See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/329571016

Study on Water Jet Trajectory Model of Fire Monitor Based on Simulation and
Experiment

Article  in  Fire Technology · December 2018

DOI: 10.1007/s10694-018-0804-1

CITATION READS

1 1,118

4 authors, including:

Jinsong Zhu Wei Li

China University of Mining and Technology China University of Mining and Technology
7 PUBLICATIONS   3 CITATIONS    73 PUBLICATIONS   573 CITATIONS   

SEE PROFILE SEE PROFILE

Zhao Ge
China University of Mining and Technology
6 PUBLICATIONS   3 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Transfer learning based fault diagnosis for rolling bearing under varying operation conditions View project

National Key R&D Program of China View project

All content following this page was uploaded by Jinsong Zhu on 14 December 2018.

The user has requested enhancement of the downloaded file.

Fire Technology
Ó 2018 Springer Science+Business Media, LLC, part of Springer Nature.
Manufactured in The United States
https://doi.org/10.1007/s10694-018-0804-1

Study on Water Jet Trajectory Model

of Fire Monitor Based on Simulation
and Experiment

Jinsong Zhu, Wei Li*, Da Lin, and Ge Zhao, School of Mechatronic

Engineering, China University of Mining and Technology, Xuzhou 221116,
People’s Republic of China

Received: 25 June 2018/Accepted: 4 December 2018

Abstract. Water is required to reach the ignition location accurately to ensure rapid
fire extinguishing in the process of automatic fire suppression. In this paper, a water
jet trajectory model was established based on the Moving Particle Semi-implicit
(MPS) method, the experimental trajectory data were also collected through the
experimental platform system and compared with the simulation trajectory data.
Under low initial velocity, the results of simulation matched well with the experimen-
tal data while there were large errors under high velocity. It was found that air resis-
tance was the main reason for deviation, the longer flight time caused by the high
initial velocity and the bigger stressed area after the water jet break-up were main
causes for the greater impact from air resistance. The binary function interpolation
compensation method was used for error compensation and the jet angle and initial
velocity were selected as error factors. The trajectory model was corrected based on
existing simulation and experimental data and verified through unused data. The
results showed that the revised trajectory model can accurately predict the position of
the jet trajectory and provide data support for automatic fire suppression systems.

Keywords: Automatic fire suppression systems, Water jet trajectory model, MPS method, Binary
function interpolation compensation

1. Introduction
Autonomous fire suppression system is an active research area to increase the
effectiveness of firefighting operations in large building space, such as airport
lobby, stadium, warehouse, exhibition halls and other important areas [1]. In gen-
eral, autonomous fire suppression systems are installed on high positions of the
covered space which consists of four parts: a flame and smoke detector, micro-
computer, control motor and fire nozzle. Once the suspicious fire or smoke area
comes into the detector’s field of vision, the system will identify the fire area and
calculate its position relative to the fire nozzle based on the recognition and loca-
tion algorithm respectively. The structural components and the workflow of the
autonomous fire suppression system are as shown in Fig. 1. The system relies on a

* Correspondence should be addressed to: Wei Li, E-mail: liwei_cmee@163.com

1
 Fire Technology 2018

reliable and stable ignition identification and location algorithm as well as the
accurate trajectory model matched with fire monitor operating parameters. This
article focuses on establishing and improving a jet trajectory model to improve the
extinguishing accuracy and the automation degree of automatic fire extinguishing
systems. Most previous researches focus on fire and smoke detection while few
study about the effect on fire suppression by the water shot out of the nozzle has
been carried out. For a complete automated firefighting process, it is not only nec-
essary to focus on how to identify and locate the fire but also how to make the
fire extinguishing agent reach the location of fire accurately, otherwise, everything
is in vain. Work on fire extinguishing performance experimental study of the
sprinkler system for fire protection in large open spaces has been conducted [2],
while systematic studies on the jet ratewater flowdrop size and distribution are
inadequate. The fluid trajectory simulation based on fluid mechanics has not pro-
duced very satisfactory results, and the main reason is that there is no reasonable
calculation model to simulate the trajectory change after jet broken [3].
For fire protection system research, the focus is on several important parame-
ters such as the trajectory of fire extinguishing agent and its placement. A. P. Hat-

Figure 1. Structural components and workflow of the autonomous

fire suppression system.
Study on Water Jet Trajectory Model

ton et al. proposed a water jet computer simulation in still air conditions and pro-
vided a useful design aid for offshore firefighting [4]. Min yong-lin proposed a new
firefighting jet trajectory model based on theoretical calculations. However, the
drag coefficient and the changing jet cross-section coefficient severely restricted the
accuracy of simulation results, and it was difficult to determine these coefficients
[5]. The fire extinguishing characteristics of a sprinkler system were studied experi-
mentally by Fan Wu et al [6]. W. M. Zhang studied the aerated water jets at a 45
into the air experimentally, and it was found that the injection of air into water
jets would significantly accelerate water jet breakups in the air [7]. Jong-Hwan
Kim et al. studied the fire source identification and fire extinguishing path of intel-
ligent firefighting robots, the robot was very close to the fire so that water can
easily reach the fire with the fire hose installed on the robot [8]. Joshua G. McNeil
carried out complete research on the automatic fire extinguishing system, includ-
ing fire source identification and positioning based on infrared cameras, fire jet
trajectory identification with visual feedback control of fire monitors, and a simple
trajectory model was developed as well [1, 9, 10]. Tao Chen et al. verified that the
fire extinguishing system installed at a high place could well achieve the fire con-
trol in the covered area through large number of experiments [2]. Tatsuya Miya-
shita built a tank fire extinguishing effect evaluation model based on the
simulation and experimental data with the help of the office software [11–13].
More studies found that the jet angle, initial velocity, fire monitor structure, and
external environment were all important factors determining the jet trajectory [14–
16]. For the installed fire protection system, the initial angle and velocity of the
fire monitor are adjusted to match the fire burning in different locations.
In this paper, the trajectories of the jet were simulated based on the MPS
method and some of the experiments were carried out with the same simulation
parameter. The simulation and experimental trajectory well-matched by contrast
the experimental and simulation data and the cause of the error were analyzed. In
order to improve the accuracy of the trajectory model, an error compensation
method based on velocity and angle was proposed. The new experimental and
simulation data were used to verify the reliability of the method. At the same
time, the feasibility of data support used in the automatic fire extinguishing sys-
tem based on simulation and error compensation method was also analyzed.
Corresponding to the flow chart at the right bottom of the Fig. 1, the organiza-
tion of this article is as follows: Sect. 2 is water jet trajectory simulation study
based on the MPS method. Experimental study with the same simulation parame-
ters presented in Sects. 3 and 4 is about the error analysis between simulation and
experiment. An error compensation model is proposed and validated simultane-
ously in Sect. 5. Finally, Sect. 6 concludes the paper with some remarks and indi-
cates future study directions.
 Fire Technology 2018

2. Water Jet Simulation

2.1. Principle of the MPS Method
The jet trajectory model was constructed by the MPS method which is a meshless
method based on Lagrange particle. The MPS Method was first proposed by
Koshizuka in 1995 as a meshless method for calculating incompressible fluid
motion [17]. The MPS method has been widely used in complex fluid such as free
surface flow including splash, as well as gear oil behavior, and it can also handle
coupled analysis of fluids and rigid bodies or fluids and powders [18–20]. MPS
method was used to track the kinematic and dynamic characteristics of particles
in real time while the particles interacting with each other through the kernel
functions. The kernel function is used to calculate the force between the particle
and other surrounding particles as well as the force depending on the distance
between particles. The longer the distance, the greater the kernel function value,
and vice versa. However, the scope of force between particles was limited to a dis-
tance re to ensure the computational efficiency, as shown in Fig. 2.
The Principle of the MPS is based on the Lagrangian method, including the
mass of fluid motion and the conservation of momentum equation, which can be
expressed as:

dq
þ qru ¼ 0 ð1Þ
dt

du
q ¼ rP þ lr2 u þ qF ð2Þ
dt

Figure 2. Particle force analysis diagram in MPS method (G

represents the gravity of the particle, F represents the air resistance
of the particle, re represents the maximum range of the kernel
function, r represents the distance between the particles).
Study on Water Jet Trajectory Model

where u is the fluid velocity, q is the fluid density, l is the dynamic viscosity coef-
ficient and P is the fluid pressure.
The water particles are affected by the external force F:

1
F ¼ CD qair S ð3Þ
2m

where the mass of particle is m, the coefficient of air resistance CD ¼ 0:5, the den-
sity of air is qair and the projected sectional area is S.
Here is the kernel function established by Shakibaeinia [20], which is expressed
as:
(r
rp
e
1 rp < re
xðrp Þ ¼ ð4Þ
0 rp  re

where the distance between particles is rp and the interaction radius is re (12 times
as large as the diameter of particle).
In the MPS method, each calculation step was divided into explicit and implicit
steps. The first step considered the effects of gravity and boundary conditions,
performed the first explicit correction of the particle’s velocity and coordinates.
Then, in order to make the particles density number return to the initial state to
satisfy the incompressible initial hypothesis, the pressure Poisson equation was
established with the change of the particles density number caused by the first
step. After acquiring the pressure field, the particles density number was corrected
implicitly in the second step. After these two steps, the particle density number
returned to the original value, and the next cycle continued until the completion
of calculation. The water is considered incompressible in the process of the being
expelled from the nozzle to reach the ground, which means that the particle den-
sity number is a constant. Each calculation step is divided into two steps based on
this constant and the calculation flow is shown in Fig. 3. Firstly, for each particle,
u0 is velocity, r0 is position and n0 is particle density number under the initial con-
ditions. After that, the new velocity u and position r are calculated under the
action of volume force and boundary conditions. Furthermore, the new particle
density number n is generated because of changes in position and velocity in the
meanwhile. Secondly, the Pressure-Poisson equation is used to correct the particle
density to the original value n0 based on the initial conditions. At this point, one
single step of the entire calculation process based on the MPS method has been
completed.

2.2. Condition of Simulation

Jet trajectories in large space are extremely dependent on the initial velocity of the
fire monitor as well as the initial angle. In order to analyze the relationship
between trajectory characteristics and pressure, and compared that with experi-
mental data items to analyze the feasibility of the trajectory model, the simulation
 Fire Technology 2018

Figure 3. Simulation flow of the MPS method (u0, r0, n0 represent the
initial velocity, position and particle density number, respectively. u*,
r*,n*, represent the velocity, position and particle density number
after a single display calculation, respectively. u , r represent the
velocity, position after an implicit calculation, respectively).

condition parameters were as consistent as possible with the experimental environ-

ment parameters. The large-space water jet simulation model was conducted based
on the MPS method under the following conditions corresponding to the experi-
mental parameters: set flow as 16.60, 43.00, 67.95 and 95:77 m3 /h (initial velocity
being 5.40, 14.00, 22.08 and 31:12 m/s), set angle as 10 ; 20 ; 30 and 40 . In this
case, the influence of wind was not considered for the breezy environment and the
high randomness of wind velocity and direction.
Study on Water Jet Trajectory Model

Figure 4. Structural diagram of the experimental platform.

3. Experiment Study
3.1. Experimental Platform
It is generally arduous to enforce fire extinguishing experiments due to relevant
safety regulations. It must be stated that the experimental site of this study was in
Xu-gong Construction Machinery Group (XCMG) Comprehensive Test Site and
the safety and security guidance were provided by XCMG Research Institute. The
experimental platform consists of water-tank, pump, frequency converter, fire
monitor, solenoid valve, flow meter, pressure and other fittings, which were instal-
led in XCMG Comprehensive Test Site, as shown in Fig. 4. The water tank can
provide 10 tons of water in one experiment and the water was pressurized into the
pipeline through the pump controlled by a frequency converter. The exit diameter
of the experimental fire monitor was 33 mm and the water flowing into the moni-
tor can be controlled by the pump. In other words, the velocity of water shot out
of the fire monitor was determined by the inverter frequency. The main purpose
of this experiment is to explore the effects of the water exit velocity and the spray
angle of the fire monitor on the jet trajectory.
The jet trajectory image can be obtained by a high-speed camera acquisition
system which could take 7400 pictures per second with the maximum resolution of
1280  800. The maximum jet height and range of the water trajectory can be
analyzed with the image measurement software Mind Vision.
 Fire Technology 2018

3.2. Experimental and Trajectory Image Processing

The experiment was carried out under breezy and weak sunlight environment so
as to minimize the impact during jet image capture. In this study, the experimen-
tal and simulation parameters were as identical as possible to facilitate the com-
parison and analysis of results. The camera was placed on the side of the jet
trajectory and the camera perspective was perpendicular to the trajectory plane to
obtain proper images. In order to obtain stable trajectory parameters, each image
was taken after the jet trajectory was stabilized. Moreover, in order to overcome
the interference of lens distortion, the lens distortion was corrected before image
processing. As an example, the velocity and the angle of water shot out of the fire
monitor were 14 m/s and 20 respectively, and the jet trajectory image and mea-
surement results are shown in Fig. 5.
The basic principle of jet trajectory range measurement is scale conversion. The
starting point of the jet trajectory was 0.8 m above the ground after establishment
of the experiment system. The height of the jet trajectory start-point from the
ground and the range of jet trajectory in the image can be easily obtained by
Mind-Vision software. The jet trajectory is identified in the software by means of
manual calibration. As shown in Fig. 5, there are detailed scales around the jet
trajectory image in the software, and the jet trajectory range in the image can be
obtained easily and accurately by manual calibration. Therefore, the real range of
jet trajectory can be obtained by scale conversion.
With the help of the image measurement software Mind-Vision, the pixel length
of any object in the image can be easily obtained and displayed. Specifically, the
height of the green matrix box represents the height of the water jet’s starting
point from the ground, and the height and length of the red rectangle represent
the height and range of the jet trajectory respectively. The height of the water jet
starting point from the ground can be measured before each experiment and the
height is 0.8 m in this research. Therefore, the actual height and range of the jet
trajectory can be obtained easily with the help of the proportional conversion
principle.
The scale conversion is expressed as:

H
R¼ r ð5Þ
h

where H is the height of the jet start-point from the ground, h and r are the start-
ing-point height and drop point range of jet trajectory in the image, respectively.
In Fig. 5, the height and width of the water jet trajectory are 1.8 m and 15.3 m
respectively.

4. Results and Discussions

In order to verify the accuracy of trajectory simulation results based on the MPS
method, the experimental and simulation trajectory were compared and analyzed,
and the results are shown in Table 1. As a case of comparison, Fig. 6 shows the
Study on Water Jet Trajectory Model

Figure 5. Water jet trajectory image and measurement results based

on the Mind-Vision.

experiment and simulation trajectories in the same coordinate system with the ini-
tial velocity and jet angle being 14.0 m/s and 20 respectively. It can be seen that
the jet trajectory between experiment and simulation are very similar in shape.
The range of the simulation and experiment trajectories are 14.3 m and 15.3 m
respectively, and both the height are 1.8 m.
As can be seen from Table 1, the simulated jet trajectory based on MPS
method matched the trajectory image obtained through experiments consistently.
Under initial speeds of 5.4 m/s and 14.0 m/s, the difference between the trajectory
of the experiment and the simulation is very small. In most cases, the result is
exactly the same, and the maximum height error is no more than 0.5 m. The
range values of the trajectories also exhibit such features, and the maximum error
is less than 1.2 m and the maximum relative error is 6.5%. Under initial speed are
22.1 m/s and 31.1 m/s, the trajectory height error between experimental and simu-
lation data increases obviously, and the trajectory range error also increases. The
relative height error value exceeds 12.3% generally and the maximum range rela-
tive error reaches 22.7%. As the water jet angle increases, the errors of height and
the range between experimental and simulation data also increase. Obviously, the
results cannot provide reliable data support for the water to reach the fire position
accurately.
Previous research found that the total time of the water jet fly in the air-space
and the functional area of the air resistance on the water trajectory were the two
main factors which caused the error [21]. The water jet at a high speed could fly
in the air-space for longer time result in more significant impact of air resistance
on the surface of the trajectory, and it is worth noting that the initial angle also
affects the flight time of the water jet. Furthermore, after the trajectory fly length
exceeds the water jet broken length, the effect of air resistance to the trajectory
also significantly increases on account that the surface area of the air resistance on
the trajectory increases obviously.
The analysis shows that the random direction and size of water jet diffusion
result in the difference between simulation and experimental data under actual
environment. Therefore, the water jet error between experimental and simulation
must be corrected based on the MPS simulation model. However, the water jet
diffusion has always been a difficult problem in academic research and not the
 Fire Technology 2018

Table 1
Comparison of Data for Experimental and Simulated Water Jet
Trajectories (Exp and Sim are the Experimental and Simulation Value
Respectively, Dev is the Relative Deviation)

Height (m) Range (m)


Frequency (Hz) Initial velocity (m/s) Angle ( ) Exp Sim Dev Exp Sim Dev

10 5.4 10 0.8 0.8 0.0 2.7 2.6 3.7

20 0.9 0.9 0.0 3.3 3.2 3.0
30 1.0 1.0 0.0 3.6 3.5 2.8
40 1.4 1.4 0.0 3.7 3.6 2.7
20 14.0 10 1.0 1.0 0.0 9.5 9.8 3.2
20 1.8 1.8 0.0 15.3 14.3 6.5
30 3.3 3.4 3.0 16.5 17.5 6.1
40 3.4 3.2 5.9 19.5 18.3 6.2
30 22.1 10 1.5 1.4 6.7 23.2 20.9 9.9
20 3.6 3.5 2.8 27.5 31.8 15.6
30 5.5 6.1 10.9 33.5 38.4 14.6
40 5.7 6.4 12.3 36.5 41.0 12.9
40 31.1 10 2.0 2.1 5.0 30.2 34.8 15.2
20 5.5 5.8 5.5 42.5 49.3 16.0
30 10.5 11.1 5.7 47.5 58.3 22.7
40 14.5 16.0 10.3 55.5 63.3 14.1

focus of this article. Furthermore, modifying the simulation model based on the
experimental results is an easier way to establish the jet trajectory model, and it
also reduces the difficulty of the model and the calculation load. Previous studies
find that the initial velocity and the angle of the water jet are the main factors
directly causing the error, this two water jet parameters are selected to modify the
simulation model, make the simulation results more accurate and related to the
experimental results more closely.

5. Error Compensation
5.1. Interpolation Compensation Principle of Binary Function
To improve the fire extinguishing efficiency, the jet trajectory range and height
error caused by the longer flight time and jet diffusion need to be reduced. The
analysis shows that the water jet errors mainly depend on the initial velocity and
angle. Therefore, a binary function piecewise interpolation method was used for
error compensation based on the two parameters of initial velocity and initial
angle as error factors.
The method mentioned above is an extension of piecewise interpolation com-
pensation for unary functions [22, 23]. As an example to explain the method,
divide the initial velocity v into n parts, the position of a point i is marked as vi
and the error @ 1 ðvi Þ ¼ f 1 ðvi Þ in here. L1 ðxÞ represents the interpolation function
Study on Water Jet Trajectory Model

Figure 6. Comparison of experimental and simulated water jet

trajectories (Pictures a and b are experimental and simulation
trajectory images respectively, with the initial velocity and jet angle
are 14.0 m/s and 20° respectively).

within interval ½vi ; viþ1 , and L1 ðvi Þ ¼ f 1 ðvi Þ; L1 ðviþ1 Þ ¼ f 1 ðviþ1 Þ the interpolation is
expressed as:
v  vi
L1 ðvÞ ¼ f 1 ðvi Þ þ ½f ðviþ1 Þ  f 1 ðvi Þ ð6Þ
viþ1  vi 1

where i ¼ 1; 2; . . . ; n, vi  v  viþ1 .
Divide the initial velocity v into n parts and divide the initial angle h into m
parts. As a result, the plane composed of variables v and h can be divided into
n  m block areas. The error of the area numbered (i, j) is @ ¼ f ðv; hÞ and H 1 ðv; hÞ
is the interpolation function. When v is a constant, it can be obtained from the
piecewise interpolation compensation for unary functions that:
v  vi
L1 ðv; hÞ ¼ f ðvi ; hj Þ þ ½f ðviþ1 ; hj Þ  f 1 ðvi ; hj Þ ð7Þ
viþ1  vi 1

v  vi
L1 ðv; hjþ1 Þ ¼ f ðvi ; hjþ1 Þ þ ½f ðviþ1 ; hjþ1 Þ  f 1 ðvi ; hjþ1 Þ ð8Þ
viþ1  vi 1

After interpolation calculation of h, the binary function error compensation value

can be obtained, as shown in Eq. (4).
v  vj
H 1 ðv; hÞ ¼ f ðv; hj Þ þ ½L1 ðv; hjþ1 Þ  L1 ðv; hj Þ ð9Þ
vjþ1  vj

5.2. Results Analysis

With the aid of the above error compensation principle, binary function error
compensation model of the water jet trajectory height and range was established
 Fire Technology 2018

Table 2
Partial Error Compensation Results and the Deviation After
Compensation (Exp and Sim are the Experimental and Simulation
Value Respectively, Com is the Error Compensation Value and the Dev
is the Deviation After Error Compensation)

Height (m) Range (m)

Angel ( ) Initial velocity (m/s) Exp Sim Com Dev Exp Sim Com Dev

20 13.5 1.6 1.7 0.12 0.02 9.6 9.6 0.03 0.03

18.3 2.5 2.6 0.11 0.01 21.6 23.2 0.23 1.37
26.4 4.1 4.3 0.55 0.35 38.2 40.2 3.50 1.50
30 13.5 2.1 2.2 0.09 0.01 15.6 16.3 0.85 0.15
18.3 5.5 5.7 0.18 0.02 26.5 30.3 2.65 1.15
26.4 7.9 9.3 1.63 0.23 36.5 44.5 6.35 1.65
40 13.5 2.7 2.8 0.15 0.05 16.3 17.3 0.96 0.04
18.3 5.0 5.4 0.55 0.15 30.2 36.3 7.55 1.45
26.4 9.5 13.0 3.20 0.30 38.5 44.2 7.96 2.26

based on the experimental and simulation data. In order to verify the error com-
pensation model, unused experimental and simulation data were substituted into
the model for comparison. Firstly, after the initial velocity and the angle of the
water jet were determined, the jet trajectory was simulated based on the MPS
method and the trajectory height and range were measured by means of the Mind
Vision software. Secondly, experimental trajectory are also measured and calcu-
lated with the same parameters of the simulation. Furthermore, error compensa-
tion values of the height and range were calculated by substituting the initial
velocity and angle values into the error compensation model respectively. Finally,
add the height error compensation value to the simulation trajectory height value,
and the same calculation for the trajectory range. Partial error compensation
results and the deviation after compensation are shown in Table 2.
Table 2 shows the result of revised jet simulation data, the error of simulation
data and experimental data can be well corrected based on binary function inter-
polation compensation model. It can be seen that this method reduces the error
between simulation and experimental data with the lower initial velocity effec-
tively. With the help of error compensation model, the height error of the jet tra-
jectory between experiment and simulation does not exceed 0.35 m even under
high initial velocity, and the range error of the trajectory is at most 2.26 m. How-
ever, as the initial velocity increases, the effect of error compensation model
decreases, which may be caused by too little experimental data. The reasons for
errors may be that there is too little data for the error compensation model estab-
lishment. However, due to limitations of the experimental site and the environ-
mental conditions, it is difficult to obtain a large amount of experimental data.
Similarly, in the course of the experiment, the high randomness of the wind may
also be one of the causes for such results. Fortunately, the modified simulation
Study on Water Jet Trajectory Model

trajectory height and range are closer to experimental results, the errors have little
impact on actual demand of the firefighting preferably.

6. Conclusion
Jet trajectory model based on the MPS method predictions shows satisfactory
agreement with experimental data of the fire monitor water jet trajectory. The
model can be used to predict the trajectory of fire monitor in the fire-ground,
which could make the fire suppression more accurate and efficient. However, the
range and height difference between simulation and experimental data becomes
increasingly obvious as the initial velocity increases under the influence of air
resistance. After the water jet break-up, owing to the longer flight time and
greater force area, the air resistance has a greater effect on the trajectory. How-
ever, the influence of air resistance on the water jet trajectory is inevitable since
the wind is everywhere. Therefore, the trajectory simulation model needs to be
modified to eliminate the negative effects of these factors, such as the broken
length and the wind speed and direction during the flight time.
A binary function interpolation compensation method was developed for simu-
lation error compensation. This algorithm was developed for use on an automatic
fire suppression system that helps to correct the jet trajectory model so that the
fire extinguishing agent can accurately reach the ignition point after the fire was
discovered and located. After adding the error correction model, the relative error
of the jet trajectory height decreased from 1.5 m to 0.3 m and the range relative
error also decreased from 10.8 m to 2.26 m. It greatly enhanced the prediction
accuracy of the height and range of the jet trajectory. However, there was also
some errors in the revised model, especially when the initial speed was higher.
This may be caused by deficient error compensation for experimental data and
random fluctuations in size and direction of the wind. It needs to be pointed out
that the error has little effect on the efficiency of fire extinguishing because the fire
extinguishing agent will spread out into an elliptical area with a large area at the
terminal of the jet trajectory.
To summarize, the jet trajectory model established based on MPS method can
provide effective data support for long-distance automatic fire extinguishing after
being corrected via the binary function interpolation method. In the actual fire
extinguishing process, the existential error does not affect its wide application in
urban fire vehicles, fixed fire extinguishing systems, and fire extinguishing forcast-
ing systems. With more experimental and application data accumulation,results of
the firefighting operation supported by the model would be better.

Acknowledgements
This work is supported by National Key R&D Program of China
(2016YFC0802900), Xu-gong Construction Machinery Group (XCMG) Research
Institute and a Project Funded by the Priority Academic Program Development
 Fire Technology 2018

of Jiangsu Higher Education Institutions, Top-notch Academic Programs Project

of Jiangsu Higher Education Institutions.

References
1. McNeil JG, Lattimer BY (2017) Robotic fire suppression through autonomous feed-
back control. Fire Technol 53(3):1171–1199
2. Chen T, Yuan H, Su G, Fan W (2004) An automatic fire searching and suppression
system for large spaces. Fire Saf J 39(4):297–307. https://doi.org/10.1016/j.fire-
saf.2003.11.007
3. Sallam K, Dai Z, Faeth G (2002) Liquid breakup at the surface of turbulent round liq-
uid jets in still gases. Int J Multiph Flow 28(3):427–449. https://doi.org/10.1016/S0301-
9322(01)00067-2
4. Hatton A, Leech C, Osborne M (1985) Computer simulation of the trajectories of large
water jets. Int J Heat Fluid Flow 6(2):137–141. https://doi.org/10.1016/0142-
727X(85)90051-7
5. Min Y (2011) Pitching angle-based theoretical model for the track simulation of water
jet out from water fire monitors. J Mech Eng 47(11):134(in Chinese)
6. Wu F, Cui Y, Qu F, Mai L (2016) Experimental study on fire extinguishing characteris-
tics of automatic sprinkler system. In: Sixth international conference on intelligent sys-
tems design and engineering applications, pp 389–392
7. Zhang W, Zhu DZ (2015) Far-field properties of aerated water jets in air. Int J Multiph
Flow 76:158–167. https://doi.org/10.1016/j.ijmultiphaseflow.2015.07.006
8. Kim JH, Lattimer BY (2015) Real-time probabilistic classification of fire and smoke
using thermal imagery for intelligent firefighting robot. Fire Saf J 72:40–49. https://
doi.org/10.1016/j.firesaf.2015.02.007
9. McNeil JG Lattimer BY (2015) Real-time classification of water spray and leaks for
robotic firefighting. IGI Global
10. McNeil JG, Lattimer BY (2016) Autonomous fire suppression system for use in high
and low visibility environments by visual servoing. Fire Technol 52(5):1343–1368
11. Miyashita T, Sugawa O, Wada Y, Ishikawa R, Kawaguchi Y (2012) Three-dimensional
simulation model for water and fire-foam discharge using mps method. J Jpn Soc Saf
Eng 51 (in Japanese)
12. Miyashita T, Sugawa O, Wada Y, Ishikawa R, Kawaguchi Y (2013) Development of
two-dimensional simple simulation model and evaluation of discharge ability for water
discharge of firefighting. Bull Jpn Assoc Fire Sci Eng 62:13–19(in Japanese)
13. Miyashita T, Sugawa O, Imamura T, Kamiya K, Kawaguchi Y (2014) Modeling and
analysis of water discharge trajectory with large capacity monitor. Fire Saf J 63:1–8.
https://doi.org/10.1016/j.firesaf.2013.09.028
14. Hong-yong Y (2002) Fire location and suppression with automatic hydrant in large
space. In: Proceedings of 2nd NRIFD symposium—science, technology and standards
for fire suppression systems, vol 6, pp 17–19
15. Nam S, Braga A, Kung H, Troup J (2003) Fire protection for non-storage occupancies
with high ceiling clearances. Fire Saf Sci 7:493–504
16. Xin Y, Thumuluru S, Jiang F, Yin R, Yao B, Zhang K, Liu B (2014) An experimental
study of automatic water cannon systems for fire protection of large open spaces. Fire
Technol 50(2):233–248
Study on Water Jet Trajectory Model

17. Koshizuka S, Oka Y (1996) Moving-particle semi-implicit method for fragmentation of

incompressible fluid. Nucl Sci Eng 123(3):421–434
18. Koshizuka S (1995) A particle method for incompressible viscous flow with fluid frag-
mentation. J Comput Fluid Dyn 4:29–46
19. Liu J, Koshizuka S, Oka Y (2005) A hybrid particle-mesh method for viscous, incom-
pressible, multiphase flows. J Comput Phys 202(1):65–93. https://doi.org/10.1016/
j.jcp.2004.07.002
20. Shakibaeinia A, Jin YC (2012) Mps mesh-free particle method for multiphase flows.
Comput Meth Appl Mech Eng 229–232:13–26. https://doi.org/10.1016/
j.cma.2012.03.013
21. Funada T, Joseph D, Yamashita S (2004) Stability of a liquid jet into incompressible
gases and liquids. Int J Multiph Flow 30(11):1279–1310. https://doi.org/10.1016/j.ijmul-
tiphaseflow.2004.07.001
22. Okafor A, Ertekin YM (2000) Derivation of machine tool error models and error com-
pensation procedure for three axes vertical machining center using rigid body kinemat-
ics. Int J Mach Tools Manuf 40(8):1199–1213. https://doi.org/10.1016/S0890-
6955(99)00105-4
23. Wang S, Yun J, Zhang Z, Liu Y, Zhang Q (2003) Tranjin university modeling and
compensation technique for the geometric errors of five-axis cnc machine tools. Chin J
Mech Eng 16(2):197–201

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.

View publication stats