A Computational and Experimental Method for Determining the Current in the Braid of a Control Cable During a Short Circuit

Abstract

Non-equipotentiality in a grounding device can cause thermal heating in the screens of control cables that are grounded on both sides of high-voltage substations. At the same time, there is currently no approach for assessing the thermal endurance of cable screens that takes into account the configuration of the grounding device, the properties of the ground, and the connection. This paper presents a methodology for the experimental and computational determination of the thermal endurance of control cable shields in secondary circuits of 220–500 kV substations under short-circuit (SC) conditions. The method is based on full-scale imitation experiments using a sinusoidal current generator and verified numerical modeling in the ORU-M software. The potential and current density distribution in the cable shields were determined. The results showed that current densities in some circuits exceed permissible levels, confirming the risk of thermal damage. It was found that reconfiguring the grounding system—by densifying ground electrodes and increasing connections between grounding points—can reduce current density to acceptable values. The presented method allows for reliable assessment of the thermal endurance of cable shields without decommissioning the substation, making it suitable for the design and modernization of high-voltage facilities.

1. Introduction

1.1. Background of the Study

Reliable operation of an electric power system is largely determined by the failure-free operation of relay protection and automation (RPA) and compliance with the requirements of electromagnetic compatibility (EMC) for the entire complex of secondary equipment—control, measurement, and process protection systems operating in electric networks of different voltage classes. In order to protect secondary equipment and circuits, shielded cables grounded on both sides are used. However, this type of protection from external EM interference increases the probability of thermal damage due to the occurrence of a return path between the grounding circuit and the shield of the control cable. Thus, the noise immunity and thermal resistance of shields of control cables directly affect the survivability of RPA and the stability of the power system in transient and emergency modes. At the design stage, it is necessary to check cables for thermal resistance.

Emergency events may occur in electric power systems (EPSs), leading to mass load shedding and significant socio-economic losses. Short circuits are among the most frequent initiating factors of major technological incidents [1]. At SC currents, a significant rise in the potential of grounding devices is possible, which causes equalizing currents on the screens of control cables and can lead to their overheating, insulation degradation, and failure of secondary circuits.

For example, in 2023, an accident occurred in Astana in October. The destruction of the busbar disconnector (BD) header of a bypass bus led to the fall of the blade to the ground and the development of a single-phase SC. The resulting currents and the associated potential rise in grounding circuits caused thermal damage to the control cable and voltage sampling cabinet. The disruption of the secondary circuits was accompanied by the disconnection of two 220 kV lines supplying ACHP-2; the power of the plant was reduced from approximately 530 MW to 120 MW, leaving a significant part of the city without power supply for several hours [2]. This case clearly demonstrated how a local error in estimating the allowable thermal loads on the control cable shield can trigger a cascade of failures evolving into a system blackout.

Particular attention must be paid to the thermal resistance of control cables in nuclear power plants. For example, the accident in America at Browns Ferry NPP (USA, 1975), where a fire in the cable duct led to a significant loss of power generation, as well as significant direct and indirect economic losses, and recovery work took 19 months [3].

The importance of thermal effects in building electrical cables and their possible hazards has been emphasized in [4], showing how overheating can degrade safety and performance. While these studies focus on installations in buildings, the same principle is critical for control cable braids in substations, where excessive SC currents can produce comparable thermal stresses. Cable thermal behavior has been widely studied across operating and fault scenarios, with the finite element method (FEM) used to model coupled electromechanical–thermal effects and their impact.

Bonding of metallic shields in power cables is regulated by [5], which sets out objectives and configurations (single-point, cross-bonding, and impedance bonding) to control induced voltages and currents, coordinate with sheath-voltage limiters, and manage losses in normal or quasi-steady conditions. However, it focuses on power cable circuits and does not specify methods for the thermal stability of control cable shields under fault conditions in high-voltage substations. In modern substations, where control cables are grounded at both ends and routed between regions that can rise to different potentials during a fault, this omission creates a practical design risk; equalizing currents can flow through the screens, potentially exceeding short-time thermal limits.

Standards [6,7] regulate general grounding and bonding practice—covering system grounding objectives, step and touch safety limits, soil-resistivity measurements, and ground grid design for substations. They do not include regulations for the thermal stability of both end-grounded control cables. Likewise, [8] addresses installation practice and EMC-oriented earthing, but it does not provide a standardized, practical, measurement-validated method for designers to verify the thermal admissibility of control cable shields during faults.

Conversely, Ref. [9] standardizes how to calculate thermally permissible SC currents for cable conductors, screens, sheaths, and wire braids, including non-adiabatic corrections via material-dependent coefficients. It provides tabulated constants and analytical formulas and notes that non-adiabatic treatment can materially increase permissible SC current, especially for screens and sheaths, while adiabatic assumptions remain acceptable for a very short time. However, it does not specify how to determine the actual SC current flowing in a given control cable shield in a substation, which depends on grounding topology, bonding and soil parameters, and mutual impedances among parallel return paths.

Thus, to the best of our knowledge, no standardized, practical method or measurement-validated practice exists for designers to verify and ensure the thermal admissibility of control cable shields during SC events in high-voltage substations; our study proposes and validates such a method and mitigation.

1.2. Related Work

The thermal endurance of control cables shields at SC faults can be assessed by means of standardized calculation formulas, modeling the use of the finite element method, as well as field experiments. Each of these methods has its advantages and limitations, which affect their practical applicability in engineering work.

In thermal endurance calculations, according to the standard in [9,10], heating is considered adiabatic; i.e., the heat remains inside the conductor for the entire short circuit time.

Thus, according to [9], heating in the shield of the control cable is determined by the following expression:

$${\theta }_f=\sqrt{{\left(\beta +{\theta }_i\right)}^2+{\displaystyle \frac{2·U^2·\tau ·(\beta +20)}{\epsilon {\left(\tau \right)}^2·L·{\sigma }_c·{\rho }_{20}}}}+\beta$$

(1)

where ε(τ)is the coefficient of heat losses into neighboring elements; θ_f and θ_i are the final and initial temperatures, °C; σ_c is the specific volumetric heat capacity of the conductive element at 20 °C, ${\displaystyle \frac{\mathrm{J}}{\mathrm{K}·{\mathrm{m}}^3}}$; τ is the current flow time, s; ρ₂₀ is the resistivity of the screen material at 20 °C, Ω·m; β is the inverse of the temperature coefficient of resistance of the conductive element at 0 °C, K; U is the voltage between the ends of the cable screen, V; and L is the length of the control cable, m.

Transient cable heating has been approached through analytical computation and thermal equivalent circuit modeling [11]. The temperature limits of the insulation ultimately limit the permissible current. Thus, material thermal properties are central, while installation and the environment can change achievable current and losses [12,13]. Much of the literature uses FEM, with comparatively few studies that pair measurements and calculations [14,15,16,17].

In [15], the calculations of steady-state current-carrying capacity based on IEC 60287 and the FEM model are compared. The author shows that for single-core cables laid in trefoil or flat formation, IEC and FEM calculations provide almost the same results. However, in the case of a group of cable ducts with multiple lines, the FEM results show stronger electromagnetic and thermal interference than predicted by the IEC standard. Thus, it can be concluded that computational methods may not take into account characteristic factors such as cable location in the conduits, contact resistances, grounding device configuration, etc.

A more accurate methodology is the use of finite element methods, which allow us to simultaneously take into account the shape of a screen, as well as the resistances of individual connections. The authors of [16] conducted a comparative analysis of the thermal resistance of OPGW cables in a two-dimensional (2D) model to analyze thermal losses in cables.

In [17], the authors extended the previously experimentally validated model of an underground cable line to include connections and the environment to account for possible defects in coupling fabrication or contact weakening during operation. The study considers the thermal effects of earth fault currents flowing through the shields of medium-voltage cables and their couplings. Using a 3D model, contact resistances at the joints due to installation defects or the loosening of clamps are taken into account. The simulation includes realistic scenarios with repeated automatic switching, typical of distribution networks. The results confirm that such factors significantly increase the thermal load on cable shields and require a comprehensive analysis similar to the one used in this study.

Nonetheless, it is labor-intensive to perform such calculations at the substation; the need to take into account the configuration of the substation, as well as the topology of control cable routing, complicates the calculations.

Direct full-scale tests provide the most reliable results. In [18], field experiments were conducted by conducting artificial single-phase SC. During the experiment, the authors measured the potential difference in the grounding grid and proposed a way to reduce the potentials in the secondary cables by densifying the grounding. Field research on the effect of applied voltage on the thermal resistance of cable insulation, conducted in [19], showed the relationship between the applied voltage level and cable insulation ignition. According to the results, at levels above 140 V AC, a stable arc breakdown occurs, significantly increasing the risk of ignition of thermally weakened cables. At voltages above 280 V, arcing processes are accompanied by instantaneous protection activation and ignition.

However, such full-scale experiments to determine the thermal stability of the screens of control cables in substations are accompanied by the withdrawal of equipment from operation and the risk of damage to the insulation. Practical implementation is associated with the risk of equipment damage, long preparation (taking sections out of operation), and high costs for downtime in the network during the experiment.

This methodology provides sufficient efficiency for the practical work of the designer, minimal risk for the equipment, and the verifiable mathematical reliability of the results, which makes it the preferred tool for computational verification of the thermally permissible SC currents of the shields of control cables.

1.3. Novel Contributions

In this paper, we propose a reliable engineering approach to assessing the thermally permissible SC currents of shields of control cables, combining the field verification of the grounding network with detailed software modeling of the distribution of currents in the grounding system at SC. Our approach in this work is to overcome each method’s limitations by combining a controlled experimental injection with computer modeling. We perform a low-power simulated fault on the grounding system to gather measurements and then use a validated model to extrapolate the results to worst-case fault scenarios. This hybrid methodology provides reliable results while minimizing risk and cost. It effectively bridges the gap between purely theoretical analysis and hazardous full-scale testing.

The scientific novelty of the study is as follows:

• This paper introduces, to our knowledge, the measurement-validated method for estimating control cable shields’ thermal stability in HV substations during faults that combines a full-scale field experiment with a calibrated digital model, one that can predict real fault behavior, including substation characteristics (soil, grid, transformer neutral, shields, etc.). The proposed method is already in use by Kazakhstan’s national grid company (Astana, Kazakhstan).
• We show the potential of soils typical of the Republic of Kazakhstan to provide thermally permissible SC currents for control cables by reducing the potential gradient on the grounding system of the substation.

This paper is organized as follows: Section 1 describes the background of the paper, related prior works, and the novelty of our study. Section 2 details the experimental setup, measurement procedures, modeling techniques, and the thermal failure criterion used for cable shields. Section 3 and Section 4 present the model validation, the effect of bonding configuration, and the thermal limit evaluation, with comparisons to permissible levels. Section 5 concludes the paper with a summary of findings and specific recommendations for substation design and cable shield protection.

2. Materials and Methods

2.1. Permissible Level and Criteria of Thermal Endurances of Control Cables

To reduce the level of electromagnetic interference, shielded control cables are used, which are grounded on both sides (Figure 1). However, there is a subsequent problem associated with ensuring the thermal resistance of the shield of the control cable. This problem arises due to the non-equipotentiality of the SC protection device.

During a short circuit, the current flowing through the equipment causes an increase in the potential at the equipment grounding point. Simultaneously, the potential on the shields of the control cables also rises, with the magnitude of this increase depending on the length of each cable. As a result, certain current flows through the shield, which can compromise the thermal endurance of the cables. The thermal stability of the control cables is defined as the maximum current flowing through the shield at which the cable does not heat above the permissible temperature values according to Equation (2).

$$I_{screen}<I_{perm}$$

(2)

where $I_{screen}$ is the current flowing through the screen at SC, A; $I_{perm}$ is the permissible current, A.

This study will examine the thermal resistance of control cables during single-phase short circuits occurring in control system circuits due to potential differences in screens through inductive coupling and coupling through total resistance.

2.2. Experimental Setup for Simulated Ground Fault

The standard in [20] focuses on measuring and verifying the performance of grounding systems, including techniques to inject test currents and measure ground impedance and surface potentials. Our work draws on these standards by injecting a test current and measuring the shield voltage, essentially performing an actual fault current path through the substation’s grounding network and cable shields.

To determine the current distribution across the grounding device and the shield of the control cable, experiments are conducted to simulate a single-phase ground fault using a sine-wave generator (SWG) that simulates the fault current. Figure 2 schematically illustrates the experiment at the substation.

One output of the SWG is connected on one side to the grounding point of the power equipment to which the test cables are connected, and on the other side to the neutral of the power transformer or to an external electrode. The flow of SC current through the grounding device outside the substation (the SC current component of the power system) and through the neutral of the transformer is simulated.

By driving current from the yard into the neutral/remote ground, we imitate the two main return paths of a real line-to-ground fault: (1) current spreading out through the earth grid and exiting the substation grounding system and (2) current returning through the transformer neutral grounding point. In effect, the SWG current splits between the earth path and the metallic return path, reproducing the conditions of a genuine fault without the destructive high magnitude.

During the injection, we measured both the voltage and current distribution in the system using selective instruments to avoid interference from the power frequency background. A selective voltmeter is used to measure the voltage between the SC simulation point and the grounding point of the control cable screens in the relay protection panels in the control room (CR) or substation control building (SCB), taking into account background interference. This measured voltage is essentially the ground potential rise (GPR) that will drive current through the cable shields and any other parallel paths. With the help of selective current clamps, the components of the SC current are measured for various equipment elements: the current flowing in the cable shield, the current in the grounding grid at various locations, and the current in the transformer neutral connection.

The magnitude of the test current was chosen to be large enough to produce measurable effects yet small enough to avoid any harm. In our tests, the SWG injected a sinusoidal current on the order of tens of amperes (at 50 Hz). This level is safe for equipment and did not appreciably heat the conductors during the short test duration, but it produces voltages of a few volts to a few tens of volts in the grounding system that can be accurately measured. We verified that this injection current and resultant voltages were well below any hazardous thresholds. Then, we linearly scaled up the observations to represent a real fault. This scaling is valid because the system’s response is essentially linear in the low-frequency steady state.

2.3. Numerical Modeling and Calibration

In parallel with the field experiment, we constructed a detailed model of the substation’s grounding network and cable system using the ORU-M software (version 2.3.2) (a specialized tool for modeling multi-level grounding systems). The model included the mesh of grounding conductors (ground grid and vertical rods) represented as a network of interconnected segments, the soil model (with an empirically determined resistivity profile), and the control cable shields. The control cables in this substation use a braided or ribbon copper shield (often a few mm² of cross-sectional area); in the model, we represented each cable shield as an equivalent conductor running from the control building to the yard equipment connected to the ground at both ends.

For modeling convenience, we set the shield as a cylindrical conductor of 1 mm² cross-section copper, which is topologically equivalent to the actual braided shield tape. The choice of 1 mm² is arbitrary for simulation; since we are interested in current distribution and current density, it is easy to scale the results to the actual cross-sectional area of each cable’s shield.

Due to the fact that substations use a variety of control cables with different cross-sections, the current density flowing through the shield of the control cables was used as a parameter of thermal stability.

The program calculates currents and potential distributions in similar locations and compares the results with measurements obtained at the substation. If there is a discrepancy between the results, changes are made to the calculation model in the program. If necessary, additional clarifying measurements are taken at the survey site. If the model is reliable, the distribution of currents and potentials during a short circuit at various points of the substation is calculated.

The data obtained on the currents flowing through the screen of the control cables is compared with the permissible value of the thermal stability of the control cables according to Formula (3).

$$J_{perm}={\displaystyle \frac{0.8}{\rho }}·\sqrt[3]{{\displaystyle \frac{{∆\theta }^2}{49\tau }}}={\displaystyle \frac{312}{\sqrt[3]{\tau }}}=470{\displaystyle \frac{\mathrm{A}}{{\mathrm{m}\mathrm{m}}^2}}$$

(3)

where ∆θ is the permissible overheating of the control cable, NYCY, ∆θ = 130 °C; τ is the current flow time, s; and ρ is the resistivity of the screen material at 30 °C, Ω·m.

The ambient temperature is taken as the initial temperature, equal to 30 °C. At the TsGPP substation, NYCY brand (PVC insulation) cables are used as control cables, and according to their specifications, the permissible cable temperature in the case of an SC is θ = 160 °C. The response time of the backup protection was taken as the time of the SC current flowing through the control cable screens.

In our case, we considered a fault time as 0.3 s (backup protection time), which is the worst-case scenario. Thus, about 470 A/mm² was the benchmark.

The critical feature of our approach is that the model is built and calibrated on real field-experiment data. By matching the measured ground-potential rise from the field experiments, the model becomes a digital copy of the grounding of the substation based on experiments.

A critical aspect of the model is capturing the return-path network and mutual impedances. A grounding system offers multiple parallel return paths for fault current: through the soil, through any metallic connections like cable shields or cable armor, through structural steelwork, and through the neutral conductor of transformers. These paths are not independent; there are electromagnetic couplings between them. To account for this, our model uses a full electromagnetic computation in which all conductors are represented in their physical positions, and the mutual coupling is calculated. This approach is similar to methods used in FEM simulations, where the self- and mutual impedances of each segment are derived from geometric and material data. By doing so, the model solves for how much current goes through the soil and through the cable shields based on their impedances. Initially, the model’s soil parameters (two-layer soil resistivity) were estimated from ground-resistivity measurements at the substation.

In summary, our methodology involved (1) a field injection test to measure how fault current flows in a real substation grounding and cable system, (2) a detailed simulation model tuned to those measurements, and (3) to judge the acceptability of the fault currents for the cable shields. This approach provides both a diagnosis (identifying problematic places) and a solution space (testing improvements in the model) in a way that adheres to modern standards and safety practices.

3. Results

Based on the results of the survey conducted at KEGOC JSC’s 220–500 kV substations, a model of the grounding devices was created in the ORU-M grounding device parameter simulation environment (see Figure 3).

In the model, grounding conductors are represented as solid-steel cylindrical conductors with a diameter of 14 mm. At substations, control cable screens are the ribbon type, so in the calculation model, the screen was modeled as a copper cylindrical conductor.

The reliability of the model was confirmed by conducting field simulation experiments directly at the substations (see Figure 4).

The results obtained were recalculated for real SC current. At the same time, simulations of real SC currents at frequency of 50 Hz were performed on the design model of the grounding system at the same locations as in the simulation experiments.

Thus, the following data were obtained, for example, for the 500 kV substation of the TsGPP based on simulation experiments and modeling results (see

Table 1. Comparison of field experiment results and the calculated model of the grounding device.

Cable Route	Point of Application	Maximum Voltage on the Cable (Experiment), kV	Maximum Voltage on the Cable (Model), kV	Permissible Voltage Level on the Cable, kV
OS-500 kV
OS-SCB	CB-2	1.129 ± 0.11	1.220 ± 0.06	2
OS-220 kV
OS-SCB	CB-1	1.880 ± 0.2	1.96 ± 0.1	2
OS-110 kV
OS-CR	TD-9	1.484 ± 0.15	1.44 ± 0.07	2

).

The SC current parameters for each outdoor switchgear (OS) are different (see

Table 2. Routes with shield current density exceeding the permissible level (t = 0.3 s).

Cable Route	Point of Application	The Highest Current Density in the Screen by Experiment, A/mm2	Permissible Level of Current Density in the Screen, A/mm2	Conclusion
500 kV substation “TsGPP”
OS-220 kV (current SC I = 20.54 kA, neutral currents AT-1 I = 5.94 kA, AT-2 I = 5.95 kA)
OS—SCB	CB-16	570	470	Reconstruction of the grounding network
OS-110 kV (current SC I = 26.36 kA, neutral currents AT-1 I = 5.44 kA, AT-2 I = 5.45 kA)
OS—CR	CB-3	680	470	Reconstruction of the grounding network
	BD-1-2	670
	CB-5	1180
	BD-1-7	1160
500 kV substation “Shymkent”
OS-220 kV (current SC I = 22.45 kA, neutral currents AT-1 I = 4.45 kA, AT-3 I = 3.65 kA)
OS-CR-220	BD-1	656	470	Reconstruction of the grounding network
	LD-1	650
	CB-3	515
	BD-5-2	540
220 kV substation “No.7”
OS-110 kV (current SC I = 36.41 kA, neutral currents AT-1 I = 6.92 kA, AT-2 I = 6.92 kA)
OS-CR-110	CB-433	590	470	Reconstruction of the grounding network
	CB-132	620
	BD-132	530
	CB-157	620

) and correspond to the single-phase SC currents and their components provided by the relay protection and automation service of each substation.

Error bars in Table 1 reflect the uncertainty from two sources: the selective voltmeter used to measure the ground potential rise (instrument accuracy ~10%) and the model error (5%).

Table 2 shows the results in control cable circuits where the current density exceeded the permissible level.

The simulation results also provided diagrams of potential distribution during a short circuit for each substation. Figure 5 shows an example of a diagram for the 110 kV switchgear of the 500 kV TsGPP substation.

By modeling short circuits for each piece of equipment in the design model of the grounding of substations, control cables were identified in which current densities exceeding the permissible value of the thermal resistance of the control cable flowed through the screens, causing thermal damage (see Figure 6a–f).

4. Discussion

The results obtained from the field simulation tests and subsequent modeling confirm that, under real operating conditions at 220–500 kV substations, the distribution of potentials across the grounding network during single-phase short circuits is significantly uneven. The non-equipotentiality of the grounding determines the magnitude of the equalizing currents flowing through the screens of the control cables and, as a result, their thermal damage.

A comparison of the results obtained with the calculated permissible density showed that the limits were exceeded on a number of routes (see Table 2). Our study identified specific locations at the 220 kV and 110 kV switchgears where thermal stability was not ensured.

In our measurements, peak shield current densities reached up to ~1180 A/mm² on certain routes, far exceeding the ~470 A/mm² thermal limit. Such exceedances confirm that, under present grounding configurations, the thermal stability of some control cable shields is not ensured. The most vulnerable circuits were those with the shortest runs between yard equipment and the control room and those connected to grounding points with a large potential rise (equipment with few connections to the main ground grid). These configurations create low-impedance return paths between regions of unequal potential, allowing large fault currents to be diverted through the cable braid. This finding is consistent with observations in medium-voltage networks: for example, ref. [21] reported that a continuous metallic screen path can attract most of the fault current back to the source via the cable shields instead of the earth electrodes. In our case, a similar effect was evident—the cable shields effectively formed unintended return conductors for fault current.

A cross-site comparison clarifies the relative roles of soil resistivity and fault magnitude in setting screen currents. With similar SC magnitudes and with the same cable routing paths, the “TsGPP” substation (OS-220 kV) exhibited lower shield currents (up to ~500 A/mm²) than “Shymkent” (OS-220 kV) because its lower earth resistivity provided a more conductive ground return, reducing the end-to-end driving voltage along cable routes and shifting current away from the shields. In contrast, at Substation “No. 7”, the higher SC magnitude led to higher shield currents (~1180 A/mm²) than at Shymkent, even though their earth resistivities were similar, indicating that source magnitude can override soil effects when the network remains non-equipotential. These observations are consistent with our model’s point: shield current scales with both the fault-time potential difference between shield bonds and the absolute fault current, moderated by the available return impedances.

Within the same site, the “TsGPP” substation, a controlled model comparison with identical SC magnitude and cable routing, showed higher shield currents at OS-110 than at OS-220 due to OS-220’s denser grounding topology, which lowered the fault-time potential difference along the cable route.

To ensure the thermal stability of control cables, it is necessary to equalize the potential (see Figure 5) in the grounding of substations during a short circuit. To do this, it is necessary to strengthen the local grounding at the grounding points of the control cable shields in the switchgear (see Figure 7a,b).

Experimental studies presented in [22,23] show that improving the configuration of the grounding significantly reduces the level of thermal heating of control cable screens. The authors believe that the necessary thermal stability of the screens can be achieved by improving the geometry of the grounding network, in particular by locally compacting its elements and adding horizontal jumpers connecting the cable screen grounding points to a common control point.

The reconfiguration of the grounding models (see Figure 7b), taking into account the proposed measures, demonstrated the expected effect: a decrease in the potential gradient and, as a result, a reduction in current density to an acceptable level.

The proposed methodology allows various conditions to be considered using modeling (densifying of the grounding network, changes in the cable routing, etc.), while the model is linked to the actual grounding network at the substation via field experiments.

This has immediate practical importance, as shown by an actual incident in 2023 (Astana), where underestimating cable shield thermal loads contributed to a cascading outage. Applying the proposed method enables us to find and fix grounding weak points before a fault, when upgrades are relatively inexpensive. By contrast, post-fault costs include unserved energy (lost revenue) and the repair of damaged equipment and cables, which are far higher than proactive grounding improvements.

The results of the field experiments and numerical modeling confirmed that current design practices for grounding devices practically ignore the influence of currents and overvoltage that occur on the screens of control cables during short circuits. At the same time, regulatory documents on the organization of grounding devices [24,25] also do not regulate issues related to the assessment and assurance of the thermal stability of shielded control cables.

The limitations of this study are that it did not consider the high-frequency component of the SC current; this aspect requires separate research. High-frequency content in the fault current can alter the return path. As frequency rises, the inductive reactance of conductors and frequency-dependent resistance grow, so do current flows to lower-inductance routes over higher-impedance alternatives.

In addition, field tests were conducted with low SC currents; with high currents, additional nonlinear effects must be taken into account.

Our field experiments injected a sinusoidal current at power frequency to simulate the SC condition, which neglects the DC-offset and high-frequency transient components of real fault currents. Actual short-circuit waveforms in power systems are asymmetrical, especially in the first few cycles, and contain decaying DC, as well as higher-frequency components. These fast transients could, in principle, induce brief spikes of shield current or resonant responses in loops, potentially aggravating interference. However, the worst-case heating is often governed by the overall fault energy, which is dominated by the power frequency component for fault durations above a few cycles. A more detailed dynamic thermal modeling of the cable system (e.g., time-domain finite-element simulation of transient heating) could refine the predictions of insulation temperature.

Because soil resistivity, ρ, varies with moisture and temperature, the fault-current partition between earth and metallic paths is environment-dependent. A higher ρ increases ground potential rise, tending to increase shield current; a lower ρ shifts current into the earth grid and reduces shield current. In our study, the measurements were taken under summer soil conditions, and the reported current partitions correspond to this measured ρ. When ρ is varied within plausible seasonal bounds, the absolute shield currents shift accordingly, but the trend remains: equipotential grounding (denser, low-impedance bonds) consistently reduces end-to-end shield voltage and screen current, independent of season. For sites with strong seasonal swings (dry/frozen periods), we recommend periodic re-measurement of ρ and quick model re-calibration. In addition, corrosion increases electrode and bond resistances over time, raising ground potential.

Nevertheless, the data obtained allow us to conclude that the proposed method is currently the most flexible and technologically advanced tool for engineering assessments of thermally permissible SC currents in control cable shields at high-voltage substations. Its implementation in the regulations for the design and reconstruction of switchgear will reduce the risk of thermal damage to secondary circuits, increase the reliability of relay protection and automation, and prevent cascading accidents similar to the 2023 incident in Astana. This work has already been applied by Kazakhstan’s national transmission system operator; integrating this approach into future IEC/IEEE standards would close the current gap between substation grounding design and control cable protection.

5. Conclusions

This work presented a combined computational and experimental approach to evaluate and ensure the thermal safety of control cable shields in high-voltage substations during SC faults. By injecting a controlled test current and calibrating a detailed simulation model, we were able to accurately determine the distribution of potential in the substation grounding system, including the often-overlooked paths through cable braids. The major findings and implications are as follows:

• Control cable shields grounded at both ends can carry significant fault currents due to ground potential differences in a fault scenario. This confirms that shield loops in secondary circuits may pose a thermal weak point in system protection. It is not sufficient to assume the ground grid alone will carry all fault return current; designers must account for alternate return paths like cable shields.
• The experimental simulation method proved to be a practical and safe way to gather data on the grounding system. This approach can be applied in other substations with more strict regulations as part of periodic testing to identify latent issues in grounding or bonding.
• Analysis of the research results showed that the current densities arising in the screens of control cables during short circuits in the 220 kV and 110 kV switchgears at substations may exceed the permissible regulatory limits (up to ~1180 A/mm²). The main reason for this is the uneven distribution of potentials across the grounding network.
• Analysis and cross-comparisons identify three primary factors of shield current: grounding topology, soil resistivity, and fault magnitude. For similar fault levels, “TsGPP” substation showed lower shield currents than “Shymkent” substation because its lower soil resistivity favored earth-return paths; by contrast, Substation “No. 7” exhibited higher shield currents than “Shymkent” substation due to a larger fault current despite comparable resistivity. “TsGPP” substation shows that shield currents are lower where the grounding grid is denser and better bonded (more equipotential).
• As practical recommendations, the authors suggest local compaction of grounding devices and an increase in the number of horizontal connections between the grounding points of the cable screens and the SCB (CR). This will reduce the potential level at the grounding device and, as a result, reduce the amount of current flowing through the control cable screens (see Figure 7a,b).

Thus, the proposed methodology is a combination of full-scale simulation experiments and modeling. The obtained results for current densities in the screens confirm the methodology’s suitability for the rapid and reliable determination of the thermal stability of control cables during the design and reconstruction of grounding devices. Inclusion of this procedure in engineering inspection regulations will allow for the design of an economically viable configuration of the grounding circuit, thereby ensuring the required reliability margin of secondary circuits in high-voltage substations.