Experimental Research and Numerical Analysis of Marine Oil Leakage and Accidental Ignition in Fishing Vessels

Featured Application

This work expands the existing limitations of ship engine room fire assessment, providing more support for ignition source control in enhance fishing vessel safety.

Abstract

The hazard of highly combustible marine oil leakage greatly increases fishing vessel operation risks. This research integrates an experiment to explore the coupling mechanism of a typical heated surface of an engine room as a source to ignite marine oil. A numerical model is established that depicts the dynamic process of and variations in the combined effects regarding multiple factors of oil ignition under actual experiment. The leaked marine oil is ignited with a heated surface, relevant models are applied to reproduce the results, and the influences of specific parameters of a fishing vessel’s engine room are analyzed. The results indicate that the leaked oil boils violently on the heated surface, and a vapor film forms on the oil surface. Increased heated-surface temperatures lead to a significant difference in the initial ignition occurrences of marine oil, and the distance between the ignition height and oil is closely related to the engine room environment. The ignition probability of marine oil shows a gradually increasing trend with elevated heated-surface temperatures. The ignition height presents a downward trend with the increase in the heated-surface temperature, while the engine room’s humidity in air inhibits the upward transfer of heat; however, the degree of inhibition is limited accordingly. The results evidence that this comparative work can be an effective approach to reveal the impacts of marine oil, heat source, ventilation velocity, and humidity on initial ignition characteristics. Additionally, this work provides a basis for setting up emergency planning with appropriate monitoring equipment and further preventing vessel fires due to oil–thermal ignition.

1. Introduction

Fishing vessels are affected by the wind, waves, ice, and other factors during navigation, which can cause maritime accidents. Statistically speaking, fishing vessel fires are the greatest contributor to accidents, with enormous losses. On 10 December 2020, a fishing vessel named Lucky Angel was trawling for shrimp in the Gulf of Mexico when a fire broke out in its engine room. The fishing vessel suffered a total constructive loss of nearly USD 120,000 [1]. According to statistical data from 2017 to 2021, about 63% of all accidental fires originated in the engine room, and 56% of such fires occurred due to the mutual contact of leaked oil and a heated surface [2,3]. Further data related to fishing vessel fires indicate that nearly two-thirds of these fishing vessel engine room fires occur on the main and auxiliary engines or associated components such as turbochargers. When a fishing vessel is sailing or operating near a port, the oil container or piping system in the engine room breaks and leads to oil leakage. Once the marine oil of the engine room flows near the high-temperature surface of mechanical equipment under high-load operation, it is easy for a flowing oil fire to be triggered due to contact with the heated surface. Compared with ordinary cabins, a fishing vessel’s engine room is special in terms of application, layout, and internal environment. Therefore, the damage degree and difficulty of the fire caused by the engine room are far greater than those of other areas, such as the passenger cabin and cargo hold. For the accurate prevention of fire accidents in a fishing vessel’s engine room, on the one hand, relevant fire prevention measures should be taken based on the fire types induced by different physical hazards in the engine room. At present, there are relatively perfect classification standards and coping methods for the risk classification of a fishing vessel engine room. On the other hand, the ignition source functions as the energy source of a dangerous material combustion reaction in the early stage of a fire. Therefore, it is necessary to study the induction of a fire accident in a fishing vessel’s engine room. Based on the knowledge gained, the prevention and elimination of the ignition source is the most critical correction to cut off the fire chain in a fishing vessel’s engine room.

Due to shipping operations, fishing ships greatly rely on marine diesel and bunker oil [4]; however, accidental fires in the engine room caused by fuel leakage constitute a tough problem. Fabiano et al.’s statistical research of ship accidents from 1987 to 1998 [5] revealed that the main causes of engine room failure are engine trouble (30%), collisions (24%), and stranding (12%). McNay et al. [6] indicated that engine room fires are primarily caused by flammable fuel leakage and its contact with unprotected heated surfaces. The work also suggested that more cost-effective safety measures should be developed by addressing systemic causes of ship engine room fires. Hazards connected to the handling of combustible materials in ship engine room areas mainly originate from the complicated nature of activities that occur there: the possibility of fire increases in the event of piping system and equipment failures or external events [7]. Ahn et al. [8] presented a framework for evaluating specific scenarios associated with human errors in ships, and the Cognitive Reliability and Error Analysis Method (CREAM) method was applied in the case of a ship fire. Spyrou et al. [9] proposed a risk model for assessing ship fires at the initial stage of design, and this model indicated that the reliability index of fire safety systems is a core factor affecting the potential for large consequences in a ship fire. Zhang et al. [10] developed a dynamic evolutionary model for quantifying the domino effect of ship fires, which indicated that fuel ignition would cause a whole engine room fire in a short period due to narrow space. In ship fire accidents, the risk of the main engine fuel system is the highest, because the leakage of marine fuel from the piping system or diesel supply tanks is the main failure cause [11]. Bao et al. [12] applied a Bayesian Network (BN) method to analyze the probabilities of ship fires, and this work revealed the symptoms of poorly conditioned vehicles, including aging electrical lines and heavy oil stains in the ship engine room. The leakage of marine fuel, which may be ignited by a heated surface, is the main contributor to ship engine room fires. Antão et al. [13] conducted an assessment of specific risk-influencing factors regarding the occurrence of ship accidents based on a global sample involving 936 events from the period of 2005–2017. Fires, collisions, and groundings represented nearly 50% of the overall number of ship accidents. Some other dynamic risk assessments of different types of ships were carried out, illustrating the common problems of piping failures and potential ignition sources from the machinery system in the engine room [14,15,16,17]. In fact, an accidental release of marine fuel can lead to a stratified vapor cloud in the ship engine room. Two kinds of hazards can take place: one referring to mechanical damage caused by overpressure resulting from an explosion, and the other concerning heat exposure from fire development [18]. Therefore, it is of great value to understand the safety management of small fishing ships [19], including the piping system design and environmental conditions related to engine operations. Many previous numerical studies have been conducted on ship fires. For example, Su et al. [20,21] completed ship engine room fire simulations by using large eddy simulation (LES). The numerical models were performed in a fire dynamics simulator (FDS), and the main engines and auxiliary machines were constructed in an engine room. The simulated results indicated that the temperature distribution in the ship engine room is significantly different, and the negative pressure would have an effect on ship fire development. Kang et al. [22] focused on how to arrange ship fire control with minimal changes to existing design, and their study showed a framework for using a numerical method in the initial stage of ship fire safety. Meanwhile, it is also important to effectively estimate the risk index and the possibility of fires in real time. Park et al. [23] developed a prediction model by using ANASYS CFX and PHAST data [24] that can evaluate the consequences of ship engine room fires at the beginning stage. Lan et al. [25] reconstructed a 3D model of a ship engine room based on the distant-water training ship Yukun in Dalian, China. They found that a flame would appear in a mechanically ventilated engine room. The flame propagation was subject to a comparative effect from the momentum, buoyancy, and pressure of the ship engine room’s airflow. The arrangement of ventilation conditions in the ship engine room has a great effect on flammable material ignition and flame propagation, which affects the combustion characteristics in ship fire accidents [26]. Numerical models for ship fires were built by CFD-based tools based on standards and cases [27,28], which showed better improvements by combating with experimental data [29,30]. In another experiment, it was found that the general behavior of marine oils was steady combustion after ignition in open space, where the low and intermediate fractions began to evaporate, causing a large number of bubbles [31]. Wang et al. [32] carried out a series of tests in a sealed ship engine room, and the experimental results indicated that the evaporated fuel was not involved in chemical reactions during the burning process, resulting in a burning rate of marine fuel lower than the mass loss rate. Liu’s experimental work [33] found a new flame spreading phenomenon of “continuous-flow around-broken-retract” in ship fires. Further, this study proposed a prediction relationship between the flame drag length of a long and narrow rectangular oil pool, which was based on a dimensionless fitting curve. In a real experiment, it was verified that a ship fire would be affected by the airflow of the engine room [34]; in addition, the influences of air supply volume and inlet height on marine fuel ignition and flame characteristics were verified. Obviously, leaking marine oil can be ignited by a heated surface, which is relevant to flammable oil properties, the heated-surface temperature, and environmental parameters [35,36,37,38]. Recently, Lekomtsev et al. [39,40] carried out experimental and numerical works on the complicated phase transition process of crude oil, which was described to confirm the effectiveness of a comparative method used in the design of technical characteristics. However, in-depth works have not yet generated widely common results, and a practical model for marine oil ignition by a heated surface in a ship engine room fire needs to be further explored. With the combination of reasonable ignition and fire scenario modeling, consequence-based experiments, and targeted emergency planning by decision makers [41], accidental risks in the engine room area of fishing vessels can be reduced.

The current research aims to determine the correlation between marine oil, flow state, fishing vessel engine room wind conditions and environmental humidity, and the multiple factors based on heated-surface ignition. With experimental data and a numerical simulation, this research develops an energy migration model of heated surfaces so as to reveal the time-varying evolution mechanism of marine oil ignition under the coupling effect in an engine room. A new relationship between ignition probability and heated-surface temperature is presented based on comparative data, and the dynamic process of evaporation and the phase change of leaking marine oil on a heated surface are analyzed. In addition, this research proposes a prediction model of marine oil ignition characteristics (e.g., ignition occurrence, heated-surface temperature, wind velocity, humidity, etc.) that covers key factors in field-flow ignition. This research expands the ignition source assessment of fishing vessel engine rooms, providing more support for fire prevention in enhancing fishing vessel operation safety.

2. Materials and Methods

2.1. Experimental Set-Up

2.1.1. Experimental Site

The object of this research is a distant-water double-deck trawler, which is mainly used for single bottom trawling in waters of West African countries such as Morocco. The fishing vessel has the following characteristics: made of steel, double bottom (except engine room), double deck, forward steering tower, stern type, aft side hangar, single machine, large-diameter propeller, single rudder, and stern slipway trawler with stern portal and gantry. The stern line of the vessel has a wide, whole-deck surface fitted with a stern slipway, a rounded straight stern sealing plate, and a ball stern in the front of the hub. The bow line of the vessel has a large outward flapping, a forward-leaning stem, and an SV-shaped ball bow. The total length of the fishing vessel is 33.2 m, the design draft is 3.5 m, and the design displacement is 510.81 t. The fishing vessel’s engine room is located on the lower deck, with a length of 10.0 m, a height of 3.5 m, and a total area of about 48.0 m². In the middle of the fishing vessel’s engine room, a gear box, high-elastic coupling, main engine, and hydraulic pump station are arranged successively. The generator sets are arranged on the left and right sides of the engine room. According to the operating environment of the real fishing vessel’s engine room, this experimental work is completed in a full-scale ship engine room laboratory, as shown in Figure 1. The layout of the laboratory is constructed in equal proportion to the fishing vessel’s engine room. The laboratory wall is made of a marine steel plate and the floor is made of a patterned steel plate, which are consistent with the interior of the real fishing vessel’s engine room. The engine room has shutters to exhaust the air. The laboratory is equipped with a top exhaust system and a bottom air supply system, which can be used to simulate the internal airflow organization of the engine room during fishing vessel operation. Temperature and humidity monitors were set up in the laboratory, which were applied to control the environmental parameters in different scenarios. Meanwhile, a humidity regulator was set up, which can be used to simulate the humidity conditions in the engine room under different scenarios. In addition, since the temperature in the engine room is higher than that in a conventional cabin, a temperature adjustment instrument was set in this experiment to adjust the laboratory temperature to be consistent with the real operation condition of the fishing vessel’s engine room.

2.1.2. Experimental Apparatus

In this experiment, a typical heated-surface simulator in the fishing vessel’s engine room was developed. The simulator mimics the high-temperature heated surface formed by typical equipment in the engine room, as shown in Figure 1. This heated-surface simulator has multiple functions, such as fast heating speed and high heating efficiency, and it can control the initial temperature of the heated surface and the real-time change temperature of output of the heated surface. It is mainly divided into two parts. One part is the surface of the heating table made of a titanium alloy material, and the shell is made of cold-rolled steel. The surface of the heating table was treated by electrostatic spraying, and there are heat dissipation holes on the side to ensure safety during the heating process. The other part is a man-made temperature controller. The main function of the temperature controller is to adjust the temperature of the heated surface, and it can be connected with a computer outside the laboratory in order to save the temperature data in real time. The heated-surface simulator in the engine room laboratory has an adjustable temperature range of 0 K to 1100 K with an accuracy of ±1.0 K. The heated surface and temperature controller were connected with a power output line, and the outer layer of the connection line had a protective cover. Except the simulated typical heated-surface platform, no other open flame ignition device was set. A K-type thermocouple was arranged above the heated surface, as shown in Figure 1. In this experiment, the purpose of arranging thermocouples is to collect the temperature change in the vertical space after the leaked marine oil contacts the heated surface. The K-type thermocouple was fixed on a self-machining bracket. A total of 3 columns were arranged, and 10 K-type thermocouples were installed in each column. The K-type thermocouples were close to the liquid surface of the marine oil. When the heated-surface ignition occurs, a flame spreads to the position, and the thermocouple will produce a response of temperature rise. There are two ways to determine that the leaked marine oil is ignited by the heated surface: one is to calculate the temperature rise of the K-type thermocouples as the time scale, and the other is to use images collected by a high-speed camera and the distance scale (using the thermocouples as the distance scale). Two heat flow meters were placed near the heated-surface simulator in the fishing vessel engine room to measure radiative and convective heat flow with a maximum measurement of 100 kW/m². In this experiment, a high-precision electronic balance was used to measure the mass transient change of marine oil on the heated surface. An electronic balance was placed below the heated surface of the engine room to collect data on the mass loss of the leaked marine oil during ignition. The mass loss rate of the leaked marine oil is one of the significant indexes reflecting its combustion characteristics. The range of the electronic balance is from 0 g to 5000 g, and the measurement accuracy is 0.01 g. The electronic balance was preheated before the experiment and calibrated with standard weights. An observation port was opened on the side of the laboratory to capture the whole process of marine oil ignition by the heated surface. Because the heated-surface ignition moment is quite short, the high-speed camera was adopted for capturing up to 1000 fps images. In addition, a TP700 data acquisition instrument was used to communicate with Modbus, with a total of 64 channels. The data collected by the K-type thermocouples, high-speed camera, heat flow meter, and other equipment were transmitted to the data acquisition instrument outside the laboratory through the connection line. The measuring point information collected by the data acquisition instrument was then sent to the computer for recording and storage for analysis.

2.1.3. Marine Oil Properties

According to the specifications of the main diesel engine in the fishing vessel’s engine room, marine diesel was selected as the type of leakage oil in this experiment. The main diesel engine in the fishing vessel’s engine room is an in-line four-stroke main engine, which adopts the form of closed cooling. The main diesel engine has 6 cylinders and uses 0# diesel. The marine oil is collectively referred to as marine diesel oil and heavy fuel oil. For an accidental scenario of main diesel engine failure and oil leakage, light marine diesel oil was used in this experiment, whose chemical determination was tested. The flash point temperature of the marine diesel oil used in the experiment is about 63 °C, which indicates that the flash of disappearance of the liquid surface occurs when the oil contacts with the flash point temperature under normal conditions. The initial boiling temperature is 170 °C, and the vapor pressure at 20 °C is 2.0 hPa. The marine diesel has a low/high flammability of around 1% to 6%, which is also an important flammability indicator. The other chemical properties of the marine diesel used in this experiment are shown in

Table 1. Main properties of leaked marine oil diesel in this experiment.

	Density at 15 °C (kg/m3)	Flash Point (°C)	Kinematic Viscosity at 50 °C (cSt)	Pour Point (°C)	Water Content (v/v)	Cetane Number
Marine diesel oil	991	63	180	24	0.5%	40~60

. In this experiment, the external oil tank was connected with a peristaltic pump, and the amount of leaked marine oil on the heated-surface simulator was controlled by a flow controller. The experimental volume ranged from 10 mL to 20 mL. According to the typical heated-surface temperature in the fishing vessel’s engine room, a total of 30 tests were conducted based on each adjusted heated-surface temperature. Therefore, the number of tests can ensure a comprehensive analysis of marine oil ignition characteristics by a heated surface in a fishing vessel’s engine room.

2.2. Numerical Procedure

2.2.1. Physical Model

The comparative analysis in this study aims to compare the numerical simulations with the experimental results, so as to obtain the law of marine oil leakage and ignition on a typical heated surface in a fishing vessel’s engine room more accurately. The diesel engine in the fishing vessel’s engine room was equipped with a fuel pump, oil pump, fresh water-cooling pump, oil cooler, fresh water cooler, and expansion joint supply. Meanwhile, the main diesel engine was equipped with a remote instrument, and the front end of the main engine was equipped with a hydraulic pump station (speed-increase gearbox) through a short shaft. Before establishing the geometric model, this research investigated the real operating environment of the fishing vessel’s engine room and the equipment with a high-temperature heated surface to obtain accurate geometric parameters. The main engine, generator set, and other auxiliary engines in the engine room were designed according to the continuous power of the diesel engine. The fluid-phase thermodynamic package utilized in this numerical analysis was FLUENT, and an equal-scale scene model of a typical heated-surface accident area of the fishing vessel’s engine room was constructed. As shown in Figure 2, the regional scene model includes equipment such as ship diesel engine, generator set, steam boiler, and related piping system. For the ventilation system in the fishing vessel’s engine room, the lower outlet and the top outlet of the engine room were set in the numerical model. The vent is the top outlet, located at the top of the entire fishing vessel’s engine room, with a radius of 0.4 m. In the numerical model, the air outlet was set at the lower part of zone model, and the diameter was set at 0.3 m. The computational domain size set was 3.0 m (length) × 3.0 m (width) × 3.5 m (height). The computational domain was mainly set for the area where the oil leaks and the heated surface ignites, which is the main diesel engine and oil supply pipeline. By simplifying the size of the computational domain, the mesh size was set more finely. In ANSYS 19.0, the fishing vessel’s engine room model was segmented. Due to the complexity of the interior structure of the engine room, the numerical model was divided into non-structural grids with strong adaptability. In the current model, the minimum grid size was set at 0.05 m, and the total number of grids in the entire computational domain was 252,000.

2.2.2. Governing Equation

In the computational fluid domain of a typical heated surface in the fishing vessel’s engine room, the whole calculation process follows the basic equations of mass conservation, momentum conservation, and energy conservation. According to the law of mass conservation, the mass increment of the fluid in the computational domain should be equal to the difference between the mass of inflow and the mass of outflow. Based on this, the integral form of fluid flow continuity equation is deduced as Equation (1). The first term on the left of Equation (1) represents the increment of mass inside the control body, and the second term represents the net flux from control surface into the control body.

$$\frac{\partial }{\partial t}\underset{\mathrm{v}}{\overset{}{\iiint }}\rho \mathrm{d}x\mathrm{d}y\mathrm{d}z+\underset{\mathrm{A}}{\overset{}{∯}}\rho v·\mathrm{ndA}=0$$

(1)

where V is the control body, m³; A is the control surface, m²; ρ is the density, kg/m³; v is the flow velocity of the leaking marine oil, m/s; t is the time, s.

The motion equation involved in the process of igniting marine oil on the heated surface in the fishing vessel’s engine room is the momentum conservation equation, which represents that the time rate of change of momentum in the system is equal to the sum of the external forces acting on it, as shown in Equation (2).

$$\left\{\begin{array}{c}\rho \frac{\mathrm{d}u}{\mathrm{d}t}=\rho F_{bx}+\frac{\partial p_{xx}}{\partial x}+\frac{\partial p_{\mathrm{yx}}}{\partial y}+\frac{\partial p_{\mathrm{zx}}}{\partial z}\\ \rho \frac{\mathrm{d}v}{\mathrm{d}t}=\rho F_{by}+\frac{\partial p_{xy}}{\partial x}+\frac{\partial p_{yy}}{\partial y}+\frac{\partial p_{\mathrm{zy}}}{\partial z}\\ \rho \frac{\mathrm{d}w}{\mathrm{d}t}=\rho F_{bz}+\frac{\partial p_{xz}}{\partial x}+\frac{\partial p_{yz}}{\partial y}+\frac{\partial p_{zz}}{\partial z}\end{array}\right.$$

(2)

where F is the component of mass force per unit mass on fluid in three directions, N/kg; p is the component of the internal stress tensor of fluid.

The fluid motion of leaking marine oil on a heated surface follows the first law of thermodynamics, so its energy equation can be shown in Equation (3). As defined by the turbulence model used in the numerical calculation, J_j′ is the diffusion flow of component j. S_h covers the chemical reaction heat and other customized volumetric heat source terms. The first three terms on the right of Equation (3) describe energy transport by heat conduction, component diffusion, and viscous dissipation, respectively.

$$\frac{\partial }{\partial t}\left(\rho E\right)+\frac{\partial }{\partial x_i}\left(u_i\left(\rho E +p\right)\right)=\frac{\partial }{\partial x_i}\left(k_{\mathrm{eff}}\frac{\partial T}{\partial x_i}-\sum _{{\mathrm{j}}^{\prime }}{h}_{{\mathrm{j}}^{\prime }}{J}_{{\mathrm{j}}^{\prime }}+{ u}_j{\left({\tau }_{ij}\right)}_{\mathrm{eff}}\right)+{ S}_{\mathrm{h}}$$

(3)

$$E=h-\frac{p}{\rho }+\frac{u_i^2}{2}$$

(4)

where k_eff is the effective heat transfer coefficient; k_t is the turbulent heat transfer coefficient, W/m²⋅K.

The leaked marine oil in the fishing vessel’s engine room absorbs the heat transferred by the high-temperature heated surface after being converted from liquid to vapor, during which a phase transition occurs. The model computed the numerical simulation of the phase transition processes, following the principles below.

• Regarding the evaporation process of leaked marine oil (T_l ≥ T_sat), the mass transfer rate of phase transition is shown in Equation (5).

  $${\dot{m}}_{\mathrm{lv}}=coeff·{\alpha }_{\mathrm{l}}{\rho }_{\mathrm{l}}\frac{\left(T_{\mathrm{l}}-T_{\mathrm{sat}}\right)}{T_{\mathrm{sat}}}$$

  (5)
• Regarding the condensation process of leaked oil (T_v < T_sat), the mass transfer rate of phase transition is shown in Equation (6).

$$\dot{m_{\mathrm{vl}}}=coeff·{\alpha }_{\mathrm{v}}{\rho }_{\mathrm{v}}\frac{\left(T_{\mathrm{sat}}-T_{\mathrm{v}}\right)}{T_{\mathrm{sat}}}$$

(6)

where α_v is the volume fraction of the phase; ρ is the density of the phase, kg/m³; T_l is the temperature of the liquid phase, K; T_v is the temperature of the gas phase, K; T_sat is the saturation temperature, K; and coeff represents a coefficient, which equals to reciprocal relaxation time.

The numerical process of the energy source term S_h is shown in Equation (7).

$$S_{\mathrm{h}}=-\dot{m}·L_{\mathrm{v}}$$

(7)

where m is the quality, kg; L_v is the latent heat of evaporation, kJ/g.

Equations (8)–(10) are used to calculate the evaporative latent heat of leaking marine oil in the fishing vessel’s engine room.

$$L_{\mathrm{v}}=h_{\mathrm{q}}^{\mathrm{s}}-{ h}_{\mathrm{p}}^{\mathrm{s}}$$

(8)

$$h_{\mathrm{p}}^{\mathrm{s}}{=h}_{\mathrm{p}}^{\mathrm{f}}+{\int }_{T_{\mathrm{ref}}}^{T_{\mathrm{sat}}}c{p}_p\mathrm{d}T$$

(9)

$$h_{\mathrm{q}}^{\mathrm{s}} {=h}_{\mathrm{q}}^{\mathrm{f}}+{\int }_{T_{\mathrm{ref}}}^{T_{\mathrm{sat}}}c{p}_q\mathrm{d}T$$

(10)

where c_p is the specific heat capacity, J/kg·K; T_ref is the reference temperature, K; and h is the enthalpy of formation, J.

In this research, the auto-ignition is modeled using the transport equation for a thermal ignition species, as shown in Equation (11).

$$\frac{\partial \rho Y_{\mathrm{ig}}}{\partial t}+\nabla ·\left(\rho \overrightarrow{v}{Y}_{\mathrm{ig}}\right)=\nabla ·\left(\frac{{\mu }_{\mathrm{t}}}{S_{\mathrm{ct}}}\nabla Y_{\mathrm{ig}}\right)+\rho S_{\mathrm{ig}}$$

(11)

where Y_ig is the mass fraction of a passive species representing radicals which form when marine oil in the domain breaks down; S_ct is the turbulent Schmidt number; and S_ig is the source term for the ignition species.

The abovementioned source term for the ignition species can be developed as below:

$$S_{\mathrm{ig}}={\int }_{t=t_0}^t\frac{1}{T_{\mathrm{ig}}}\mathrm{dt}$$

(12)

where t₀ is the time at which marine oil is introduced into computational domain, s; T_ig is the ignition delay with the units of time, s.

When modeling the ignition delay of the leaked marine oil on the heated surface, the chemical reactions were allowed to take place when the ignition species reached a value of 1 in the domain. For the ignition delay model, the relevant equations were built into FLUENT, which reproduces Arrhenius correlations. If the ignition species is less than 1 when using the ignition delay model, the chemical source term is suppressed by not activating the combustion model at that particular time step. It means that the energy release will be delayed. This approach does not apply to solving for typically low-temperature chemistry, but it is reasonable for a high-temperature chemical model in current research. The equations related to the ignition delay model in this work can be shown by Equation (13).

$$T_{\mathrm{delay}}=\left(\frac{C_1+0.22\overline{S_{\mathrm{p}}}}{6N}\right)\exp \left[E_{\mathrm{a}}\left(\frac{1}{\mathrm{R}T}-\frac{1}{17,190}\right)+{\left(\frac{21.2}{p-12.4}\right)}^{e_{\mathrm{p}}}\right]$$

(13)

where T_deley is the ignition delay time, s; C₁ is a pre-exponential constant number with a value of 0.36; N is the operation speed of a device that forms a heated surface, m/s; E_a is the effective activation energy, kJ/mol; and e_p is the pressure exponent, Pa.

The equation for the effective activation energy E_a is expressed by Equation (14).

$$E_{\mathrm{a}}=\frac{E_{\mathrm{h}}}{CN+25}$$

(14)

where E_h is the activation energy, KJ/mol; CN is the cetane number.

2.2.3. Initial and Boundary Conditions

In this numerical procedure, in addition to a geometric model of the equipment and piping system, the oil supply pipeline was set up near the typical high-temperature heated surface in the fishing vessel’s engine room. The purpose was to simulate the ignition situation of marine oil leakage flowing to a heated surface when the oil supply pipeline is broken. This research focuses on the ignition process of leaked marine oil on a heated surface, as well as the influence of the interior environment of the fishing vessel’s engine room. Marine diesel is a kind of mixed oil with complex fuel components. It was found that the cetane (C₁₆H₃₄) in the marine diesel occupies the highest proportion, and this component is an important index in measuring the ignition performance. The equipment housing in the diesel engine system forms a local high-temperature heated surface, which was set at a height of 1.0 m from the ground in the numerical model. In addition, the size of the heated surface in the fishing vessel’s engine room was 0.5 m×0.5 m. According to the measurement of the real environment in the fishing vessel’s engine room, the initial environmental parameters are shown in

Table 2. Environmental parameters of engine room in fishing vessel.

Parameter	Internal Temperature (°C)	External Temperature (°C)	Relative Humidity (%)	Internal Airflow Velocity (m/s)	Atmospheric Pressure (MPa)
Value	45	32	60	1.0	0.1

below.

Since the environment of this numerical model is in a fishing vessel’s engine room, the computational domain was divided into a fluid domain and a solid domain. The solid domain represents the mechanical equipment in the fishing vessel’s engine room, and the surface of the main diesel engine equipment was set as a high-temperature heat transfer surface, that is, an initial fixed temperature was set. This simulation set different heated-surface temperatures ranging from 440 °C to 525 °C. The heated-surface temperature was set at an interval of 5 °C, for a total of 18 heated-surface temperature changes. According to the specific content involved in this research, the entrance of the fishing vessel’s engine room was set as the velocity inlet, which is usually suitable for the flow of incompressible fluids. The parameters to be set include material velocity and volume ratio flowing into the computational domain through the inlet. They were used in this numerical model to set the ventilation velocity and humidity conditions. Based on the airflow organization in the engine room, four different wind speeds were: 0 m/s, 1.0 m/s, 3.0 m/s, and 5.0 m/s, respectively. The outlet boundary was set as a pressure outlet, and no additional pressure value was added at the outlet, that is, the normal pressure was set. In this numerical modeling, there were two types of walls in the physical model of engine room constructed. One type of wall is the heated surface, which was set to a fixed temperature and heat flux. The other type of wall is the equipment and engine room’s wall, which was set as a fixed wall. For such walls, a fixed heat generator rate was set, which means how much heat is being added to the body from within. In order to further explore the influence of environmental humidity on the ignition process of leaked marine oil, three scenarios with different humidity conditions were set in the numerical model, with respective humidity values of 60%, 75%, and 85%. The data monitored during the simulation of the engine room model include the phase transition volume of the gas/liquid phase and the standard temperature of the fluid domain. The monitoring points were set up in different areas of the numerical model, especially in the area where the leaking marine oil heat meter ignition occurred. Thirty monitoring points were set up in the vertical space where the heated surface of the main diesel engine was formed, which is consistent with the experimental arrangement. Based on this, various results such as temperature data can be generated in the computational domain to meet the needs of this research.

3. Results and Discussion

3.1. Evaporation Mechanism of Marine Oil Ignition by Heated Surface

The liquid phase change of marine oil leakage on a heated surface of a fishing vessel’s engine room is directly related to the formation of a gas phase, so it is necessary to observe the liquid phase flow and change process of marine oil leakage. Figure 3 shows the image data collected during the experiment and numerical simulation. Figure 3a presents the evaporation process of leaked marine oil on a heated surface when there is no mechanical ventilation in the engine room. It is found that when marine oil is leaked on a heated surface, its phase state changes rapidly. The temperature of a typical heated surface inside the engine room is higher than the initial temperature of the marine oil. When the temperature difference between the typical heated surface and the leaked marine oil is less than 278 K, the solid–liquid interface is in a state of natural convection. Normally, the temperature of a typical heated surface of fishing vessel’s engine room is higher, which causes different forms of boiling when marine oil comes into contact with it. When the temperature difference between the heated surface and leaking marine oil reaches 278 K to 308 K, nucleate boiling occurs in the solid–liquid interface. If the typical heated-surface temperature of a fishing vessel’s engine room continues to rise, and the difference between it and temperature of the marine oil is greater than 308 K, it will enter a state of transition boiling. This state is between nuclear boiling and film boiling, and the heated-surface temperature at this time is higher than that of nuclear boiling. According to detection of the equipment surface in the engine room, it is found that the heated-surface temperature formed by the high-speed operation of mechanical equipment is usually several times or even ten times the temperature in the engine room. In the case of higher heated-surface temperatures of the fishing vessel’s engine room, the temperature difference between the heated surface and marine oil is larger. In this scenario, leaked marine oil will boil violently on the heated surface, and a vapor film will form on the oil surface. This state is called film boiling, which occurs at temperatures above the transition boiling condition. The formation of a gas phase layer between the marine oil and heated surface can be clearly observed from the image collected by the high-speed camera, and the formation time is very short. When film boiling occurs on the heated surface, the flammable vapor produces a different distance from the leaked marine oil surface. Since there is no mechanical ventilation in the engine room, the vapor flow of the leaking marine oil on the heated surface is slower. White vapor is produced by the heating of marine oil, and it accumulates above the heated surface. After the marine oil vapor is fully mixed with the surrounding air of the engine room, a combustible mixed vapor cloud is formed that has a clear cloud profile, as shown in Figure 3a.

The liquid phase volume of the marine oil was monitored, and its change is shown in Figure 3b. The movement of the liquid molecular flow of the marine oil on the heated surface of the engine room exists from the beginning and synchronizes with the gas phase change of the marine oil. Figure 3b shows the location of initial marine oil accumulation, and the thickness and accumulation range of the marine oil above the heated surface can be observed from the image. After the leaked marine oil is affected by the heated surface of the fishing vessel’s engine room, it gradually begins to flow to two sides. The two sides are more symmetrical, and the whole flow displays a horn shape. According to the numerical simulations, a small number of liquid phase molecules of the marine oil appear in the left air in the engine room, and their shape is basically similar to the marine oil vapor formation profile. In the space above the heated surface, the spot liquid phase molecules of the marine oil gradually increase. Subsequently, the liquid phase molecules in the air begin to disappear, and the leakage marine oil at the bottom shows a normal flow and fluctuation state. The liquid phase molecules located on the heated surface show a normal flow state, and the gas phase molecules in the environment of the engine room gradually increase. The change in the liquid phase medium after the oil leakage is heated on the heated surface shows that the liquid phase movement presents a symmetrical change at the initial stage. Subsequently, the leaked marine oil begins to flow slowly to both sides of the high-temperature heated surface, and the flow state is completely unconstrained. It can be seen from the phase diagram of the marine oil that there are some small fluctuations on the marine oil surface. Due to the small-diameter range of the marine oil accumulation areas, some of marine oil begins to flow to the areas in direct contact with the heated surface. At this time, the leaked marine oil layer at the heated surface is relatively thin. With the continuous flow of leaking marine oil, the thickness of the oil layer at the edge gradually increases. Meanwhile, the thickness of the oil layer in the middle decreases, and the marine oil reaches a pseudo-dynamic equilibrium state on the whole. When the marine oil is subjected to high-temperature heated-surface action to reach its saturation temperature, molecules overflow from the liquid to become vapor. In the environment of the fishing vessel’s engine room, it begins to diffuse, and a part of the oil vapor re-forms a liquid state after collision, but it does not completely fall onto the heated surface. There are a large number of liquid phase molecules in the air. The number of molecules that re-form the liquid phase is much smaller than that of molecules evaporating into the gas phase after being acted on by the heated surface, which has a strong promotion effect on ignition. Heated gaseous molecules of marine oil are transferred in two ways: the transport of components caused by the macroscopic overall flow of the fluid, or molecular diffusion, which includes the conventional diffusion motion from a region of high concentration to a region of low concentration and thermal diffusion caused by a temperature gradient, as shown in Equation (15).

$${\dot{\mathrm{m}}}_{\mathrm{A}}^{″}=Y_{\mathrm{A}}\left({\dot{\mathrm{m}}}_{\mathrm{A}}^{″}+{\dot{\mathrm{m}}}_{\mathrm{B}}^{″}\right)-\mathsf{\rho }{D}_{\mathrm{AB}}\frac{\mathrm{d}{Y}_{\mathrm{A}}}{{\mathrm{d}}_{\mathrm{x}}}$$

(15)

where ${\dot{\mathrm{m}}}_{\mathrm{A}}^{″}$ is the mass flux of component A per unit area, kg/(s·m²); Y_A is the mass fraction; and D_AB is the diffusion coefficient, m²/s.

Figure 3c,d show the process of evaporation and phase change of the leaked marine oil on the heated surface after mechanical ventilation, respectively. Figure 3c is the screen collected in the experiment, and Figure 3d is the cloud image generated by the numerical simulation. With the change in time, there are two distinct periods of growth of the marine oil evaporation. In the earlier period, the evaporation of the leaked marine oil in the engine room under low wind velocity conditions is higher than the change value under high wind velocity conditions. The analysis indicates that the influence of mechanical ventilation on the engine room environment increases after the mechanical ventilation is increased. This results in heat loss from the heated surface of the equipment and is not conducive to the endothermic evaporation process of the marine oil. Therefore, the marine oil ignition time will be shortened under low wind velocity in the engine room. When the inflection of marine oil evaporation is reached, the evaporation amount of marine oil will jump obviously. It is believed that this is due to the transition from natural evaporation to boiling evaporation. The marine oil in the fishing vessel’s engine room is transferred by the heated surface and begins to spread into the engine room environment after reaching the vaporization temperature. At this time, the ventilation in the engine room takes away the oil vapor phase medium above the heated surface, thus promoting the transformation of the liquid phase of the marine oil into the gas phase, and finally the evaporation amount of the marine oil above the heated surface increases with the strength of the wind force in the engine room. The evaporation rate of the marine oil leakage decreases gradually in the later period because the marine oil vapor gathers above the liquid phase surface, and the decrease in the concentration difference leads to a decrease in the rate. With the weakening of the mechanical ventilation in the fishing vessel, the smaller relevant ventilation volume takes away less oil vapor above the surface of the liquid phase, and the marine oil evaporation rate decreases accordingly. The results of the experiment and numerical simulation show high agreement.

Figure 4 shows the process in which marine diesel oil leaks to the heated surface, forming an initial ignition and causing flame propagation. Figure 4a presents the initial ignition of leaked marine oil when the heated-surface temperature reaches 460 °C. Figure 4b shows the initial ignition of leaked marine oil when the heated-surface temperature rises to 500 °C. When the typical heated-surface temperature of the engine room changes, there is a significant difference in the hot-surface ignition position of the marine oil. Marine diesel oil is a multi-component mixture: it is composed of C₁₂~C₂₃ alkanes and contains aromatic hydrocarbons and a small amount of organic sulfur, nitrogen oxides, and other substances. Once marine diesel oil is in contact with a high-temperature surface, it produces a large amount of white flue gas mixture in a short time. Related to the flow characteristics of marine diesel, the combustible gas mixture formed on the heated surface of marine diesel has a wide range. The evaporation products produced by marine diesel oil on a heated surface can be fully mixed with the air in a fishing vessel’s engine room. The mixed medium can form stable combustible mixed clouds in the vertical space above the heated surface. At this time, there is continuous heat transfer from the heated surface to the liquid surface, coupled with the thermal feedback effect around the engine room. As shown in Figure 4a,b, when marine diesel oil leaks for a certain period of time, the mixed vapor cloud experiences an instantaneous initial ignition above the heated surface. This instantaneous ignition phenomenon is difficult to directly observe and can only be captured through high-speed photography technology to collect the instantaneous impact, showing the occurrence location of the ignition. Because of the different heated-surface temperatures, the initial igniting positions of leaking marine oil are very different. The distance between the Ignition height and marine oil is closely related to the heated-surface temperature and environmental factors. After the initial ignition forms a core in the air, it gradually grows and produces a bright flame, as shown in Figure 4c,d. Due to the high initial ignition height, the flame spreads along this path to the remaining oil surface through the mixed vapor between the initial flame and the heated surface. If the initial ignition position is close to the heated surface, the remaining marine oil will directly ignite and form stable combustion behavior.

3.2. Ignition Probability of Marine Oil on Heated Surface

The ignition probability of leaked marine oil in the fishing vessel’s engine room was carried out with experimental tests. In these tests, diesel was selected as the type of marine oil, and the dosage was fixed at 15 mL. A total of 30 tests were performed to determine the typical heated-surface temperature of each identified fishing vessel’s engine room. Aimed at the experiment of the ignition probability of marine oil on a typical heated surface in the engine room, the temperature of the heated-surface platform was first raised to a fixed value. Secondly, the marine diesel oil needs to flow through the drainage tank to the heated surface. In this step, the amount of marine diesel oil was selected as 15 mL, and the heating step of the heated surface was 5 °C. After the marine diesel oil leaks onto heated-surface platform, there is a need to observe whether there is an ignition phenomenon to record the situation of marine oil on the heated surface. By observing the experimental phenomenon, it can be seen that marine diesel oil leaks onto the heated surface with a temperature of 465 °C, which does not result in ignition every time. Other hot-surface temperatures show different ignition times and flame behaviors. If the heated-surface platform continues to warm, at a constant temperature of 480 °C, the number of ignition times of leaked marine oil will be greatly increased. In experimental scenarios where the typical heated-surface temperatures are above 500 °C, marine diesel can be ignited nearly every time. The marine oil leaks above the heated surface formed by the machinery, and the occurrence of ignition behavior can be regarded as a random event. The ignition probability P_i of marine oil ranges from 0 to 1, which indicates that the ignition probability of marine oil occurring on the heated surface during the leaking process cannot be simply expressed as a linear function. In this research, the logic model was applied to represent ignition probability, as shown in Equation (16). The logical transformation corresponding to ignition probability can be expressed as a linear function of an independent variable through the change in function. The ignition probability problem should conform to the logistic regression function, which can be shown by Equation (17).

$$\mathrm{Logit}\left(P\right)=\ln \left(\frac{P}{1-P}\right)$$

(16)

$$P_{\mathrm{i}}=\frac{\exp \left({\beta }_0+{\beta }_{1}X\right)}{1+\exp \left({\beta }_0+{\beta }_{1}X\right)}$$

(17)

where β is the logistic regression coefficient; X is the linear function.

Figure 5 shows the ignition probability distribution of leaked marine oil on the heated surface in the fishing vessel’s engine room. In this research, marine diesel oil with an experimental dosage of 15 mL was taken as an example, and another variable was the high-temperature heated surface at different temperatures. The ignition probability of marine oil presents a gradually increasing trend with the increase in the typical heated-surface temperature. This suggests that the phenomenon of leaked marine oil ignited by the heated surface of equipment in a fishing vessel’s engine room is a probabilistic accident. Not every time the leaked marine oil touches the surface of equipment in the engine room will the ignition accident occur, but the ignition probability changes at different temperatures. With the temperature of a typical heated surface in the engine room increasing gradually, the ignition probability distribution of marine oil on a heated surface can be divided into three stages. During the first stage, the typical heated-surface temperature does not reach a high value, generally below 450 °C. In this heated-surface temperature range, the ignition probability of marine oil is accordingly low. Taking this experiment as an example, in the 30 igniting tests usually completed, about one to three ignitions occur, and the rest of the time the oil is not ignited. When the heated-surface temperature rises to more than 460 °C, the ignition probability of marine oil will fluctuate greatly. When the heated-surface temperature goes up to nearly 475 °C, the ignition probability of marine oil is 50%. This indicates that when marine oil leakage occurs in a fishing vessel’s engine room, if the temperature near the equipment surface is close to 475 °C, there is a 50% probability that a fire accident will occur. If the heated-surface temperature continues to rise to about 500 °C or above, the leaked marine oil can be almost completely ignited on the heated surface. This indicates that once the marine diesel oil leakage accident occurs in the fishing vessel’s engine room, and the heated-surface temperature of equipment exceeds 500 °C, ignition will inevitably occur. Whether it can further induce a fire accident in the engine room needs to be determined according to factors such as the amount of marine oil and surrounding items. Based on the fitting results of the experimental data, this research proposes a prediction model for ignition probability that is suitable for evaluating the ignition of leaked marine oil on a heated surface, as shown in Equation (18). The correlation coefficient R² of this prediction model reaches 0.973, which indicates that this model can assist in the investigation and judgment of fire accidents caused by marine oil leakage in a fishing vessel’s engine room.

$$P_{\mathrm{i}}=\frac{0.994}{1+\exp \left[-0.194\left(T_{\mathrm{s}}-T_{50\%}\right)\right]}$$

(18)

where T_s is the heated-surface temperature of fishing vessel’s engine room, K; T_50% is the hot surface temperature at 50% ignition probability, K.

3.3. Effect of Ventilation on Marine Oil Ignition in Engine Room

The mechanical ventilation equipment will be turned on during the daily operation of a fishing vessel’s engine room, and the ventilation volume will be adjusted according to different operation scenarios. This section focuses on the differences in wind velocity in the engine room under different ventilation rates and the change in the ignition process of leaking marine oil. According to airflow organization in the fishing vessel’s engine room, four types of wind velocity in the engine room were set in this research: 0 m/s, 1.0 m/s, 3.0 m/s, and 5.0 m/s. The change in airflow organization in the engine room is directly reflected in the change in ignition characteristics of leaked marine oil. Figure 6 shows the variation in ignition height of leaked marine oil under different wind velocities in the engine room. The solid points in Figure 6 represent experimental data, and the hollow points represent numerical simulation data. On the whole, it can be seen that in the scenario of the same wind velocity, the ignition height of the marine oil on the heated surface shows a downward trend with the increase in the heated-surface temperature. This indicates that the increase in the heated-surface temperature causes the marine oil above the heated surface to evaporate rapidly, and the flammable vapor it forms quickly mixes with the air near the surface. Due to the high temperature of the surface itself, it can provide a lot of heat in a short time, so the ignition occurs in the area closer to the heated surface. As the heated-surface temperature continues to rise, it is found that the ignition height of the marine oil will no longer fluctuate greatly, but basically maintain in the area of 0.29 m to 0.51 m above the heated surface. According to the experimental data, there is a fitting relationship between the ignition height of the marine oil and the heated-surface temperature, as shown in Equation (19). The correlation coefficient R² of the present model reaches 0.944. The fitting relationship can be used to determine the initial ignition location of marine diesel at different heated-surface temperatures. In the scenario where the typical heated-surface temperature of a fishing vessel’s engine room is consistent, the ignition height of leaked marine oil increases with the wind condition strength in the engine room. The mechanical ventilation in the engine room enhances, which restricts the ignition process of the marine oil on the heated surface to a certain extent. Since the air supply and exhaust system in fishing vessel’s engine room supply air at the bottom and exhaust air at the top, more marine oil vapor is carried higher by the airflow movement. Under enhanced ventilation conditions, marine oil vapor forms a volume fraction with ignition conditions at a higher position above the heated surface. Once the ignition occurs, the location of the ignition occurrence is far from the heated surface, which is significantly different from the accident scenario without mechanical ventilation.

$$H=H_0+0.149\exp \left(-\frac{T_{\mathrm{s}}}{\mathsf{\pi }\mathrm{g}}\right)$$

(19)

where H is the height of marine oil ignition occurrence on the heated surface, m; H₀ is the surface height of the marine oil, m; T_s is the heated-surface temperature in the engine room, K; and g is the gravity acceleration, m/s².

The ignition delay time of marine oil is of great significance in characterizing its accidental ignition characteristics. The ignition delay time refers to the time required for the system of the combustible vapor mixture to reach an ignition condition, from the initial temperature rise to the initial ignition phenomenon. Different marine oils show different ignition characteristics on high-temperature heated surfaces, and the time required for ignition is varied. The leakage of diesel on a heated surface and the occurrence of local ignition are two basic necessary premises. The first is that there is a suitable mixture ratio of vapor/air in the ignition area, which can promote the formation of an initial flame. On the other hand, environmental factors in the fishing vessel’s engine room are important. After the initial flame is formed, such a flame can stabilize itself in the region where it occurs at a low-velocity flow field. Then, through flame transmission, it can achieve stable propagation to an unburned gaseous mixture in space. There are complex influencing factors in the fishing vessel’s engine room that are different from the time required for the initial ignition of marine oil in other scenarios. The experimental tests and numerical simulations of the relationship between the ignition time of leaked marine oil and heated-surface temperature were carried out under different wind velocity conditions. Through the combination of experimental and numerical results, this research explores the change rule of time required for marine oil ignition above the heated surface under different engine room wind velocities. Figure 7 shows the change in the ignition delay time of leaked marine oil with heated-surface temperatures in different engine room wind velocity scenarios. The ignition delay time represents the time required for the initial ignition of a vapor/air mixture on the heated surface, which directly determines the risk degree of the engine room fire accident. It can be seen from the Figure 7 that under the same wind condition, the ignition delay time of the leaked marine oil on the heated surface shows a downward trend with increasing heated-surface temperature. This is because the vapor/air mixture is acted upon by the increasing heated-surface temperature, which undergoes a violent chemical reaction in the region closer to thermal boundary. This violent chemical reaction take place precisely because the combustible gaseous mixture is heated to a certain temperature by the surrounding heat source, and such a heat release is greater than the heat release to the surrounding environment. It ultimately leads to the automatic acceleration of the chemical reaction, forming the initial flame. Meanwhile, Figure 7 indicates that at the same heated-surface temperature, the enhanced engine room’s wind velocity results in a longer ignition delay time for the leaked marine oil. Due to the increase in the amount of ventilation in the fishing vessel’s engine room, the wind velocity increases, which causes the air distribution in the area around the heated surface to change. This change causes the marine oil vapor to form a concentration higher than the heated surface, so the time required for the ignition of the vapor/air mixture increases. It brings about new fire accident risks in the fishing vessel’s engine room. When the ventilation condition in the engine room is enhanced, the ignition time after the leaked marine oil contacts the heated surface will be prolonged, and it is more necessary to strengthen the monitoring and control measures of the area where ignition may occur.

3.4. Effect of Humidity on Marine Oil Ignition in Engine Room

In order to further explore the influence of environmental humidity in the fishing vessel’s engine room on the ignition process of leaked marine oil, three scenarios with different humidity conditions were arranged, and the humidity values were set at 60%, 75%, and 85%. The humidity values correspond to different scenarios of a fishing vessel’s port stay and sea operation. Figure 8 presents the variation in ignition height of the leaked marine oil on the heated surface under different humidity conditions in the fishing vessel’s engine room. When the heated-surface temperature is the same, the ignition height of the leaking marine oil goes up with the increase in humidity. However, the magnitude of this change is not as significant as the effect of increased wind velocity on the ignition height. This indicates that the humidity in the air inhibits the upward transfer of heat; however, the degree of inhibition is limited. When the typical heated-surface temperature is 460 °C, the ignition height of marine oil at 60% humidity is approximately 0.481 m. If the humidity in the engine room is moderately increased to 75%, the ignition height of the marine oil on the heated surface increases to 0.574 m, with an increase rate of 19.33%. If the humidity in the engine room is further increased, the ignition height of the leaking marine oil at this time goes up to 0.652 m. The numerical simulations have good agreement with the experimental results, and the CFD-based data are lower than the experimental data. When the typical heated-surface temperature of the fishing vessel’s engine room rises to 475 °C, the ignition probability of the marine oil is close to 50%. At this heated-surface temperature, the ignition height of the leaked marine oil ranges from 0.401 m to 0.502 m, depending on humidity. Compared with the scenario where the hot-surface temperature is lower, the ignition position of the leaking marine oil is closer to the heated surface of machinery. With the further heating of the surface, the temperature reaches about 500 °C, and the marine oil above this temperature will achieve a fully ignited state. According to experimental data, under different humidity conditions, the ignition heights of the marine oil above the heated surface are 0.289 m, 0.335 m, and 0.413 m, respectively. The increase in the heated-surface temperature causes the marine oil leaking on the heated surface to evaporate rapidly, and the flammable vapor it forms quickly mixes with the air near the surface. The heated surface can provide a lot of heat within a short time, so the ignition behavior occurs in the area closer to surface. With the continuous increase in the surface temperature in the engine room, it is found that the ignition occurrence of the marine oil does not change greatly. Based on the fitting of comparative results, this research proposes a prediction model for the ignition occurrence of marine oil on a hot surface that is applicable to different humidity variation scenarios, with an R² of 0.926. As shown as Equation (20), the heated-surface ignition area can be determined when the marine oil leakage in the fishing vessel’s engine room enhances the humidity condition.

$$H=H_0+0.292\exp \left(-\frac{{7T}_{\mathrm{s}}}{5\mathsf{\pi }\mathrm{g}}\right)$$

(20)

The ignition delay time of marine oil is composed of liquid evaporation time and chemical reaction induction time. The difference in the ignition delay time represents the difference in the ignition characteristics of marine oils. Meanwhile, different characteristics of the ignition delay time of marine oils directly determine the protection focus in the fire process of a fishing vessel’s engine room. In this research, the heated-surface ignition time of marine diesel oil was measured experimentally, and the data collected by thermocouples and images captured by a high-speed camera were used to determine the ignition delay time. The temperature distribution and time variation of the marine oil vapor above the heated surface were obtained by thermocouple data collection. The images of the ignition moment and evolution process of the initial flame were collected with millisecond accuracy by image acquisition with a high-speed camera. Figure 9 shows the change rule of the ignition delay time of the leaking marine oil above the heated surface when the humidity of the fishing vessel’s engine room changes. It is found that the increase in humidity in the engine room has a significant impact on the ignition delay time of the marine oil on the heated surface. Taking the heated-surface temperature of 465 °C as an example, when the humidity is 60%, the ignition delay time of the leaking oil is approximately 3.36 s. When the humidity increases to 75% and 85%, the ignition delay times of the marine oil on the heated surface go up to 7.05 s and 8.49 s, which are nearly twice as high. This change rule is particularly obvious when the heated-surface temperature is relatively low, and the increase rate decreases gradually when the heated-surface temperature rises. When the typical hot-surface temperature rises to 505 °C, the ignition delay times of marine oil under different humidity conditions are 0.84 s, 2.10 s, and 3.57 s, respectively. The water in the fishing vessel’s engine room absorbs environmental heat and converts it into vapor, and the leaked marine oil above the heated surface absorbs the environmental heat and converts it into flammable vapor. Since the specific heat capacity of seawater is higher than that of marine oil, the relevant temperature rises due to absorbing the same heat is lower than that of marine oil. Therefore, in the case of absorbing the same heat, the seawater and engine room environment have a more obvious temperature difference. This results in seawater absorbing more heat than marine oil during convective heat transfer. In the case of the same heat transferred by the heated surface in the engine room, the thermal feedback of the engine room is inhibited to a certain extent with the increase in humidity. As a result, the ignition delay time of leaking marine oil on the heated surface is increased. According to the analysis of the experimental and numerical results, the change in the ignition delay time on the heated surface after the marine oil leakage is predicted by the change in the heated-surface temperature, which has an important role in preventing a fire accident in the fishing vessel’s engine room.

4. Conclusions and Recommendations

The leaked marine oil ignited by the heated surface of machinery in the fishing vessel’s engine room provides deadly risks that significantly change the environmental conditions. This research reveals the effects of temperature, wind velocity, and humidity on the heated-surface ignition characteristics of leaked marine oil in a fishing vessel’s engine room. It explores the new correction relationship for predicting the heated-surface ignition probability and ignition occurrence of marine oil with corresponding engine room condition parameters. The noteworthy findings of this study are as follows:

• Elevated heated-surface temperatures lead to significantly different initial ignition positions of marine oil, and the distance between the ignition height and oil is closely related to heated-surface and environmental factors. Higher initial ignition occurs without mechanical ventilation, and the flame spreads along a path to the remaining oil surface. A lower ignition position occurs under ventilation, forming stable combustion.
• The ignition probability of marine oil shows a gradual increasing trend with elevated heated-surface temperatures. The prediction model developed in Equation (14) can be applied to assist in the investigation of fire sources caused by marine oil leakage in a fishing vessel’s engine room.
• The ignition height of the marine oil on the heated surface presents a downward trend with the increase in heated-surface temperature. This indicates that at the same heated-surface temperature, the enhanced engine room’s wind velocity results in a longer ignition delay time for the leaked marine oil.
• The engine room’s humidity inhibits the upward transfer of heat; however, the degree of inhibition is limited. It is found that the increase in humidity in the fishing vessel’s engine room has a significant impact on the ignition delay time of the marine oil on the heated surface.

For the scientific prevention and accurate measures of fire accidents in fishing vessel engine rooms, corresponding fire prevention measures should be taken according to the leaked marine fuel types induced by different physical hazards in engine rooms. At present, the ignition source is the energy source in the initial stage of the combustion reaction, and it is necessary to induce a fire accident in a vessel’s engine room. Based on our findings, the prevention and elimination of the heated surface is the most critical link to cut off the reaction chain in a fishing vessel fire. This work expands the existing limitations of engine room fire assessment, providing more support for ignition source control in enhance fishing vessel operation safety. Based on the findings of the experimental and numerical works, combined with the research on the initial ignition mechanism and flame propagation characteristics of leaked marine diesel in engine rooms, the following recommendations are proposed:

• Identify leakage sources of fuel. This requires inspecting marine diesel and lubricant oil piping lines for loose fittings and missing bolts on flanges and non-metallic hoses in areas where the high temperature takes place. The safety supervisor should regularly assess these components to identify potential leakage sources of marine fuel.
• Shield potentially heated surfaces. As the materials used to insulate the heated surfaces will degrade in marine operation processes, resulting in them becoming soaked in oil, it is recommended that they are checked consistently. Even if the insulation of the piping system shows to be in good condition, careful checking is essential in finding hidden, smaller heated spots which could trigger an initial ignition if in contact with marine fuel.
• Transform the prevention model. As the ventilation conditions in an engine room are enhanced, the ignition position of leaked fuel above a heated surface gradually increases. This means there should be different requirements for the location of fire monitoring sensors. It is recommended to add thermographic examination or temperature sensors above the potential heated surface of equipment, which can be used to monitor the formation of initial fire nucleus.
• Regulate the ventilation system. The strength of the ventilation condition increases the evaporation rate of leaked diesel on a heated surface and the particle velocity in the environmental flow field. As a result, the evaporative phase change rate of leaked diesel on a heated surface in an engine room will be promoted. Decision makers should ensure that the ventilation system is closed to avoid airflow conditions that promote the evaporation of leaked marine fuel and accelerate the possibility of fire.
• Perform evaluation under multiple indexes. Due to the increase in the humidity in a vessel’s engine room, the ignition delay time of leaked fuel above a heated surface increases. Because hazards still exist, it is recommended that monitoring data based on multiple indexes should be established to evaluate environmental safety. The influence of humidity on the evaporative phase variable of fuel on a heated surface and the rise in the engine room temperature must be considered comprehensively. Due to the inhibition effect of humidity on engine room temperature and evaporation phase transition, it is necessary to strengthen safety inspections and find the existence of leaked fuel in advance to better avoid fire risks.