WWW.IJITECH.ORG
ISSN 2321-8665
Vol.03,Issue.05,
July-2015,
Pages:0656-0661
Channel and Energy Modeling for Self-Contained Wireless Sensor
Networks in Oil Reservoirs
DASARI NARESH NAYUDU1, ESARAM PARVATHI2, D. SUBBARAO3
1
PG Scholar, India, E-mail: [email protected].
Assistant Professor, India, E-mail: [email protected] & 3HOD, India.
2
Abstract: Wireless communication is the prerequisite for the
highly desired in-situ and real-time monitoring capability in
underground environments, including oil reservoirs, groundwater aquifers, volcano’s, among others. However, existing
wireless communication techniques do not work in such
environments due to the harsh transmission medium with
very high material absorption and the inaccessible nature of
underground environment that requires extremely small
device size. Although Magnetic Induction (MI) communication has been shown to be a promising technique in
underground environments, the existing MI system utilizes
very large coil antennas, which are not suitable for
deployment in underground. In this paper, we propose a
metamaterial enhanced magnetic induction communication
mechanism that can achieve over meter scale communication
range by using millimeter scale coil antennas in the harsh
underground environment. An analytical channel model for
the new mechanism is developed to explore the fundamentals
of metamaterial enhanced MI communication in various
underground environments. The effects of important system
and environmental factors are quantitatively captured,
including the operating frequency, bandwidth, and parameters
of metamaterial antennas, as well as permittivity,
permeability, and conductivity of underground medium. The
theoretical model is validated through the finite element
simulation software, COMSOL Multiphysics.
Keywords: Channel Modeling, Oil Reservoirs, Magnetic
Induction (MI) Communications, Wireless Sensor Networks,
Self Contained, Wireless Energy Transfer.
I. INTRODUCTION
Currently the crude oil and natural gas provide more than
50% share of the total global energy consumption. However,
the oil reserve is diminishing faster and faster due to the
increasing energy demands all over the world. Meanwhile,
the current achievable oil recovery rate, i.e. the ratio of
recoverable oil to the total oil in the reservoirs, is less than
60% even using the latest extraction technology which
aggravates the global energy crisis. Hence, to guarantee the
global energy security, it is of great importance to increase
the current oil recovery factor. Increasing recovery factor
requires optimally developing the oil/gas fields and optimally
managing the recovery process, which necessitates the realtime and high resolution physical chemical information
throughout the whole oil reservoirs, such as pressure,
temperature, and fluid type. Better understanding of such
information in oil reservoirs can significantly increase the
recovery factor. For instance, during the recovery process,
water is injected into the oil reservoir to maintain the
pressure. On the one hand, if the injected pressure is too low,
the oil cannot be effectively extracted. On the other hand, if
the pressure is too high, the reservoir can be destroyed.
Hence, by deriving the real-time pressure information inside
oil reservoirs, the water injection can be optimally managed
and the oil recovery factor is envisioned to be increased by
15%-25% However, to date, no existing technology can
provide the real-time and in-situ monitoring capability inside
oil reservoirs.
The accuracy of remote monitoring are far from
satisfactory, while the intra-wellbore monitoring systems
cannot derive the measurements deep inside the oil reservoirs
.A natural idea to collect the accurate real-time information is
to utilize the wireless sensor network (WSN) paradigm ,
where wireless sensors are deployed throughout the whole oil
reservoir. Such kind of in-situ and direct measuring
techniques have drawn increasing attention from the industry
and research community. Currently, most of the efforts have
been put towards the development of the sensors that are
small and strong enough for the deep oil reservoir monitoring
This kind of micro wireless sensor motes have been
successfully injected into oil reservoirs However, it is still
unknown how to continuously report the measurements from
the micro wireless sensors deep inside oil reservoirs in real
time. Existing wireless communication and networking
solutions do not support WSNs in the oil reservoirs.
Specifically, shows the typical oil reservoir environment and
the oil extraction process, where a wellbore is drilled to the
underground oil reservoirs at the depth of around 6,000 ft the
hydraulic fracturing process using high pressure fluid to open
several long but very narrow fractures growing from the
wellbore to the targeted rock formations, so that the oil and
natural gas can be extracted. Main objective of this paper is to
design an wireless sensor node for real time monitoring of oil
reservoirs.
Copyright @ 2015 IJIT. All rights reserved.
DASARI NARESH NAYUDU, ESARAM PARVATHI, D SUBBARAO
The wireless Sensor networks are envisioned to be
deployed deep inside oil reservoirs to collect and report the
physical and chemical information in real time Zigbee Based
Smart Sensing Platform for Monitoring Parameters in oil
reservoirs has been designed and developed. This system
guarantee the global energy security, it is of great importance
to increase in the current oil recovery factor. Through this
path loss in oil reservoirs can be minimized Wireless
communication in challenging environments, especially in
underwater, have been extensively explored. However, the
results cannot be applied to the oil reservoir environment due
to the following reasons. Firstly, the highly confined space
can significantly affect the Field distribution, which leads to
very complicated channel characteristics in oil reservoir.
Fig.1.The structure of oil reservoir and the system
Moreover, most underwater communication system utilizes
architecture of the MI-based wireless sensor network.
acoustic waves, which is different from the magnetic
induction. Secondly, we can artificially modify the
The remainder of this paper is organized as follows. Related
electromagnetic properties of the signal propagation medium
Work in Section II. Next, in Section III, Metamaterialin the oil reservoirs. Thirdly, the power issue is much more
Enhanced Mutual Induction .After that, in Section IV, the
critical for the sensors in oil reservoir. For the sensors in oil
energy consumption and transfer model of the whole sensor
reservoir, once they are injected into the fracture, there is no
network system is developed based on the derived channel
way to replace them. Thus we need to design self contained
models. Finally, this paper is concluded in Section V.
sensor network.
However, it is still unknown how to continuously report
the measurements from the micro wireless sensors deep
inside oil reservoirs in real time. Existing wireless
communication and networking solutions do not support
WSNs in the oil reservoirs. Specifically, Fig. 1 shows the
typical oil reservoir environment and the oil extraction
process, where a wellbore is drilled to the underground oil
reservoirs at the depth of around 6,000 ft. The hydraulic
fracturing process using high pressure fluid to open several
long but very narrow fractures growing from the wellbore to
the targeted rock formations, so that the oil and natural gas
can be extracted the length of the fracture can reach up to 100
m; while the typical width and height of the fracture are 0.01
m and 1 m, respectively. Since the WSN is envisioned to be
deployed in the fractures during the hydraulic fracturing
process, the following research challenges are imposed to the
existing wireless communication and networking solutions:
 Extremely high path loss in oil reservoirs: Since the
oil reservoirs consist of crude oil, gas, soil and rocks, and
other injected fluids, electromagnetic (EM) waves
experience
significant
material
absorptions.
Consequently, the path loss in the oil reservoir is
extremely high.
 Infeasibility in radiating wireless signals: Due to the
very narrow fracture, the size of wireless sensors should
be no larger than 1 cm. Consequently, the antenna on the
sensor node can only radiate EM waves with a frequency
of hundreds of GHz or even several THz. The
propagation of such high frequency signals in the oil
reservoirs is extremely difficult [12].
 Prohibitively short system lifetime: Similarly, due to
the size limitation, the capacity of the battery on the
wireless sensor is extremely small. The small battery
cannot power the device for a long enough operating
period, which causes a prohibitively short system
lifetime of the wireless sensor networks in oil reservoirs.
II. RELATED WORK
The concept of the WUSNs is first proposed in, where the
applications, system architecture, and research challenge is
discussed. The channel models for EM wave-based wireless
communication in soil medium are developed. To address the
problem of extremely high path loss of EM waves in pure
underground environments, the hybrid WUSNs consisting of
both aboveground data sink and shallowly buried
underground sensor nodes are proposed. To reduce the high
path loss due to the medium absorption, the magnetic
induction (MI) technique has been widely used in many
challenging environments, such as body area, and
underwater. The theoretical model of path loss and channel
capacity for MI communication has been reported, which
show the significant improvement over EM wave-based
solution if only underground channel is available. However,
MI communication system is highly directional and has
limited channel capacity which limits its application. To
address the challenges, a tri-directional coil antenna is
proposed to increase the channel capacity and create an
omnidirectional receiving antenna. In order to extend
communication range, the MI waveguide presented is
adopted.
The theoretical system model for MI waveguide in
underground environments is provided and the numerical
simulation shows that the communication range can reach
tens of meters when the diameter of the coil antenna is 30 cm.
Another way to extend the MI communication range is to use
very large coil antenna. In, a MI-based through-the-earth
communication system is developed for mine disaster rescue,
where the wireless device is bigger than a human. However,
all the above MI techniques cannot realize the underground
wireless communications among millimeter scale sensors,
which are essential to the envisioned applications in this
paper. Metamaterials proposed it can manipulate the EM
fields by adjusting the value of the dielectric permittivity and
International Journal of Innovative Technologies
Volume.03, Issue No.05, July-2015, Pages: 0656-0661
Channel and Energy Modeling for Self-Contained Wireless Sensor Networks in Oil Reservoirs
magnetic permeability of the material to nearly any value,
as the second layer, and the space outside the sphere as the
including positive, negative, or even near-zero values. It has
third layer. The MI coil is located at the center of first layer.
been shown that metamaterials can greatly improve the
In this section, we fist derived the field distribution in each
efficiency of magnetic induction-based wireless energy
layer by using spherical wave equations. Then the mutual
transfer the experiment in shows that the efficiency of
induction and self induction can be calculated.
wireless power transmission can be as high as 47% by using
IV. ENERGY MODELING AND SELF-CONTAINED
metamaterial. While without using metamaterial the
SYSTEM ANALYSIS
efficiency is only 17%. The corresponding theoretical model
Based on the derived channel models, in this section, we
is developed in where the loss of metamaterials is taken into
show that the sensors deep inside oil reservoirs can utilize the
consideration. In, by using a small metamaterial slab with −1
energy transferred from the large dipole to perform sensing
permeability at 63.87 MHz, the sensitivity of magnetic
tasks and wirelessly transmit the measurements back to the
resonance imaging (MRI) is significantly increased.
base station. Specifically, the energy transfer model and
However, all above works consider the strong coupling
energy consumption model for the proposed oil reservoir
mode, where two coils are very close to each other. In
sensor network system are first developed. In order to
underground wireless communications, to enlarge the
calculate the transmitting/receiving time that is required by
coverage range, we consider the loose coupling to achieve
the energy consumption model, the uplink channel capacity is
longer transmission distance. Moreover, due to the different
derived to obtain the achievable data rate. Then, based on the
objectives, the system bandwidth for energy transfer and MRI
models, the self-contained property of the proposed wireless
is too small for wireless communications. Besides wireless
sensor network is validated by balancing the transferred
energy transfer and MRI, metamaterials have also been used
energy and the consumed energy.
in antenna design for the EM wave-based terrestrial
A. Energy Transfer Model
communication systems. In, metamaterials are introduced to
The downlink channel that is discussed in the last section is
miniaturize the antenna in portable devices. In, the radiations
utilized
here to realize the energy transfer. If the transmission
of metamaterial-enhanced electrical dipole antenna are
power
is
PdipoleTX, the received power at the MI sensor node
theoretically investigated. However, all the above antennas
coil
is P RX = PdipoleTX・ Ld2cp (dd2c), where Ld2cp (dd2c) is the path
operate at UHF band that are not suitable in underground.
Such antennas lose their efficiency if HF band signals are
loss of the downlink channel given and the transmission
used. As discussed in Section I, MI communication is a
distance is dd2c. Assuming the energy conversion rate at the
promising candidate solution in underground environments if
sensor node is η, the available power at the micro sensor node
is:
the near field can be enhanced by the metamaterials. To our
best knowledge, this is the first paper which analyzes the
(1)
channel characteristics of metamaterial enhanced MI
Then
the
energy
transferred
to
the
sensor
node
(denoted
as
communications in underground environments.
node n) is Ern = Psensoravail (dd2c) Tc, where Tc is the charging
time. An ultra capacitor with capacity Emax is utilized by each
III. METAMATERIAL-ENHANCED MUTUAL
sensor to store the energy. The sensors far away from the
INDUCTION
dipole antenna require more time to receive enough energy
In order to capture the channel characteristics of
than those close to the dipole. However, since the sensor
metamaterial enhanced MI communication, in this section,
nodes that are far from the dipole do not need to relay as
we first analyze the EM field (especially the magnetic field)
many packets as the sensor nodes near the dipole, they also
intensity distribution around the transmitting antenna and the
requires less energy.
receiving antenna. We validate the theoretical result by
matching it with the FDTD simulation results obtained by
COMSOL Multiphysics. Then self induction of each
metamaterial enhanced coil antenna and the mutual induction
between the transceivers can be obtained, which are the key
parameters in MI communications. Generally, there are three
kinds of structures utilizing metamaterials in antenna design:
the rectangular structure utilizes a metamaterial slab in
between the transmitter and receiver the cylindrical structure
utilizes a metamaterial cylinder to surround the transceivers,
and the spherical structure uses a metamaterial sphere to
enclose the antenna. For underground MI communications in
this paper, we adopt the spherical structure to create a
unidirectional Coverage. The spherical structure is also
favorable for deployment. Hence, the metamaterial enhanced
MI coil antenna considered in this paper is illustrated in Fig.
2: the single-turn MI coil with radius of a is enclosed by a
metamaterial sphere with and internal radius of r1 and
external radius of r2. We define the space inside the
Fig.2. Comparison between unidirectional coil and trimetamaterial sphere as the first layer, the metamaterial layer
directional coil in downlink channel.
International Journal of Innovative Technologies
Volume.03, Issue No.05, July-2015, Pages: 0656-0661
DASARI NARESH NAYUDU, ESARAM PARVATHI, D SUBBARAO
oil reservoirs. The 3-dB bandwidth in the oil reservoir sensor
network can be obtained by:
(3)
Where L and C are the coil inductance and the resonance
capacitor. According to,
, while C =
,
where f0 is the resonant frequency. By solving (3), we can
obtain two frequencies: fU that is the upper cutoff frequency;
and fL that is the lower cutoff frequency. Then the bandwidth
of the uplink channel is:
Fig.3. Comparison between unidirectional coil and tridirectional coil in uplink channel.
B. Energy Consumption Model
We select a widely used energy consumption model given
in [10], where the total energy consumption in a sensor node
consists of three parts: the communication consumption, data
processing consumption, and sensing consumption. Since the
energy used for data processing and sensing constitutes a very
small portion, only the communication energy consumption is
considered in this paper. The duty cycle of the proposed
sensor network is denoted as T0. During each cycle, all the
sensor nodes report their measurements to the base station in
turns. It costs Et for sensor nodes to transmit the information
and Er for the relay nodes to receive the information.
According to, Et = Ptx (Ton−tx+Tst) +Pout Ton−tx and Er = Prx
(Ton−rx+Tst). Ptx/rx is the power consumption of the
transmitter/receiver; Pout is the output transmit power; Ton−tx/rx
is the transmit/receive time; and Tst is the startup time of the
transceiver. We assume the proposed sensor network has a
linear topology along the long and narrow oil reservoir
fracture. There are N sensor nodes in the network. We use
sequence number n to denote each node. The sequence
number of the sensor node who is the closest to the dipole
antenna is 1. The next one is 2, so on and so forth. Then, the
energy consumption in T0 for each sensor node can be
calculated by
(2)
Now the only unknown parameter in the energy model is
Ton−tx/rx, i.e. the transmit/receive time, which is determined by
1). How much data needs to be transmitted/received and 2).
how fast the data is transmitted/received. We assume the
length of the measurement data that needs to be reported is a
determined constant. Then, to complete the energy
consumption model of the oil reservoir sensor networks, we
only need to find out the data rate to report this data.
Uplink Channel Capacity: The upper bound of the data rate
to report the sensing data is determined by the channel
capacity of the uplink channel. It is shown that the MI energy
transfer requires a single frequency to maximize the
efficiency, while the MI communication needs a certain
bandwidth. Thus, there should be an maximum value for the
channel capacity. We have investigated the bandwidth and
channel capacity of the MI-based wireless sensor networks
with relay coils. In this paper, we further develop the result to
cover the unique MI channel in the fracture environments in
(4)
Fig.4. Optimal μ for different mutual inductions.
Fig.5. Received energy & consumed energy.
According to, the channel capacity can be calculated by:
(5)
where Pt is the transmission power and Nnoise is the noise
power. Since the coils are loosely coupled to each other, 4π2 f
2 2
M is much smaller than Rc. Hence, 4π2 f 2M2/Rc is very
small and can be neglected. Then, the closed form of (5) can
be approximately obtained:
International Journal of Innovative Technologies
Volume.03, Issue No.05, July-2015, Pages: 0656-0661
(6)
Channel and Energy Modeling for Self-Contained Wireless Sensor Networks in Oil Reservoirs
Optimal Μ1 To Maximize Uplink Channel Capacity: As
Fig. 5 shows the received energy in six hours and the
discussed previously, the effective permeability inside the
consumed energy by each sensor node. The sequence number
fracture, μ1, can be adjusted by changing the additive in the
of the sensor node also indicates the distance between the
high-μ prop pants. In we prove that higher μ 1 has no effect on
sensor node and the dipole antenna in meters. The result
the downlink channel but can significantly reduce the path
shows that even the last sensor node that is farthest from the
loss in the uplink channel. In this subsection, we analyze the
base station can receive enough energy for scheduled
effect of μ1 on the uplink channel capacity. On the one hand,
operation. It should be noted that although the last sensor
that the high-μ environment can reduce path loss in the uplink
node receives the minimum energy, it only sends out one
channel. On the other hand, (4) shows that the high-μ
packet from itself and does not relay packets. Meanwhile,
environment can also reduce the bandwidth of the channel
although the first sensor nodes need to relay the data from all
therefore; there exists an optimal value for μ in the fracture to
the other sensors, it also receives the maximum amount of
maximize the uplink channel capacity. Here we utilize the
energy that is more than enough. Interestingly, the bottleneck
numerical method to identify the optimal value. We set the
of the proposed system is neither the last nor the first sensor
transmission power as -20 dBm. The measured underground
node, but the sensors numbered 36, 37, and 38 in Fig. 5.
noise level is around -100 dBm. Due to the high temperature
These nodes receive similar energy as the last node, but they
and complicated components in the oil reservoir, we set the
need to relay the packets from the sensor nodes behind them.
noise power as -80 dBm. Since there is no measured noise
Therefore, the duty cycle should be long enough to allow
enough charging time so that those ―bottleneck‖ nodes can
level in oil reservoir at this moment, we use a higher noise.
receive enough energy to work. We find that if the last sensor
The distance between the transmitter and receiver is 1 m.
is closer to the base station, the charring time can be
Other parameters are the same as previous sections. Due to
significantly reduced. For example, if we deploy the last
the random orientation of the tri-directional coil, the mutual
sensor at the distance 35 m, the duty cycle can be reduced to
induction M is not a constant. We consider the largest,
1.5 hours.
smallest and average mutual induction. The results are shown
V. CONCLUSION
in Fig.4. It indicates that the channel capacity reaches its
The real-time and in-situ oil reservoir monitoring system
maximum value when μ is around 20μ0. It’s worth noting
requires a self-contained and micro-sized wireless sensor
that this result is a local optimized value that is subjected to
network. However, due to the micro device size and the harsh
the transmission power, noise power, coil size, number of
oil reservoir environments, existing wireless communication
turns, and transmission distance. The global optimal value
and networking solutions do not work. In this we propose a
can be found by taking all the variables into account.
Zigbee-based oil reservoirs monitoring system serves as a
However, it is out of our scope in this paper.
According to
reliable and efficient system for efficiently monitor the
Fig. 4, with the optimal μ1, the achievable data rate can be
environmental parameters Although this paper proves the
higher than 8 kbps. Then the transmit/receive time can be
proposed sensor system can be self-contained if the deepest
calculated based on the data length and the data rate, which
sensor is within 40 m inside the oil reservoir, there is still
completes the energy consumption model for oil reservoir
significant room to extend the coverage distance if the key
sensor networks.
system parameters can be optimally determined, including the
global optimal effective permeability inside the fracture, the
C. Self-Contained System Performance
system operating frequency for energy transfer and
So far, the energy transfer model and energy consumption
communication, among others. Since the sensor nodes are
model are derived. In this subsection, we utilize the two
densely deployed while their sensing results are highly
energy models to check whether the proposed oil reservoir
correlated, optimal sampling schedule throughout the whole
sensor network is self-contained or not, i.e. whether the
network can be designed to enable the sensor system provide
sensors deep inside oil reservoirs can utilize the energy
more accurate measurements, consume less energy and cover
transferred from the dipole to wirelessly transmit sensing data
deeper sections inside the oil reservoir fractures
back to the base station. The system parameters are set as
follows. As suggested, Ptx=31.5 mW, Prx=1.8 mW, and
VI. REFERENCES
Tst=800μs. Here we set Pout=-20 dBm. According to (3) and
[1] Hongzhi Guo, Student Member, IEEE, and Zhi Sun,
(4), the bandwidth and channel capacity of the proposed
Member, IEEE, ―Channel and Energy Modeling for Selfsystem are higher than 1.6 kHz and 8 Kbps, even in the worst
Contained Wireless Sensor Networks in Oil Reservoirs‖,
case. We set the length of the packet is 8 Bytes. Then
IEEE Transactions on Wireless Communications, Vol. 13,
Ton−tx/rx=8 ms. Thus, the transmitting energy is Et = 0.273 mJ
No. 4, April 2014.
and the receiving energy is Er=15.8 μJ. The capacity Emax of
[2] Z. Sun and B. Zhu, ―Channel and energy analysis on
the ultra capacitor is 0.1 J and the energy conversion rate η is
magnetic induction based wireless sensor networks in oil
90%. There are 40 sensor nodes deployed in the fracture and
reservoirs,‖ in Proc. 2013 IEEE ICC.
the interval between two neighbors is 1 m. noted that, for the
[3] D. Chapman and W. Trybula, ―Meeting the challenges of
downlink channel, we use tri-directional coil antenna as
oilfield exploration using intelligent micro and nano-scale
receiver. The path loss here we use is the maximum value
sensors,‖ in Proc. 2012 IEEE International Conference on
which is almost the minimum value for the uni-directional
Nanotechnology.
coil antenna as shown in Fig.3. We assume the charging time
[4] M. A. Braik, H. Nasr, and F. Saghir, ―New wireless
Tc is six hours. During each cycle, all the sensor nodes are
sensor network monitors water-injection wells off Abu
first charged and then report their measurements back.
Dhabi,‖ Oil & Gas Journal, vol. 106, no. 44, Nov. 2008.
International Journal of Innovative Technologies
Volume.03, Issue No.05, July-2015, Pages: 0656-0661
DASARI NARESH NAYUDU, ESARAM PARVATHI, D SUBBARAO
[5] W. C. Lyons and G. J. Plisga, Standard Handbook of
Petroleum and Natural Gas Engineering. Gulf Professional
Pub., 2005.
[6] G. J. Elbring, L. W. Lake, M. F. Wheeler, R. Banchs, and
S. P. Cooper, ―Analysis of real-time reservoir monitoring:
reservoirs, strategies and modeling,‖ Sandia National
Laboratories, Tech. Rep., 2006.
[7] D. L. Gysling and F. X. T. Bostick, ―Changing paradigms
in oil and gas reservoir monitoring—the introduction and
commercialization of in-well optical sensing systems,‖ in
Proc. 2002 Optical Fiber Sensors Conference Technical
Digest, vol. 1, pp. 43–46.
[8] A. Tamburini, M. Bianchi, C. Giannico, and F. Novali,
―Retrieving surface deformation by psinsartm technology: a
powerful tool in reservoir monitoring,‖ International J.
Greenhouse Gas Control, no. 4, pp. 928–937, 2010.
[9] O. Duru and R. N. Horne, ―Modeling reservoir
temperature transients, and matching to permanent down hole
gauge (PDG) data for reservoir parameter estimation,‖ in
Proc. 2008 SPE Annual Technical Conference and
Exhibition.
[10] I. F. Akyildiz and M. C. Vuran, Wireless Sensor
Networks. John Wiley & Sons Inc, 2010.
[11] Micro sensor motes successfully travel through a
canadian
heavy
oil
reservoir.
Available:
http://www.incas3.eu/news
[12] M. A. Akkas, I. F. Akyildiz, and R. Sokullu, ―Terahertz
channel modeling of underground sensor networks in oil
reservoirs,‖ in Proc. 2012 IEEE GLOBECOM.
International Journal of Innovative Technologies
Volume.03, Issue No.05, July-2015, Pages: 0656-0661