Introduction. Experimental identification of the properties of smoke generated during the thermal decomposition of insulation from non-flame-propagating (NG) cables is an important research problem.

Goals and objectives. This study experimentally evaluates specific smoke-generation coefficient of NG cables and sizes of solid smoke particles formed during their thermal decomposition.

Materials and methods. Experimental methods were used to measure the number, size, and mass of suspended solid smoke particles and the specific smoke-generation coefficient of substances and materials in a small-scale experimental setup. The study was conducted using samples of a control NG cable, KVVGng(A) 4 × 1.5, designed for fixed connections to electrical devices rated at up to 660 V AC and frequencies of up to 100 Hz.

Results and discussion. Fine solid smoke particles smaller than 2.5 μm, which pose the greatest hazard to evacuees during a fire, were found to form throughout the entire period of thermal decomposition. 
At the initial stage of thermal decomposition, the specific smoke-generation coefficient reached maximum values of 772–982 Np·m2/kg. After reaching its maximum, this coefficient decreased sharply and then reached a quasi-­steady level of 314–471 Np·m2/kg. 
Thus, thermal decomposition of NG cables produces an extremely hazardous gaseous environment containing fine smoke particles smaller than 2.5 μm, which can adsorb highly toxic gases generated during the process (carbon monoxide, hydrogen cyanide, hydrogen chloride, phosgene, and acrolein). The results can be used to determine the time until escape routes become untenable due to visibility loss and to assess the effects of exposure to smoke particles on the human body.

Conclusions. When the time until escape routes are blocked by hazardous fire factors is calculated in accordance with fire-safety regulations, failure to account for fine smoke particles may lead to underestimation of toxic exposure among evacuees.

Introduction. The combustion characteristics of gaseous mixtures in water sprays are analyzed. The relevance of this study is attributable to the wider use of dispersed water to reduce the explosion hazard associated with accidental releases of flammable gases. The aim of this work is to provide an analytical review of studies on the influence of dispersed water on the explosion characteristics of flammable gaseous mixtures.

Analysis of studies addressing the phlegmatizing effect of dispersed water. Droplet size is shown to be the key parameter determining the phlegmatizing effect of dispersed water. Phlegmatization is observed at droplet sizes below 20–30 µm. Phlegmatization reduces the maximum explosion pressure, the explosion pressure rise rate, and the normal burning velocity. However, complete phlegmatization is generally attainable only for near-limit mixtures. Water droplets with diameters up to 1,000 µm can be used to suppress gaseous detonation due to the fragmentation of large droplets in a shock wave.

Analysis of studies reporting the combustion-promoting effect of dispersed water. Combustion intensification was observed under the action of water droplets larger than 50 µm; this effect was accompanied by flame acceleration and more complete combustion. In some cases, combustion-to-detonation transition can occur.

Conclusions. Based on the analysis performed, conditions are identified under which dispersed water can intensify or suppress the combustion of gaseous mixtures.

Introduction. Ensuring fire safety of marine vessels is complicated by diverse combustible loads — from flammable liquids to smoldering cables and insulation. Heat detectors (HD) are often selected formally, based on a certificate, without considering actual fire scenarios, which leads to their ineffective installation in areas where smoldering dominates.

Goals and Objectives. To determine the effective application limits of heat detectors on marine facilities and to justify that the multi-criteria approach provides reliable early fire detection under any fire development scenario.

Methods. The regulatory framework (GOST R 53325–2012, GOST 34698–2020, Russian Maritime Register of Shipping Rules, SP 484.1311500.2020) was analyzed. Experimental verification was performed using data from the R&D project “Triumph-25” at the “Hephaestus” training facility. The response times of the thermal, smoke and CO channels of the multi-criteria detector “Bark M” were recorded for various test fires (the paper presents data for seven representative fire types, including smoldering, liquid fires and cable pyrolysis).

Results. Certification tests of HD according to GOST are conducted only on test fire TF-6 (liquid fuel fire), which is correct for fast-developing fires but does not guarantee performance under smoldering. Experiments confirmed that under smoldering, the thermal channel either significantly lags or does not activate at all. The Russian Maritime Register of Shipping Rules directly prohibit systems using only heat detectors and introduce delta-factors (temperature — 5 °C, smoke — 0.5 %/m, CO — 10 ppm over 15 min), which are unattainable for conventional HD but feasible for multi-criteria detectors.

Conclusions. The multi-criteria approach, using combined information from thermal, smoke and gas channels, provides reliable early fire detection based on the analysis of several heterogeneous factors (typically at least two), which eliminates false alarms and guarantees reliable fire identification under any scenario. Recommendations: for mixed combustible loads, use multi-criteria detectors tested on all standard test fires according to GOST R 57552–2017; in sector-specific regulations, clarify the criteria for “special justification” of using only heat detectors.

Introduction. Accidental explosions of gas-air mixtures in confined spaces pose a serious hazard to personnel, building structures, and production facilities. A key factor determining the nature of damage is not only the peak overpressure, but also the time history of its development governed by gas-dynamic and deformation processes within the enclosed space.

Goals and objectives. The aim of this work is to analyze the relationship between the stages of deflagration explosion development identified using the video recording data and changes in overpressure inside the experimental cleanroom.

Materials and methods. The experiment was carried out using a laboratory unit (2,230 × 2,230 × 2,360 mm, effective volume 11.74 m3) with enclosing structures made of AMS-MZMO sandwich panels. A doorway covered with polyethylene film served as the relief structure. Propane-butane (525 l, stoichiometric mixture, 4.5 %) was supplied into the experimental cleanroom; overpressure was recorded by two APZ 3420 sensors; the process was filmed by a high-speed camera at 239 fps.

Results and their discussion. The main process stages were identified using the video recording. They included tightness loss (0.735 s), first structural deflection (0.747 s), maximum deflection (0.865 s), and opening formation (0.885 s). Maximum overpressure values were 654 Pa (sensor  1) and 520 Pa (sensor  2). Comparison between the video data and the pressure oscillogram was employed to identify a staged pattern of overpressure development associated with sequential gas-dynamic and deformation development changes in the enclosing structures.

Conclusions. A combined analysis of the video recording and the pressure oscillogram data enables a more exhaustive description of explosion loading patterns for an indoor deflagration explosion in a cleanroom; this methodology can be applied to analyze other enclosing structures.

Introduction. Mezzanine structures are subject to a regulatory requirement of 15 min for loss of load-bearing capacity under the standard temperature-time regime. At present, no approved calculation method is available for determining the actual fire-resistance limits of steel mezzanine rack structures or the limiting internal forces in load-bearing members needed to achieve the R15 rating under the applied loads, including loads from subsequent tiers.

Aim and Objectives. The aim of this study is to provide computational and experimental substantiation of the fire resistance of warehouse mezzanine racks made of thin-walled cold-formed steel profiles manufactured by PAO Severstal. To achieve this aim, the following objectives were addressed: calculation schemes and load cases were determined for two representative fragments of a mezzanine rack; the fire-resistance limits of the structural elements were calculated; fire tests were conducted to determine fire resistance according to the R criterion and to validate the calculations; and applicability limits for the obtained results were established.

Methods. Calculation schemes and load cases for two representative mezzanine rack fragments were determined, and the fire-resistance limits of structural elements were calculated using two methodological approaches: direct reduction of mechanical properties followed by recalculation of the load-bearing capacity in accordance with GOST R 56567–2015, and a method based on the provisions of TKP EN 1993-1-2–2009 for Class 4 cross-sections. Fire tests of two mezzanine-fragment specimens were conducted in accordance with GOST 30247.0 and GOST 30247.1 to determine fire resistance according to the R criterion and to validate the calculations.

Results and Discussion. It was established that unprotected thin-walled rack structures, when their cross-sections and section factors are selected correctly and effective section properties are accounted for with experimental validation, can provide an inherent fire-resistance limit of at least R8. This provides a basis for developing a design methodology for rack mezzanines with regulatory justification for use in facilities requiring a fire-resistance limit of R15.

Conclusions. The refined calculation dependencies and coefficients are not universal for all racking systems or all LSTC profiles. Their application is permissible only for structures similar to the tested specimens in terms of material, profile shape and thickness, perforation pattern, calculation scheme, joints, stress level, section factor, and fire exposure conditions.

Introduction. Current dispatch schedules for fire and rescue units are drawn up on the basis of a fixed order of priority for deploying units according to fire severity levels and, as a rule, do not take into account the location of the incident, the configuration of the road network or the current operational situation. This leads to an increase in the time taken to mobilize personnel and equipment when responding to higher-priority fires.

Aims and Purposes. The aim of the study is to develop a method, as well as models and algorithms, for forming a rational sequence of dispatching fire units. The study addresses the tasks of model development and the design of decision-support algorithms.

Methods. The research is based on systems analysis, mathematical modelling and geoinformation analysis. A method of territorial decomposition of service areas (subareas) is proposed, based on modelling the travel time of operational crews to different locations within the garrison. For real-time decision-making, a lexicographic decision-support model is developed, taking into account travel time, the number of personnel involved in breathing apparatus operations, water tank capacity, fuel consumption, and the risk of adverse events within the service area.

Results and discussion. A territorial decomposition method, an algorithm for constructing a relational data model of dispatch sequences, and a software module for its implementation have been developed. It is shown that the proposed approach reduces the estimated time of assembling units by 30–70 % compared with existing turnout schedules. The lexicographic model provides a justified selection of alternative sets of fire units under current operational conditions and improves operational and tactical response capabilities.

Conclusions. The proposed method, models and algorithms formalize decision-making in fire unit dispatch, ensure a more justified selection of operational crews, and reduce the influence of the human factor. Their application is advisable both in the development of turnout schedules and in dispatcher decision support during real-time response. Further research may focus on expanding the system of dispatch criteria and incorporating artificial intelligence methods.