1. Introduction
The current-carrying capacity of power cables is an important operational parameter of both transmission and distribution networks. The current-carrying capacity is a measure of the power transmission of power lines, including low-voltage cable lines directly connected to electricity consumers. This capacity informs about the maximum value of the electric current flowing through the power cables, which will not cause thermal damage to the insulation as a result of exceeding the permissible temperature for a long time. The continuous increase in the demand for electricity and the increasing number of generation sources integrated with the power system, as part of renewable energy sources, result in the growing interest in the possibilities of increasing the load of power cable lines. The current-carrying capacity of power cables depends mainly on: the construction of the power cables and their arrangement and environmental conditions. The influence of each of these factors on the current-carrying capacity of conductors is defined in the IEC [
1,
2,
3,
4] and IEEE [
5] standards. These standards take into account individual factors selectively and do not refer to all phenomena actually occurring in power installations. The indicated standards do not take into account, among others, the influence of: variable wind speed and direction, variable insolation, dry zones around underground power cables, as well as many other factors affecting the current-carrying capacity of conductors. The indicated standards are based on the Neher–McGrath calculation method from 1957 [
6]. However, this method is not very precise [
7]. Therefore, reliable knowledge of the current-carrying capacity can be obtained through experimental tests and numerical analysis. As reported in [
8,
9,
10], numerical calculations of thermal states of power cables with the use of dedicated software are becoming more and more important, and the provisions of the above-mentioned standards are only used to confirm the results of simplified numerical modeling. Knowledge of the actual operating conditions of power cables allows to avoid design errors and, as a result, will increase the reliability and safety of power installations [
11,
12].
The temperature of the wiring elements has a significant impact on their durability, especially their thermal endurance [
13]. The authors of the referred article pointed out that there is a real risk of the wires overheating in the air, especially in the presence of direct sunlight. The authors of the paper [
12] indicated the important role of the color of the roofing and the distance of the electrical installation from the roof surface. However, the impact of the indicated factors in the context of photovoltaic (PV) installations was not analyzed. The paper [
14] pointed to a significant problem related to the destruction of electrical equipment (including power cables and connectors). The described destructive phenomena did not refer to the contact resistance of materials. In the article [
15], it is indicated that, based on operational experience, 70% of power line failures occur in places where they are connected. This is the effect of the contact resistance and its increasing value as a result of the phenomena occurring in the contact structure, as well as the influence of external forces. In addition, a damaged electrical contact point most often causes overheating of the connector, as well as short circuits and fire. In addition, temperature is a reliable and important indicator in assessing the condition of a cable joint. In turn, the article [
16] proposed a real-time temperature monitoring system for power cable joint. The proposed solution to the problem of heating the joints is suitable only for special applications. To emphasize the importance of the problem of heating up the cable connection points, GPRS wireless communication technology was also proposed [
17].
Problems with the DC joints/connectors are the main cause of failure as well as fire risk in PV systems [
18,
19]. It is estimated that nearly 70% of all fires in PV installations are the result of improper connections in DC circuits [
18]. From a scientific point of view, a potential cause of fire in PV systems due to the DC connector is the increase in its contact resistance. An equally important issue related to the fire protection of PV installations is the occurrence of possible arc faults, very often caused by bad connections. The authors of the paper [
20] indicate the types and characteristics of arc faults in PV installations but do not indicate the methods of detecting such hazards. The paper [
21] indicates the types of short circuits occurring in PV installations based on practical events in PV farms. The authors of the paper point out the special role of protections integrated with inverters (protection against arc faults). The article [
22] provides information on the occurrence and characteristics of earth fault currents on the DC side and the related possible fire hazard of PV installations.
This article presents the influence of unfavourable environmental conditions (insolation, low wind speed) on the temperature distribution of power cables and cable connectors used in typical PV installations. The effect of contact resistance of the connections on their temperature is also investigated. The above-mentioned issues are not deeply analysed in the literature. Results of the authors’ investigation show the points of electrical installations which may be a cause of the possible fire hazard as well as electric shock. Indicating these points is a key issue to prevent fires in new installations.
The rest of the paper is organized as follows.
Section 2 presents an analysis of the influence of solar radiation on the current-carrying capacity and temperature of AC power cables. In
Section 3, an analysis of the heating of DC connectors in theoretical and practical terms is provided.
Section 4 summarizes the obtained results.
2. The Impact of Solar Radiation on the Temperature Distribution and Current-Carrying Capacity of AC Power Cables
In order to present the impact of unfavorable ambient conditions on the long-term current-carrying capacity (and thus on the temperature distribution) of power cables, a low-voltage power cable line was tested by computer simulations. The simplified model presented in
Figure 1 was implemented in a 3-D numerical model in the ANSYS Fluent software.
Figure 1 shows a power cable line laid vertically in air at a temperature of 25 °C. The given air temperature corresponds to Polish climatic conditions in accordance with the standard [
3]. Three power cables (three-phase alternating current line of nominal frequency 50 Hz) with a cross-sectional area of 35 mm
2 with a conductor made of copper and insulation and a sheath made of PVC are arranged with a spacing equal to the diameter of the power cable.
In the numerical model, all heat exchange mechanisms involved in the real system were introduced:
In order to investigate the effect of the real conditions of cables heating, the heat of solar radiation falling at an angle of 45° on the cable surface (conditions typical for Europe), convective air movement and a radiation model according to the discrete ordinates theory were modeled. An example of a computational mesh in a cross section is shown in
Figure 2.
The total number of mesh elements was approximately 9 million finite volumes. The results of particular calculation variants are presented in the further part of the article. A steady-state simulation was performed and turbulence model k-epsilon was chosen. The proper materials were set in the model in certain cell zones and the boundary conditions were as follows:
mass flow inlet (bottom side of domain),
pressure outlet (upper side of domain),
mesh interfaces between different cell zones,
heatflux (the inner side of power cables),
heatflux (external side of power cables, which is exposed to solar radiation).
In the case of heating the cable line in accordance with the presented assumptions and in the absence of electric current flow, the temperature distribution around the cables was obtained as shown in
Figure 3.
In the analyzed model,
Figure 3 shows the effect of solar radiation alone (without current load) on the temperature of the power cable especially in the L3 phase (cable on the right). Taking into account the solar radiation and the lack of current load on the cables, the cable insulation temperature in the L3 phase reaches 50 °C. Thus, the solar radiation increased the temperature of the power cable insulation by 25 °C (from ambient temperature equal to 25 °C).
Figure 4 shows the temperature distribution around power cables which are symmetrically loaded and unaffected by solar radiation. The current-carrying capacity of the cable system is equal to 152 A. The current value was achieved by iteratively increasing the Joule’s heat in the conductor until the permissible temperature (70 °C for PVC insulation) was reached.
Figure 5 shows the temperature distribution around power cables which are symmetrically loaded and under the influence of solar radiation. The current-carrying capacity of each power cable is 96 A. The current value was determined on the basis of computer calculations—in the assumed conditions it obtains the max temperature of the cable insulation equal to 70 °C (max permissible).
Comparing the simulation results presented in
Figure 4 and
Figure 5, one can notice a significant decrease in the current-carrying capacity of power cables due to the impact of solar radiation. As a result of the influence of the heat of solar radiation, the current-carrying capacity of power cables decreased by nearly 37%. If the need to reduce the load of the power cables is not taken into account (the cables would be loaded with 152 A instead of 96 A), the permissible temperature could be locally exceeded, which results in a reduction in insulation durability [
13]. Significant overheating of the cables and damage to the insulation may also cause significant leakage current and, consequently, risk of electric shock or fire.
3. The Impact of Solar Radiation and Quality of Connections on Temperature Distribution in DC Connectors
Several simulations were carried out to assess the effect of insolation and contact resistance on the temperature of DC connectors used in PV systems. For this purpose, the DC connector, commonly known as the MC4 connector, was the subject of the analysis. The connector model shown in
Figure 6 has been performed in accordance with the manufacturer’s catalog card [
23]. The numerical approach was very similar to the one presented for the power cables in previous section. The only significant difference is that the “Solar calculator” option was set in the model instead of implementing the additional heatflux zone to simulate solar radiation. Moreover, we only consider DC flow. The computational model includes a constant value of solar radiation intensity (1200 W/m
2), an ambient temperature of 30 °C (assumed by the authors’ free air temperature during a hot day in Polish climatic conditions), and low convection air movement (0.4 m/s). The parameters of the model are to reflect unfavorable heat exchange conditions in the summer months on the roof [
24]. Other important simulation parameters are: connector material (polyamide), absorptivity material (0.9), and DC wire resistance (6.1 mΩ/m). It was assumed that the contact resistance between the connectors and the connected wires (resulting from the quality of cable assembly) is neglected.
In the first stage of the investigation, the effect of solar radiation on the heating of the MC4 connector, without the flow of electric direct current, was examined.
Figure 7 shows the heating effect of the DC connector only under the influence of solar radiation. According to the obtained simulations, the DC connector reaches a maximum temperature of 55 °C.
Then, the conditions of heat exchange with the environment with the participation of solar radiation and nominal operating conditions were examined (
Figure 8).
The distribution of temperature presented in
Figure 8 shows relatively good conditions for heat dissipation to the environment. Despite the flow of electric current with an assumed typical value for the connector (15 A), the increase in the temperature of the connector is recorded in the range of 6 degrees Celsius compared with no load (
Figure 7). It is worth noting that the result of the temperature distribution in
Figure 8 was obtained for the contact resistance respected by the manufacturer (max. 0.25 mΩ) and the rated contact force of 50 N.
As mentioned in the first part of the article, a significant fire hazard from a DC connector can be caused by an increase in its contact resistance above the values allowed by the manufacturer. To investigate this effect, possible contact resistance values were determined depending on the contact force, based on the following equation:
where:
Rp—contact resistance, mΩ,
C— constant dependent on the contact area, -,
ρ—contact material resistivity, Ωm,
Fk—contact force, N,
m—constant depending on the type of contact, -.
According to the information contained in [
25], for the contact material of which the DC connector is made (tin-plated copper), the value of the
c·ρ coefficient (the product of constant dependent on the contact area and contact material resistivity) is equal to 1. The value of the constant
m is equal to 1 (the contact of the planes). The list of calculated contact resistance values
Rp depending on the contact force
Fk is provided in
Table 1.
The results contained in
Table 1 show the value of the calculated contact resistance depending on the decreasing value of the contact force (values lower than the nominal contact force).
In the analyzed DC circuit, the current value decreases with increasing resistance—this phenomenon is taken into account for values included in
Table 1. Importantly, the results presented in the first and second cases of
Table 1 are consistent with the manufacturer’s data. The reduced contact forces shown in
Table 1 are the forces that actually occur in PV installations due to poor wiring. The presented Equation (1) assumes the averaged values of the contact coating resistance. It should be emphasized that in the case of favorable corrosion conditions of the connectors (especially moisture ingress), the coating resistance value is significantly increased.
Figure 9,
Figure 10 and
Figure 11 show the temperature distribution around the DC connector, for various values of the contact resistance and the flowing electric current (cases 6, 7, and 8 from
Table 1).
As the contact resistance increases, the maximum DC connector temperature also increases. It is worth emphasizing that, according to the manufacturer’s card, the upper temperature limit of the connector is equal to 105 °C. Therefore, in the cases shown in
Figure 9,
Figure 10 and
Figure 11, this temperature has been exceeded, which results in a reduction in the durability of the connector, its possible change in structure, and possible loss of insulating properties. A summary of the calculation results of the connector heating, depending on the contact resistance, is shown in
Figure 12.
The graph presented in
Figure 12 refers to the data provided in
Table 1. The temperature of the DC connector increases with the increase in contact resistance
Rp. A significant increase in temperature is observed, in particular, for the initial increases in contact resistance (up to 1 Ω). The presented effects of excessive heating of the DC connector constitute a real fire hazard due to the possible generation of minimal ignition energy for the dust–air mixture or other diffusing mixtures present in the vicinity of the PV installation [
26]. In a typical PV system, there are many DC connection points made by a set of connectors (plug-socket). If one of these connections is poorly made, it will deteriorate during operation as a result of increased contact resistance: this place heats up when current flows and a “hot spot” occurs. Due to the appearance of heat, the contact material may diffuse or even melt until, at some point, the connection will be completely broken. In this case, an arc can form over a very small air gap, resulting in a fire hazard [
27]. The described occurrence of hot spots of connectors is not only a theoretical problem—it occurred in practice, as shown in
Figure 13.
The DC connector on the left in the photo (
Figure 13) is new and unused. The connector on the right in the photo comes from a real PV installation, where the contact resistance increased above the nominal value as a result of the power cables being stretched during power generation. Finally, part of the insulation of the connector melted, which can be seen in
Figure 13. This effect can also cause a fire hazard.