Experimental study at reduced-scale of fire spread between electrical cabinets located opposite each other

doi:10.1016/j.firesaf.2021.103319

Fire Safety Journal

Volume 122, June 2021, 103319

https://doi.org/10.1016/j.firesaf.2021.103319 Get rights and content

Abstract

The ability of an electrical cabinet fire to spread to neighbouring cabinets is a major concern for fire safety in nuclear power plants. Twelve intermediate-scale fire tests were performed to determine the fire spread conditions (FSCs) from a burning enclosure (BE) to an opposite enclosure (OE) equipped with either a non-combustible (glazed or metallic) or a combustible (poly(methyl methacrylate), PMMA) front panel. The effects on the FSCs of the separation distance (SD) between the enclosures, the target type (electrical component) contained in the OE and overhead electric cable trays were also investigated. Fire spread to the OE equipped with a glazed panel when the total transmitted heat flux led to the spontaneous ignition of the target. This outcome was not obtained for the metallic panel. For the PMMA panel, fire spread when flames coming from the BE reached the flammable gas mixture produced by the panel. For both glazed and PMMA panels, the FSCs were obtained for higher SDs when overhead electric cable trays were used. The tests also revealed that the faster and more powerful PMMA panel fire was not obtained for the minimum SD studied. Finally, total transmittances of the glazed and PMMA panels were assessed.

Introduction

About 500 fire events in Nuclear Power Plants (NPPs) were recorded in the recent version of the OECD FIRE Database [1] for the period from the 1980's to the end of 2017. More than 10% of these events involved electrical cabinets. These last ones are indeed a potential source of fire since they contain both combustible materials and live electrical circuits [2]. Nowadays, the ability of a cabinet fire to spread to neighbouring cabinets is still a major concern for fire safety in NPPs. The electrical cabinets in NPPs can commonly be laid out in multiple parallel rows, as found for example in control or switchgear rooms [3]. Several studies [[4], [5], [6], [7], [8], [9]] investigated the fire spread between adjacent electrical cabinets. In contrast, little research has addressed a scenario in which electrical cabinets are located opposite each other. One purpose of Mangs et al. [6] was to study the impact of a burning cabinet on a closed-door cabinet located opposite it 1 m away and which represented a row of neighbouring cabinets. The latter cabinet was equipped with a metal door and contained poly(vinyl chloride) (PVC) cable samples, which were attached to the door. In all the tests, the cable samples did not burn. However, even though most electrical cabinets in NPPs have metal doors, some low voltage electrical cabinets are equipped with glazed doors or doors including windows made of poly(methyl methacrylate) (PMMA) [2], e.g., in control rooms.

This work therefore aims at investigating the ability of fire to spread from a burning cabinet to an opposite cabinet equipped with either a non-combustible (glazed or metal) or a combustible (PMMA) door. To that end, this study focuses on determining the fire spread conditions (FSCs) according to the door type. More precisely, in order to study the capability of fire to spread through a glazed or a metallic door, electric components are placed as targets inside the cabinet located opposite the burning cabinet. Thus, the determination of the ignition conditions for the studied targets will provide the FSCs according to the non-combustible door type. Furthermore, to ascertain the FSCs to the opposite cabinet equipped with a combustible (PMMA) door, the ignition conditions of the door are specifically investigated.

The FSCs are studied for a separation distance (SD) between the electrical cabinets facing each other that varies in the 0.4–1 m range, as will be explained in section 2.1. Moreover, the electrical cabinets in NPPs can be connected to each other by overhead electric cable trays, as reported in Refs. [2,3]. However, the study of their impact on fire spread between cabinets located opposite each other, as recommended by Chavez [4,5], had not yet been addressed. This study therefore also proposes to examine this issue.

The first and main goal of this study is therefore to determine the FSCs from a burning enclosure to an opposite enclosure equipped with either a non-combustible (glazed or metallic) or a combustible (PMMA) front panel. The effects on the FSCs of the SD between the enclosures in the 0.4–1 m range, the target type (electrical component) contained in the opposite enclosure and overhead electric cable trays are also investigated.

This study also has two additional objectives. The first one is to characterize the PMMA front panel fire of the opposite enclosure and, in particular, to highlight the impact of the separation distance (SD) on the heat release rate (HRR). The second additional purpose is to assess the total transmittances of the glazed and PMMA front panels.

To this end, a test device comprising primarily of a burning enclosure and a closed enclosure located opposite was used to represent at a reduced-scale two electrical cabinets facing each other. In addition, measurements of the incident and total transmitted heat fluxes for the three front panel types of the opposite enclosure as well as videos were notably made.

The results are intended to provide a better understanding of the FSCs between electrical cabinets facing each other in nuclear installations, as well as in other industrial sectors. The outcomes of this work also target to support the analysis of large-scale electrical cabinet fire tests, which were carried out as part of the OECD PRISME-3 project [10].

The experimental setup, the corresponding instrumentation, the fire test matrix and the test protocol are first presented. The results and discussion are then set out in the main part of this work. This part first focuses on the tests that used the enclosure located opposite the burning enclosure without a front panel in order to allow for the measurements of the incident heat flux. The second and third sections of the major part of this work deal with the tests that involved an opposite enclosure equipped with a non-combustible (glazed or metal) and a combustible (PMMA) front panel respectively. Finally, this part ends with the assessment of the total transmittances of the glazed and PMMA front panels. A last portion of this work discusses the representativeness of the performed reduced-scale fire tests.

Section snippets

Experimental setup

The test device (Fig. 1) was composed of two steel enclosures 0.6 m wide, 0.6 m deep and 1 m high, two overhead ladder-type trays and a raised floor. The left-hand enclosure corresponded to a one-quarter scale model of the steel enclosure (1.2 m wide, 0.6 m deep and 2 m high) of the open-door electrical cabinet studied in a previous work [8]. This reduced-scaled enclosure, which is called the burning enclosure (BE), contained a gravel-packed 0.5 m × 0.5 m gas burner located at its bottom (Fig. 2

Tests without a front panel

The first three tests T1-T3 (Table 3) were conducted with an open opposite enclosure (OE), as illustrated in Fig. 7 for T3. The heat fluxes measured by HFS1 and HFS2 for these tests, ${\dot{q}}_{o, 1}^{"}$ and ${\dot{q}}_{o, 2}^{"}$ respectively, are assumed to be nearly identical to those incident to the front panels of the OE during tests T4-T12. This assumption is justified by the small thickness of the front panels studied (4 mm for the thickest one, Table 4) and the closeness of their back surface to HFS1 and HFS2 (gap

Representativeness of the reduced-scale fire tests

Table 5 gives the incident heat fluxes obtained at reduced-scale (i.e., ${\dot{q}}_{o}^{"}$ ) at 300 and 1000 s as well as the maximum ones estimated for real-scale electrical cabinet fires [13], for three distances from the cabinets of 0.4, 0.6 and 0.8 m. The real-scale electrical cabinet was used to design the reduced-scale enclosures (section 2.1). Table 5 first shows that for each of the three above distances the reduced-scale ${\dot{q}}_{o}^{"}$ at 300 s is lower than the real-scale incident heat flux. Indeed, while

Conclusion

The first and main goal of this study was to determine the fire spread conditions (FSCs) from a burning enclosure (BE) to an opposite enclosure (OE) equipped with either a non-combustible (glazed or metallic) or a combustible (PMMA) front panel. The effects on the FSCs of the separation distance (SD) between the opposite enclosures in the 0.4–1 m range, the target type (electrical component) contained in the OE and overhead electric cable trays were also investigated. These tests were also

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

The author thanks Olivier Bouygues for his assistance in carrying out all the experiments.

References (25)

Y. Wang et al.
Investigation of thermal breakage and heat transfer in single, insulated and laminated glazing under fire conditions
Appl. Therm. Eng.
(2017)
I.S. Wichman
A model describing the steady‐state gasification of bubble‐forming thermoplastics in response to an incident heat flux
Combust. Flame
(1986)
Y. Pizzo et al.
Experimental observations on the steady-state burning rate of a vertically oriented PMMA slab
Combust. Flame
(2008)
N. Bal et al.
Experimental study of radiative heat transfer in a translucent fuel sample exposed to different. spectral sources
Int. J. Heat Mass Tran.
(2013)
C.K. Hsieh et al.
Thermal radiative properties of glass from 0.32 to 206 μm
Sol. Energy
(1979)
M.F. Modest
Radiative properties of particulate media
Organisation for Economic Co-operation and Development (OECD) Nuclear Energy Agency (NEA), Committee on the Safety of...
K. McGrattan et al.
Heat Release Rates of Electrical Enclosure Fires (HELEN-FIRE), NUREG/CR-7197, USNRC, Avril
(2016)
Verification and Validation of Selected Fire Models for Nuclear Power Plant Applications, NUREG-1824
(2007)
J.M. Chavez
(1987)

J.M. Chavez

(1988)

J. Mangs et al.

Calorimetric fire experiments on electronic cabinets

Fire Saf. J.

(2003)

Cited by (5)

Enhancing electrical panel anomaly detection for predictive maintenance with machine learning and IoT
2024, Alexandria Engineering Journal
This study aims to detect electrical panel fires using the Internet of Things (IoT) framework and Machine Learning (ML) algorithms. Within the scope of the study, an experimental process was carried out using Arduino and Raspberry Pi platforms to collect essential data such as gas, temperature, and humidity. In this experimental work conducted on 3478 data points to detect electrical panel fires, Decision Tree (DT), Gaussian Naive Bayes (GNB), Gaussian Process Classifier (GPC), and Support Vector Machine (SVM) algorithms were used, and their performances were evaluated. For each algorithm, the best hyperparameters were selected using the 5 k-fold cross-validation method, and different models trained with these hyperparameters were examined. The findings indicate that the GPC algorithm has higher accuracy values than others, and its performance is consistent with a high potential for generalization. The GPC algorithm has stood out from the others with its high-performance values. For the GPC algorithm, the accuracy, precision, recall, F1 score, and Area Under the Curve (AUC) metrics values were 99.56%, 0.978, 0.989, 0.983, and 0.99, respectively. Additionally, Receiver Operating Characteristic (ROC) analysis has shown that GPC, DT, and SVM algorithms effectively distinguish positive and negative classes.
Scale modeling study of airflow distribution uniformity in large spaces with high heat flux
2022, Applied Thermal Engineering
Citation Excerpt :
In this case, the orifice plate air supply could be considered as an alternative method to maintain the temperature variation within ± 1 °C in the upper truss area of JUNO. As a primary method for verifying the rationality and reliability of heating, ventilation, and air conditioning design schemes, the model testing method has been commonly used in spacious architectures such as utility tunnels [31–33], high-rise buildings [34–36], hydropower stations [37–39] and other complicated spaces with multiple zones [40–42]. Nonetheless, there are frequent scaling problems during attempts to achieve flow similarity between the scale model and the real building.
With regard to the arrangement of air conditioning and ventilation in a building with large spaces, the uniformity of airflow distribution is often related to the air supply mode, heat flux, and location of the heat source. In this study, a 1:38 small-scale model of a large-scale sphere-shaped facility with high heat flux dissipation was constructed for investigating and analyzing the airflow distribution in the upper occupied zone. Additionally, a series of environmental parameters was experimentally obtained to characterize the influence of the location and release rate of the heat source. A dimensionless method was used to process the acquired data and explore the airflow characteristics. The experimental results revealed the evident influence of the release rate and heat source positions on the temperature distribution in the upper occupied zone. Furthermore, a reference temperature non-uniformity coefficient was proposed to assess the temperature uniformity in the occupied zone. Ultimately, the coefficients of temperature non-uniformity, reference temperature non-uniformity, and energy efficiency were reliably predicted. Consequently, a supply air temperature of 18 °C was selected as the optimal design parameter after conducting experiments to maintain the upper occupied zone temperature at 21 ± 1 °C and the corresponding minimum airflow rate could significantly reduce the energy consumption of air processing.
A Study on the Evolution of Flame Height and Air Entrainment Volume Rate for Typical Electrical Cabinet Fires in Nuclear Power Plants
2024, SSRN
Experimental study on combustion characteristics of fire in data room
2022, Journal of Fire Sciences
A consistent treatment of moving boundaries with thermal convection for lattice Boltzmann method
2021, Physics of Fluids

View full text

Experimental study at reduced-scale of fire spread between electrical cabinets located opposite each other

Abstract

Introduction

Section snippets

Experimental setup

Tests without a front panel

Representativeness of the reduced-scale fire tests

Conclusion

Declaration of competing interest

Acknowledgement

Appl. Therm. Eng.

Combust. Flame

Combust. Flame

Int. J. Heat Mass Tran.

Sol. Energy

Heat Release Rates of Electrical Enclosure Fires (HELEN-FIRE), NUREG/CR-7197, USNRC, Avril

Verification and Validation of Selected Fire Models for Nuclear Power Plant Applications, NUREG-1824

Calorimetric fire experiments on electronic cabinets

Fire Saf. J.