*2.6. Metal Hose Ignition Probability and Wind Rose Data*

Frequency analysis is a hierarchical method for calculating the probability of a fire or explosion occurring in a given structure. It can be used to identify equipment failure, faults, operating conditions, environmental conditions, and human error that contribute to fire and explosion accidents. FTA and ETA were used to calculate the failure and ignition probabilities to determine the probabilities of a pool fire, fireball, and jet fire for one year.

For FTA, the ignition probability of various flammable materials in the event of a metal hose leak was taken from the literature [38–40]. Table 3 presents the probability of a fire occurring for different release rates. The frequency analysis results for metal hoses are detailed in Section 3. Data from the UK Health and Safety Executive (HSE) were referenced to determine the failure probability of the metal hose, which was set to 1.0 × <sup>10</sup>−<sup>2</sup> [21,38].

**Table 3.** Ignition probability of a liquid according to the release rate.


A wind rose graphs the frequencies of the wind direction and speed at an observation point. It can be used to show the long-term average wind direction and speed [41,42], which are important factors that affect the location and extent of damage caused by a jet fire [42–44]. In this study, the wind direction was expressed as a probability, and it was combined with the frequency analysis for the fire risk assessment. Twelve wind directions were selected, and their probabilities were set according to the wind data from Goheung, which is where the aerospace facility is located. The software WPLOT [45] was used to analyze the wind rose data. Figure 5 shows the wind rose for Goheung. Table 4 presents wind data sourced from KMA from 2009 to 2018 in 12 directions according to frequency. Wind rose data is the probability data about wind direction and speed. For example, in Table 4, when the wind direction is 345 to 15 degrees, the probability of blowing with a wind speed of 0 to 1 m/s can be expressed as 26.6538%.

Wind speeds >5 m/s accounted for less than 1% of the 10 years of data and thus, were excluded from the analysis. Although a higher wind speed can lead to greater damage, the frequency was so low that the risk was negligible. The remaining wind speed data were used to quantify the probability of a jet fire in each wind direction. The total probability of a jet fire was determined by summing the probabilities in each wind direction. According to the wind rose data for this aerospace facility, the highest probability of a jet fire was at 345–15◦ for a wind speed of 0–1 m/s and at 315–345◦ for wind speeds of 1–2, 2–3, 3–4, and >4 m/s.

**Figure 5.** Wind direction and speed at the aerospace facility in Goheung (Korea Meteorological Administration).


**Table 4.** Probability of jet fire, according to wind direction and speed.

#### **3. Consequence Analysis**

For the consequence analysis, a kerosene jet fire was assumed to occur under the following conditions:


Table 5 presents the fire probability for a metal hose with a hole diameter of 13 mm and release rate of 10.93 kg/s, while Table 6 presents the radii for 99%, 50%, and 1% fatality when a jet fire occurs with a leakage flow rate of 10.93 kg/s. According to Equations (5) and (6), the 99%, 50%, and 1% fatality areas and radii were 493.87 m2 and 3.20~12.94 m, 518.01 m<sup>2</sup> and 12.94~18.23 m, and 1027.71 m2 and 18.23~25.68 m, respectively. When the

analysis of the results was performed, it was confirmed that the lower the fatality, the lower the degree of damage, but the wider the damage range.

**Table 5.** Metal hose fire probability.


**Table 6.** Fatality area and distance from jet fire point according to metal hose release rate.


#### **4. Risk Analysis**

IR was obtained from the results of the frequency analysis and consequence analysis, which is calculated as follows:

$$IR = \theta \cdot p\_{\text{loc}} \cdot f\_{\text{wh}} \cdot fat\_{\text{jet}} \cdot fre \cdot p\_{\text{wind}}res \tag{10}$$

where *θ* is the overall fraction of time that a person is in a given area, *ploc* is the probability that the person is at a location, *fmh* is the fire probability of the metal hose, *f atjet fire* is the fatality probability of a jet fire, and *pwindrose* is the probability that the jet fire will be in one of the windrose (obtained through WPLOT).

For the risk analysis, a kerosene jet fire individual risk was assumed to occur under the following conditions:


In this study, Equation (10) was created based on Equation (2). Although Equation (2) assumed to have a constant velocity distribution in all wind directions, Equation (10) used in this study that considered both wind direction and speed to calculate the individual risk result value.

Table 7 presents the values of the terms in Equation (10). Figure 5 lists the IR ranges by applying different colors to visually express the IR result values derived through Equation (10). The colors in Figure 6 are used in Figures 7–9 to visualize IR according to the wind speed and direction for each fatality probability. Figures 7–9 show the results for 99%, 50%, 1% IR. For Table 7, all probability values of the wind rose are omitted since they are listed in Section 2.5.

**Table 7.** Calculation of IR for different fatality probabilities.



**Figure 6.** IR contours.

**Figure 7.** IR of 99% fatality at different wind directions and speeds: (**a**) 0–1 m/s, (**b**) 1–2 m/s, (**c**) 2–3 m/s, (**d**) 3–4 m/s, and (**e**) >4 m/s.

**Figure 8.** IR of 50% fatality at different wind directions and speeds: (**a**) 0–1 m/s, (**b**) 1–2 m/s, (**c**) 2–3 m/s, (**d**) 3–4 m/s, and (**e**) >4 m/s.

**Figure 9.** IR of 1% fatality at different wind directions and speeds: (**a**) 0–1 m/s, (**b**) 1–2 m/s, (**c**) 2–3 m/s, (**d**) 3–4 m/s, and (**e**) >4 m/s.

The IR results were calculated for the areas corresponding to 99%, 50%, and 1% fatality in each wind direction. The total IR was calculated for each wind speed. For the case of 99% fatality, the highest IR was at 345◦–15◦ for a wind speed of 0–1 m/s and at 315◦–345◦ for wind speeds >1 m/s. The lowest IR was at 255◦–285◦ for a wind speed of 0–1 m/s, 75◦–105◦ for wind speeds of 1–4 m/s, and 15◦–45◦ for wind speeds >4 m/s. For the case of 50% fatality, the highest IR was at 345◦–15◦ for a wind speed of 0–1 m/s and 315◦–345◦ for wind speeds >1 m/s. The lowest IR was at 255◦–285◦ for a wind speed of 0–1 m/s, 75◦–105◦ for wind speeds of 1–4 m/s, and 15◦–45◦ for a wind speed >4 m/s. For the case of

1% fatality, the highest IR was at 345◦–15◦ for a wind speed of 0–1 m/s and 315◦–345◦ for wind speeds >1 m/s. The lowest IR was at 255◦–285◦ for a wind speed of 0–1 m/s, 75◦–105◦ for wind speeds of 1–4 m/s, and 15◦–45◦ for a wind speed >4 m/s. The IR showed the same tendencies at 99%, 50%, and 1% fatality. The total IR was high at 0–1 m/s, which was the most frequent wind speed; this indicates a high probability of IR when a jet fire occurs.

#### **5. Results and Discussion**

In Section 4, individual risk results were calculated by considering the 99%, 50%, and 1% fatality ranges and wind direction and speed. Furthermore, in the case of Section 4, individual risks for all wind directions, speeds, and fatality ranges were, respectively, calculated and presented. In Section 5, the total individual risk considering wind speed and fatality ranges is calculated, and the safety of the aerospace facility is confirmed when metal hose jet fire occurred through comparison with the HSE risk standard. The HSE risk standard is a criteria created to ensure plant safety from unexpected accident that does not generate damage to workers, the environment, or assets [46,47].

HSE risk standard follows the ALARP (As Low as Reasonably Practicable) standard. As shown in Figure 10, ALARP is set for the safe operation of chemical facilities and plants, and there is a difference between the risk standard for workers and for the public [48–50].

**Figure 10.** HSE ALARP standard.

As a result, the total individual risk result value considering fatality and wind speed was shown in Figure 11, and it was found that the individual risk decreased as the wind speed increased and the fatality decreased. In the case of wind speed, due to the probability decreasing as the wind speed was increased, the *pwindrose* of Equation (10) was decreased and, consequently, the individual risk was lowered. Similarly, in the case of fatality, due to the fatality decreasing as the *f atjet fire* of Equation (10) was decreased, consequently, the individual risk was lowered. Finally, when compared with the ALARP standard, all individual risks were included in the ALARP standard regardless of wind speed and fatality. The individual risk from jet fire in an aerospace facility was calculated that maximum risk is 3.35 × <sup>10</sup>−5, minimum is 1.49 × <sup>10</sup>−6. According to this study's results, when a jet fire occurred from a metal hose, the risk satisfied tolerable criteria. Therefore, enhanced firewalls, extinguishing systems, and emergency shut off systems are needed to prevent jet fire accidents and to satisfy acceptable criteria. If the same accident occurs in another area, a jet fire risk assessment can be performed by implementing the same method using the wind data of the area where the rocket launch facility is located.

**Figure 11.** Total IR values according to fatality and wind speed.

In this study, the quantitative risk assessment according to the wind rose of the aerospace facility where a jet fire in a metal hose occurs was performed. Further studies should be conducted including the social risk, which is the probability of injury to the entire population in hazardous facilities when a fire occurs near an ignition source.

#### **6. Sensitivity Analysis**

Sensitivity analysis is an evaluation method that changes how the individual parameters affect the optimal solution. Sensitivity analysis was performed using *ploc* as a parameter. As a special and high-risk facility, the aerospace facility is not required for many working people, and it was an important factor in calculating individual risk results for metal hose jet fires in aerospace facilities, as well as fire and explosion scenarios that may occur in aerospace equipment. Therefore, in this study, sensitivity analysis was calculated by considering *ploc* among the various parameters used in the individual risk equation of metal hose jet fire according to wind direction and speed. In addition, the value of *ploc* was the average standard of a working person in the aerospace facility, and as such sensitivity analysis was performed by selecting for higher and lower values than the average standard that was 0.88.

For the sensitivity analysis, according to the total individual risk it was assumed to occur under the following conditions:


As a result, Figure 12 shows the sensitivity analysis considering the *ploc* value. In the case of the *ploc* value is higher than the average standard, when the maximum risk is 3.81 × <sup>10</sup>−<sup>5</sup> and the minimum is 9.71 × <sup>10</sup><sup>−</sup>6. When the *ploc* value is lower than the average standard, the maximum risk is 1.90 × <sup>10</sup>−<sup>5</sup> and the minimum is 4.86 × <sup>10</sup>−6. Finally, this

study confirmed that individual risk satisfied the tolerable criteria even if the *ploc* value

**Figure 12.** Sensitivity analysis considering to the *ploc*.

#### **7. Conclusions**

was maximum.

In this study, a risk assessment was performed on a jet fire at an aerospace facility considering the wind direction and speed. A consequence analysis was performed based on the properties of the metal hose, kerosene, and jet fire. Next, the IR was calculated for the wind rose when a jet fire occurred in a metal hose. The following conclusions can be drawn:


To summarize the results of this study, the major risk component is the metal hose, and the maximum IR in aerospace facility satisfies HSE ALARP standard. Further, using these results, it is possible to strengthen safety by applying additional firewalls, extinguishing systems, and emergency shut off systems.

**Author Contributions:** Conceptualization: I.S.Y. and E.S.J.; methodology: H.J.K. and K.M.J.; formal analysis: H.J.K.; data curation: H.J.K. and K.M.J.; writing-original draft preparation: H.J.K., K.M.J. and E.S.J.; writing-review and editing: H.Y.O. and S.I.K.; visualization: H.J.K.; supervision: E.S.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data is contained within the article.

**Acknowledgments:** This work was supported by the Korea Space Launch Vehicle (KSLV-II) funded by the Ministry of Science and ICT (MSIT, Republic of Korea).

**Conflicts of Interest:** The authors declare no conflict of interest.

### **Nomenclature**


#### **References**

