PipeProbe: Mapping Spatial Layout of Indoor Water
Pipelines
Yu-Chen Chang1, Tsung-Te Lai1, Hao-Hua Chu1, Polly Huang2
Department of Computer Science and Information Engineering1
Department of Electrical Engineering2
National Taiwan University, Taipei, Taiwan
{r96147, r96152, hchu}@csie.ntu.edu.tw, phuang@cc.ee.ntu.edu.tw
Abstract—We propose PipeProbe, a mobile sensor system for
mapping water pipelines hidden inside cement walls or under
floor coverings. PipeProbe works by dropping a tiny sensor capsule into the source of the water pipelines. As the PipeProbe capsule traverses the pipelines, it gathers accelerometer and waterpressure readings. Through a novel estimation method based on
Bernoulli’s principle, we reconstruct the 3D spatial layout of the
water pipelines. The PipeProbe system is non-intrusive and requires no alteration to the water pipeline infrastructure.
I.
INTRODUCTION
When indoor water pipes are hidden inside cement walls or
under floor coverings, diagnosing them without direct inspection is difficult. Especially, when the original diagram of the
pipeline layout is also missing, searching for the pipeline locations becomes guesswork and often requires brute-force methods such as knocking down cement walls or ripping up floor
coverings. This creates an opportunity for the development of a
mobile sensing probe, called PipeProbe, which can be dropped
into the source of a water pipeline. During its traversal of the
pipeline, the PipeProbe collects the sensor readings necessary
for the reconstruction of the 3D spatial layout of the traversed
water pipelines. In comparison to the traditional brute-force
approach, the PipeProbe system is a non-intrusive method of
mapping and locating indoor water pipelines that requires no
alteration to the water pipeline infrastructure.
Two recent projects that apply wireless sensor network
technologies for monitoring water pipes include the NAWMS
project [1] and the PIPENET project [2]. The NAWMS project
detects and locates pipe leaks by attaching vibration sensors to
the pipe surface. Similarly, the PIPENET project monitors water flow and detects leaks by attaching acoustic and vibration
sensors to large bulk-water pipelines and pressure sensors to
normal pipelines. In contrast to these projects, the PipeProbe
system does not assume that water pipe surfaces are exposed
and accessible for sensor module attachment. In the general
domain of environmental sensing, both wireless and wired sensor network technologies [3] have been used extensively, and a
wide variety of inexpensive sensor nodes have been created
with different sizes, sensor combinations, computational power, battery power, and radios. Our work focuses on a novel
mobile sensor system for mapping indoor water pipelines.
II.
mal-sized (>2 inch) indoor water pipelines. EcoMote comes
with a built-in 3-axial accelerometer. Additionally, the tiny
pressure sensor module MS5541C from Intersema [5]
(6.2x6.4mm, shown in Fig. 1(b)) has been added to the
EcoMote. When submerged in water, the pressure sensor
measures water pressure ranging from 0 to 14 bars at a resolution of 1.2 mbar. The whole package is fit into a teardrop shape
and sealed waterproof with glue and acrylic.
THE PIPEPROBE SYSTEM
The PipeProbe capsule is built from the tiny EcoMote
board [4] (shown in the Fig. 1(a)). EcoMote’s physical size is
13x11x7mm, which is small enough to be dropped into nor-
Figure 1. (a) wireless sensor platform ECO (b) MS5541C pressure sensor
module
Figure 2 shows how the PipeProbe capsule is used for the
mapping (i.e., for the reconstruction of the 3D spatial layout) of
indoor water pipelines. We assume a common home scenario
where a single water input source (e.g., a water tank) is connected to several water output faucets through a network of
water pipelines with multiple forks. The PipeProbe system
works as follows. (1) A PipeProbe capsule is dropped into the
water input source. When a water faucet is opened, the force of
the resulting water flow pushes the capsule through the different forks and sections of the water pipelines. When the capsule
flows out of the open water faucet, one mapping trip is complete. The capsule is then retrieved for reuse in subsequent
mapping trips. (2) While the capsule is flowing inside a water
pipe, it gathers pressure and accelerometer readings along its
path of traversal. These sensor readings are saved in its flash
memory. After the capsule leaves the water faucet, its sensor
readings are transferred from its flash memory to a PC for postprocessing. (3) By alternating among different open water faucets and repeatedly re-inserting the PipeProbe capsule into the
water input source for multiple mapping trips, the capsule
gathers multiple sensor readings spanning the entire pipeline
network. All of the sensor readings are aggregated on a PC for
post-processing.
During post-processing, a pipeline mapping algorithm analyzes the sensor readings and computes the 3D coordinates of
the PipeProbe capsule as it moves inside the pipeline network.
The capsule’s 3D movement coordinates also mark the 3D path
of the pipeline network. Interpolating the discrete location
samples of the PipeProbe capsule over time yields the spatial
layout of the water pipelines. The pipeline mapping algorithm
is described in Section III.
The algorithm can be described using the following steps.
Step (1) estimates the flow velocity inside the pipe from the
water pressure. To accurately measure the flow velocity, the
PipeProbe capsule must be fit into a specific tear-drop shape
such that the water flow velocity approximates the capsule
flow velocity. According to fluid dynamics, equation (2) can be
used to estimate the flow velocity (v) from the water pressure
(𝑝ℎℎℎ ) inside the pipe. This equation states that the pressure
difference between the head (𝑝ℎℎℎ𝐻 ) and the tail (𝑝𝑇ℎℎℎ ) of the
capsule is square-root proportional to the flow velocity:
𝑣 = √2(𝑝ℎℎℎ𝐻 − 𝑝𝑇ℎℎℎ ℎℎℎℎℎℎℎℎℎ ).
(2)
Step (2) traces the 3D coordinates (x, y, h1) of the capsule.
To estimate the z-axis position h1, equation (3) is derived from
equation (1) by substituting 𝑣0 =0 (corresponding to the sensor
being dropped in) and 𝜌=1 (water density). ℎ0 is the height of
the input water source, 𝑝1 / 𝑝0 are the detected pressures at the
tail of the capsule and at the input water source, v is the flow
velocity from equation (2), and g is the gravity:
ℎ1 = ℎ0 +
Figure 2. Sensor moving throught the pipeline
To prevent large positional error accumulation, several
wireless beacons are placed at strategic wall/floor locations
where water pipelines are likely to pass nearby. These wireless
beacons periodically broadcast their positions to any nearby
PipeProbe capsules. The PipeProbe capsule records all received
beacons, and during post-processing, the received positions are
used to correct the accumulated positional error.
III.
THE PIPELINE MAPPING ALGORITHM
Our pipeline mapping algorithm uses the water pressure
and 3D accelerometer readings obtained from multiple mapping trips of the PipeProbe capsule(s) to map the spatial 3D
paths of the water pipelines. The algorithm is based on Bernoulli’s principle [6], which states that for an ideal flow with
no viscosity, an increase in the speed of the fluid occurs simultaneously with a decrease in pressure. Bernoulli’s equation is
shown below:
1
1
2
2
𝑝0 + 𝜌𝑣02 + 𝜌𝑔ℎ0 = 𝑝1 + 𝜌𝑣12 + 𝜌𝑔ℎ1
.
(1)
This equation is applicable to any two positions in the same
connected pipeline network: 𝑝0 / 𝑝1 are the pressures at the two
different positions, 𝜌 is the density of the water, v0 / v1 are the
flow velocities at the two different positions, g is the local acceleration due to gravity, and h0 / h1 are the heights of the two
different positions. By setting the first position to the water
input source (𝑝0 , v0, and h0 can be measured at the water input
source) and assigning the second position to the moving position of the capsule (𝑝1 is sensed from the capsule), v1 and h1
can be derived by extending equation (1). Given the 3D accelerometer readings, the pipeline mapping algorithm computes
the 3D movement coordinates of the capsule.
(𝑝1 −𝑝0)
𝑔
−
𝑣2
2𝑔
.
(3)
To determine the (x, y) coordinates of the capsule, the 3axis accelerometer reading from the capsule is first analyzed to
determine its movement direction θ over the x-y plane. Integrating velocity v from step (1) over time yields the relative
displacement d. Vector calculus over the displacement d and
the movement direction θ produces the relative (x, y) coordinate movement.
In step (3), for each mapping trip, a 3D water pipeline path
is drawn beginning from the position of the water input source
toward the position of the water output faucet where the
PipeProbe capsule was retrieved.
IV.
FUTURE WORK
We are currently implementing the PipeProbe system and
look forward to evaluating its positioning accuracy.
REFERENCES
[1]
Y. Kim, T. Schmid, Z. M.Charbiwala, J. Friedman,and M. B. Srivastava
“NAWMS: Nonintrusive Autonomous Water Monitoring System,” in
Proceedings of the 6th ACM Conference on Embedded Network Sensor
Systems, 2008, pp. 309–322.
[2] I. Stoianov, L. Nachman, S. Madden, T. Tokmouline,and M. Csail,
"PIPENET: A Wireless Sensor Network for Pipeline Monitoring," in
Proceedings of the 6th IPSN, 2007, pp. 264–273.
[3] D. Culler, D. Estrin, M. Srivastava, “Overview Of Wireless Sensor
Networks,” IEEE Computer, Special Issue in Sensor Networks, August
2004.
[4] C. Park, J. Liu, and P. H. Chou, "Eco: an Ultra-Compact Low-Power
Wireless Sensor Node for Real-Time Motion Monitoring," in
Proceedings of the 4th International Conference on Information
Processing in Sensor Networks, 2005, pp. 398–403.
[5] http://www.intersema.ch/products/guide/calibrated/ms5541c/
J. D. Anderson. Computational Fluid Dynamics: the Basics with Applications,
McGraw-Hill,
Inc.,
Singapore,
1995.