group consists of 48 units, among which three of them have been in service for an impressive 42 years—one APA-5D and two LUZES V/D units mounted on the STAR 266M2 chassis.

The first age group of GPUs include, among others, the P70TA000A03-801, HOBART 4400, and LUZES-V/N, as well as LUZES V/D mounted on the JELCZ 442.32 chassis.

3.1. Statistics on Aircraft Defects Related to GPUs

During the operation of military aircraft, defects have occurred due to poor power quality or GPU failures. From 2014 to 2023, the airworthiness management IT system [1] recorded seventy-two defects or failures related to GPUs, which resulted in the aircraft being rendered unairworthy. The number of these events over the subsequent years is illustrated in Figure 6.

Figure 7 presents the number of defects and failures related to GPUs in relation to specific A/C types.

The highest number of defects or failures on the W-3WA type associated with the GPUs, shown in Figure 7, is probably due to the large population in the PAF of this type of helicopter and the high current draw from the power supply during start-up of this aircraft, which is a high requirement for power quality. Engine starting on the W-3WA helicopter is performed using an electric starter mounted directly on the engine gearbox.

3.2. Examples of Consequences of the GPU Improper Operation

The non-compliance of power quality [3] or GPU defects have led to the unairworthiness of various aircraft, including the following cases:

• Incorrect assessment of AC EPM unit operation on SH-2G helicopters, resulting in unnecessary unit withdrawals from service.
• Non-compliant power quality of observed onboard converters used during the ground checks of aircraft systems, e.g., radio navigation devices.
• Illumination and extinguishing of the GPU button during the CASA C 295M engine start.
• Loss of indications for engine parameters and circuit breakers tripped during the startup of the W-3WA helicopter.
• Inability to power up the PZL-130TC-II aircraft.
• Disconnection of the GP-21 GPU during the MiG-29M aircraft engine start.
• Voltage fluctuations in the AC (Alternating Current) power supply in the onboard power supply system of the M28B aircraft.

The presented examples confirmed findings from studies conducted by the AFIT (Air Force Institute of Technology) regarding the incorrect assessment of AC EPM unit performance on SH-2G helicopters [6]. The investigations were carried out using a test setup for the AC EPM unit M24021-4 type or M24021-5 type from the SH-2G helicopter, as depicted in Figure 8.

The studies revealed that the AC EPM helicopter units declared as faulty were fully operational; however, the power quality from the GPU (used as a power source for the aircraft) did not meet the requirements [3]. This discrepancy led to the unit being taken out of service and the aircraft being unairworthy.

The next presented example was caused by a seemingly minor parameter: the ripple of the direct current voltage at the GPU output (Figure 9a). It turned out that this parameter significantly affects the performance of the converter, leading to noticeable amplitude modulation of the converter’s output voltage (Figure 9b). The distorted voltage from the converter negatively impacts the operation of onboard devices during aircraft maintenance and affects the reliability of checks and adjustments. In contrast, Figure 9c shows the correct waveform of the direct current voltage at the GPU output, while Figure 9d illustrates the output voltage of the converter.

Poor power quality supplied from the GPU to A/C, particularly during transients (such as overvoltage pulses, peak pulses, undervoltage pulses, and voltage instability) poses a significant risk to modern on-board systems that utilize sensitive electronic and microprocessor-based circuits [7]. In modern aviation, the reliability of on-board systems increasingly depends on the utilized GPUs [8,9,10].

The power quality supplied to A/C has a significant impact on reliability, operating costs, and safety. Poor parameters of electrical power can disrupt the operation of the A/C on-board systems, lead to shutdowns, or, in extreme cases, cause damage. Apart from constant disruptions related to the load from properly functioning systems and electrical consumers of A/C, intermediate disruptions are also significant. These disruptions can stem from failures within the power distribution system or local electrical installation of the aircraft, as well as high-power consumers, which may result from decreased reliability and durability of switching devices or activation of overcurrent protections due to repeated (long-term) circuit overloads. Sudden changes in power demand, which generate transient states in the power supply system, are particularly critical for the on-board devices and systems.

3.3. Case Study

Figure 10 shows exemplary voltage waveform recorded during A/C startup from the GPU.

Typically, analyzing under- and overvoltage peaks involves calculating the value of the product of the non-standard time and voltages—specifically, the integral of voltage over time. Figure 11 provides an example of transient voltage analysis over time after a step load is applied to a DC power supply system.

According to Polish military standard [3], transient voltages are converted into their equivalent replacement peaks, which are rectangular DC voltage pulses at 28 V, matching the actual voltage pulses (Figure 11a, details: S₁, S₂—undervoltage peaks). The replacement peak pulse S_iz (S_iz2) takes the form of a rectangle with the same area and minimum voltage U_min1 (U_min2), but with a shorter duration (reduced time t_z) than the actual pulse S₁ (S₂)—see Figure 11b, details: S_iz1, S_iz2.

For the transient state shown in Figure 11a, the replacement peaks (areas S₁, S₂ below the lower reference level of 26 V) should be calculated according to Formulas (1) and (2).

$${\mathrm{S}}_1=\underset{0}{\overset{∆{\mathrm{t}}_{AC}}{\int }}\left[26-\mathrm{U}\left(\mathrm{t}\right)\right]\mathrm{d}\mathrm{t}$$

(1)

$${\mathrm{S}}_2=\underset{0}{\overset{∆{\mathrm{t}}_{CE}}{\int }}\left[26-\mathrm{U}\left(\mathrm{t}\right)\right]\mathrm{d}\mathrm{t}$$

(2)

The reduced time t_z of the replacement peaks should be calculated pursuant to Formula (3).

$$t_z={\displaystyle \frac{S}{29-U_{min}}}$$

(3)

The parameter determined during the transient state is the value of the product of non-standard time and voltage, which is converted into the so-called replacement peak S_iz according to Formulas (4) and (5).

$${\mathrm{S}}_{\mathrm{i}\mathrm{z}1}=26-{\mathrm{U}}_{\mathrm{m}\mathrm{i}\mathrm{n}1}·{\mathrm{t}}_{\mathrm{z}1}$$

(4)

$${\mathrm{S}}_{\mathrm{i}\mathrm{z}2}=26-{\mathrm{U}}_{\mathrm{m}\mathrm{i}\mathrm{n}2}·{\mathrm{t}}_{\mathrm{z}2}$$

(5)

These calculated pulses, shown as equivalent replacement peaks S_iz in Figure 11b (a rectangle with sides: height Umin and width t_z), are compared to the normative characteristics of the transient state, and their values should fall within the limits specified in the standard [3].

Based on the authors’ experience, it can be concluded that long-age operation of GPUs negatively affects the power quality from these sources. This is specifically true during their operation in transient states [11]. An example of the impact of GPUs’ age on changes in the parameters of DC power in a transient state is presented in Figure 12, which shows the visualization of test results for the DC power supply source of the Su-22 aircraft in a transient state over successive years in service.

In the subsequent years of the source’s operation, power quality parameters deteriorated until they reached non-compliant values. For this reason, older GPUs should be subjected to particularly careful verification of their technical condition, including the generated power quality, and especially the transient power parameters of the GPU.

Additionally, transient surge voltage loading of the GPU is an essential element for transient testing of 115/200 VAC GPU. Transient tests are performed by measuring the voltage AC at the end of the GPU dispensing cable during load changes, and are carried out analogously to the transient tests of a 28 VDC power circuit. However, in contrast to DC circuits, it is not the voltage waveform that is to be tested as a function of time, but the voltage envelope waveform. The time waveform u = f(t) is analyzed separately.

• Overvoltage peaks are above 118 V.
• Undervoltage peaks are below 112 V.

These peaks are converted into the corresponding equivalent peaks: overvoltage or undervoltage.

The peaks duration, on the other hand, is calculated by integrating the following:

• the area between the envelope of the actual waveform u = f(t) above the 118 V level
• the area between the envelope of the actual waveform u = f(t) below the 112 V level.

The resulting areas are divided by the amplitude of the peak, i.e.:

• The maximum value of the envelope of the real waveform u = f(t) is above the 118 V level.
• The minimum value of the real waveform envelope u = f(t) below the 112 V level.

The reduced value of the peak duration is thus obtained. The value of the equivalent peak is assimilated to the normative curve [3]. If at least one of the equivalent pulses of a given GPU channel crosses the normative curve, the test result is negative.

4. Discussion

4.1. Standards Analysis

Currently, a variety of GPUs are in operation in the PAF, with manufacturers of individual GPUs declaring compliance of their equipment with various standards: PN ISO6858:1997 [12], ISO6858:1982 [13], ISO6858:2017 [14], and MIL STD-704F w/CHANGE 1 [15], or only within the scope of selected parameters specified in the standards.

The analysis of the power quality requirements for GPUs according to NO 17 A206:2019 [3] and their comparison to other standards cited by GPU manufacturers showed their convergence to a large extent, but also showed existing differences, especially for transient testing.

When comparing the steady-state characteristics of the AC output (3 × 115 V voltage) specified in NO-17-A206:2019 [3], the following was found:

• Voltage—the range is the same for standards [12,13] and smaller than within the ranges defined in standard [14] and is larger than in standard [15];
• Phase displacement—the range is the same for standards [12,13] and smaller than within the ranges defined in standards [14,15];
• The crest factor—the range is the same for the compared standards [12,13,14,15];
• Total harmonic content—the range is the same for the compared standards [12,13,14,15];
• The divergence of corresponding ordinates from those of the equivalent sine—the requirement is the same for standards [12,13] and not specified in standards [14,15];
• AM phase voltage amplitude modulation—the range is smaller than for standards [12,13,14] and larger than specified in standard [15];
• Effective value of phase voltage envelope harmonic modulation depth—the range is within the ranges defined in standards [12,13,14,15];
• Frequencies—the range is the same for standards [12,13,14] and smaller for standard [15];
• Frequency drift—the range is the same for standards [12,13] and not specified in standards [14,15];
• Frequency modulation—the range is the same for standards [12,13,14] and larger than in standard [15].

When comparing the transient characteristics of the AC output (voltage 3 × 115 V) specified in NO-17-A206:2019 [3], for the following was found:

• Transient surge voltages—in the case of standards [12,13,14,15], both the ranges and the way of implementing the test (defining how to load the device under test) are different;
• Frequency transient—in the case of standards [12,13,14,15], both the ranges and the way of performing the test (determining how to load the device under test) are different.

When comparing the steady-state characteristics of the DC output (voltage 28 V) defined in NO-17-A206:2019 [3], for the following was found:

• Voltage—the range is the same for standards [12,13] and smaller than for [14,15];
• Voltage ripples:

  -

  the ripple of the DC supply maximum deviation from the average voltage level is smaller than in standards [12,13,14] and both different and larger than in standard [15], where the maximum amplitude of the ripple is specified

  -

  the root-mean-square (r.m.s.) values of the individual cyclic components of the ripple requirement is the same for standards [12,13,14,15];

When comparing the transient characteristics of the DC output (28 V voltage) specified in NO-17-A206:2019 [3], it was found that both the ranges and the implementation of the test are comparable for standards [12,13], while standards [14,15] differ in both the ranges and the implementation of the test.

In summary, the testing of GPUs will make it possible to determine their current state of repair in relation to the power quality requirements set out in NO 17 A206:2019 [3], and the planned comparison of the results with other standards will allow the identification of uniform requirements, or the possibility of comparing them and unambiguously interpreting the suitability of GPUs to power A/C in service in the PAF.

4.2. Proposed Tests

The general importance of norm specification and following have been mentioned, e.g., in [16]. The serviceability power supplied by the GPU to onboard A/C systems is determined by the quality requirements specified in standards [3]. By decision of the Minister of National Defence, No. 40/MON of 17 March 2020, Defence Standard NO-17-A206:2019 “Military Aircraft—Ground Power Supply Systems—Requirements” was introduced for implementation. This standard specifies, i.e., the requirements that must be met by the electrical power parameters at the end of the GPU’s output cable. Ensuring these parameters is a necessary condition for the proper operation of the A/C onboard systems. It is worth noting that the above-mentioned standard corresponds to the standard requirements commonly in force in NATO countries, meaning that GPUs compliant with this standard should guarantee their capability to service the A/C of the PAF.

In accordance with the NO-17-A206:2019 standard [3], testing the power quality of GPUs should be performed in steady and transient states for the following:

• Characteristics of the AC Power Supply System
• Characteristics of the DC Power Supply System

The authors suggest that to reliably assess the power quality in both steady and transient states, a specific set of tests is required. These tests, aligned with the NO-17-A206 standard [3] and within the GPU’s manufacturer-declared power capacity, are both necessary and sufficient. Importantly, they are designed to ensure the GPU operates safely without risk of overload or malfunction.

The scope of GPU testing for the DC output cable should include the following:

• For steady state at load levels of 0%, 5%, and 85% of rated power:
  • Steady-state voltage value
  • Amplitude of voltage ripple component
  • Frequency analysis of harmonic content in the ripple component
• For transient states during load changes from 5% to 85% and from 85% to 5% of rated power:
  • Overvoltage peak parameters
  • Undervoltage peak parameters

The scope of GPU Testing for the AC output cable should include the following:

• For a steady state at load levels of 0%, 5%, 85% of rated power:
  • Steady-state voltage value
  • Steady-state frequency
  • Harmonic content
  • Amplitude modulation depth of the voltage
  • Voltage asymmetry
  • Frequency drift
• For transient states during load changes from 5% to 85% and back from 85% to 5% of rated power:
  • Overvoltage peak parameters
  • Undervoltage peak parameters
  • Transient state frequency parameters

5. Methodology Verification

Power quality tests of GPUs performed by the authors are carried out on the test bench shown in Figure 13.

The GPU power quality test bench consists of a load set equipped with an aerial connector that allows the plug of the dispensing cable of the GPU under test to be connected. Depending on the type of the GPU under test, the test bench is equipped with a load set for 3 × 115 V AC voltages or a 28 V DC load set (Figure 13). An external 28 V DC power supply is connected to control the operating contactors of the load set. The control and measurement equipment are connected to the laboratory sockets of the load set.

The control and measurement equipment consists of the following:

• DC ammeter→2500 A, 2%
• AC ammeter→390 A, 2%
• Voltmeter→AC/DC, 400 Hz, 200 V, 1%
• Harmonic analyzer→THD, 250 V
• Power analyzer→3 phase, 2%
• Voltage recorder→8 channels DC/AC 250 V, 4 channels RMS 250 V, 20 ks/s.
• Oscilloscope→4 channels 250 V
• Frequency recorder→1 channel 400 Hz, 115 V

The comparative results presented below show how the proposed methodology effectively identifies the GPU with different lifetimes in relation to the power quality requirements of the standard [3]. The comparison presented is for a GPU with the same power rating operated for 4 years (No. 1) and for 20 years (No. 2). Values obtained that do not comply with the requirements of the standard [3] are written in bold.

Example test results for GPU No. 1 and No. 2 at the end of a 28 V DC dispatch cable in steady state obtained with rated currents of 0% I_z, 5% I_z, and 85% I_z are shown in

Table 4. Voltage, NO-17-A206: 2019 [3] p. 2.4.2.1.

No.	GPU Load	GPU No. 1	GPU No. 2	Required Value
		U [V]
1.	0% Iz	28.46	28.17	26 ÷ 29 V
2.	5% Iz	28.47	27.40
3.	85% Iz	28.51	24.05

,

Table 5. DC voltage ripple, NO-17-A206: 2019 [3] p. 2.4.2.2.

No.	GPU Load	GPU No. 1	GPU No. 2	Required Value	GPU No. 1	GPU No. 2
		Upmax [V]			5% of Mean Voltage [V]
1.	0% Iz	0.735	0.674	<5% of mean voltage	1.423	1.409
2.	5% Iz	1.091	0.998		1.424	1.370
3.	85% Iz	1.386	1.253		1.426	1.203

and

Table 6. Frequency analysis of harmonic content in the ripple component, NO-17-A206: 2019 [3] p. 2.4.2.3.

No.	GPU Load	GPU No. 1		GPU No. 2		Required Value
		Urms [V]	f [Hz]	5% of Mean Voltage [V]
1.	0% Iz	0.025	7.208	0.045	72.71	IAW Figure 12 [3]
2.	5% Iz	0.118	4799.73	0.136	2400.14
3.	85% Iz	0.082	4799.86	0.115	2402.28

.

The transient tests of GPU No. 1 and No. 2 were performed by loading the GPU device at the end of a 28 V DC dispensing cable with a current of 5% I_z and then pulse-loading the device with a current of 85% I_z. The 85% I_z current was held for about 5 s and the GPU load was reduced to 5% I_z.

Example results from testing GPU No. 1 and No. 2 at the transient end of the 28 V DC dispensing cable are shown in

Table 7. DC transient characteristics, NO-17-A206: 2019 [3] p. 2.4.3.1.

No.	GPU Load	GPU No. 1			GPU No. 2			Required Value
		U [V]	t [s]	tz [s]	U [V]	t [s]	tz [s]
1.	Overvoltage peak (5% ÷ 85%) Iz	20.60	0.0056	0.0034	22.87	3.460	2.1626	IAW Figure 15 [3]
2.	Undervoltage peak (5% ÷ 85%) Iz	33.60	0.0283	0.0168	No exceedance noticed

.

Example test results for GPU No. 1 and 2 at the end of a 3 × 115 V AC release cable in a steady state obtained with 0% I_z, 5% I_z, and 85% I_z rated current are shown in

Table 8. Voltage, NO-17-A206: 2019 [3] p. 2.3.2.1.

No.	GPU Load	GPU No. 1			GPU No. 2			Required Value
		U [V]
		A	B	C	A	B	C
1.	0% Iz	115.12	115.12	115.11	115.53	115.56	115.55	112 ÷ 118 V
2.	5% Iz	114.95	114.98	114.99	115.45	115.49	115.49
3.	85% Iz	114.30	114.74	115.01	114.95	116.20	116.49

,

Table 9. Voltage asymmetry, NO-17-A206: 2019 [3] p. 2.3.2.3.

No.	GPU Load	GPU No. 1			GPU No. 2			Required Value
		U [V]
		A-B	B-C	C-A	A-B	B-C	C-A
1.	0% Iz	0.00	0.01	0.00	0.03	0.00	0.02	<3 V
2.	5% Iz	0.02	0.00	0.03	0.03	0.00	0.04
3.	85% Iz	0.43	0.27	0.71	1.31	0.23	1.54

,

Table 10. Amplitude modulation depth of the voltage, NO-17-A206: 2019 [3] p. 2.3.2.4.

No.	GPU Load	GPU No. 1			GPU No. 2			Required Value
		The Crest Factor ka [-]
		A	B	C	A	B	C
1.	0% Iz	1.42	1.42	1.42	1.41	1.40	1.41	1.31 ÷ 1.51
2.	5% Iz	1.42	1.42	1.42	1.41	1.40	1.41
3.	85% Iz	1.41	1.41	1.42	1.40	1.39	1.39

,

Table 11. Harmonic content, NO-17-A206: 2019 [3] p. 2.3.2.4.

No.	GPU Load	GPU No. 1			GPU No. 2			Required Value
		THD [%]
		A	B	C	A	B	C
1.	0% Iz	0.57	0.57	0.57	2.00	2.00	2.00	<5%
2.	5% Iz	0.56	0.54	0.55	2.00	2.00	2.00
3.	85% Iz	0.70	0.78	0.65	5.1	5.3	4.7

,

Table 12. Frequency, NO-17-A206: 2019 [3] p. 2.3.2.6.

No.	GPU Load	GPU No. 1	GPU No. 2	Required Value
		f [Hz]	f [Hz]
1.	0% Iz	400.22	400.03	390 ÷ 410 Hz
2.	5% Iz	400.06	400.04
3.	85% Iz	400.07	399.98

and

Table 13. Frequency drift, NO-17-A206: 2019 [3] p. 2.3.2.7.

No.	GPU Load	GPU No. 1			GPU No. 2			Required Value
		Phase A
		Fwmin [Hz]	Fwmax [Hz]	δ [Hz/min]	Fwmin [Hz]	Fwmax [Hz]	δ [Hz/min]
1.	0% Iz	399.96	400.40	0.550	399.57	400.92	2.091	Frequency variation shall not exceed ±5 Hz and frequency drift shall not exceed 15 Hz/min
2.	5% Iz	399.75	400.25	0.369	399.49	401.46	10.74
3.	85% Iz	399.67	400.39	0.923	398.96	401.33	6.91

.

The GPU transient tests were carried out by loading the GPU at the end of a 3 × 115 V AC distribution cable with a current of 5% I_z, and then the device was pulse-loaded with a current of 85% I_z. The 85% I_z current was then held for approximately 5 s, and the GPU load was reduced to 5% I_z. Example results for GPU No. 1 and No. 2 at the end of a 3 × 115 V AC distribution cable transient are shown in

Table 14. AC transient characteristics, NO-17-A206: 2019 [3] p. 2.3.3.1.

No.	GPU Load	GPU No. 1			GPU No. 2			Required Value
		U [V]	t [s]	tz [s]	U [V]	t [s]	tz [s]
1.	Overvoltage peak (5% ÷ 85%) Iz	No exceedance noticed			110.6	0.224	0.120	IAW Figure 7 [3]
2.	Undervoltage peak (5% ÷ 85%) Iz	No exceedance noticed			123.4	0.224	0.069

and

Table 15. Transient state frequency, NO-17-A206: 2019 [3] p. 2.3.3.2.

No.	GPU Load	GPU No. 1		GPU No. 2		Required Value
		U [V]	t [s]	U [V]	t [s]
1.	Overvoltage peak (5% ÷ 85%) Iz	No exceedance noticed		372.97	0.5788	IAW Figure 10 [3]
2.	Undervoltage peak (5% ÷ 85%) Iz	No exceedance noticed		425.58	0.3057

. Non-conforming values are written in bold.

6. Conclusions

The power quality supplied by the GPU has a significant impact on reliability, costs, and the operation safety of A/C.

The GPUs operated by the PAF are highly diverse in terms of type and age. It is essential to identify individual GPUs and assess their current technical condition with regard to the normative requirements for power quality. It should be underlined that the military units operating aircraft do not possess tools for automated, comprehensive assessment (in steady and transient states) of the power quality supplied by the GPUs. The airport control devices LUK-1 (DC) and LUK-2 (AC), developed at AFIT in the 1990s, which enabled comprehensive diagnostics of GPUs in steady and transient states, have been withdrawn from service. Therefore, it is necessary to establish a unified system for monitoring the power quality of GPUs in the PAF in accordance with the NO-17-A206:2019 [3] standard.

Due to this, AFIT has initiated work that will ultimately create the possibility of developing airport control devices that enable automatic measurements and serve as the basis for implementing a system for a comprehensive assessment of the power quality supplied by GPUs. The implementation of these devices into operation in the PAF, along with the training of technicians to use them during regular airport operations, will enable ongoing monitoring of the power quality supplied by GPUs. This will enhance the operational readiness of GPUs supporting aircraft in the PAF, resulting in greater reliability and safety in operations.