FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
(See section 10, rule 13)
1. Title of the invention: CONTAMINATION SENSOR POSITIONING IN A FLUID
DISTRIBUTION SYSTEM
2. Applicant(s)
NAME NATIONALITY ADDRESS
TATA CONSULTANCY
SERVICES LIMITED
Indian Nirmal Building, 9th Floor,
Nariman Point,
Mumbai 400021, Maharashtra
India
3. Preamble to the description
COMPLETE SPECIFICATION
The following specification particularly describes the invention and the manner in which it
is to be performed.
1
2
TECHNICAL FIELD
[0001] The present subject matter relates, in general, to design of fluid distribution
systems and, in particular, to contamination sensor positioning in the fluid distribution
systems.
BACKGROUND
[0002] Fluid distribution systems, such as water distribution systems and air distribution
systems, say heating, ventilation, and air conditioning (HVAC) systems, are always
susceptible to contamination, both accidently as well as deliberately. For example, the water
supply systems were reported to have been contaminated in certain parts of Europe during the
Second World War. Such introduction of contaminants into water distribution systems can
have severe adverse health effects on a population, and can even have potential social and
economic impacts. Therefore, in recent times, a lot of interest has been shown in developing
techniques to address the problem of contamination of fluid distribution systems.
[0003] Conventionally, one of the most widely used techniques for detecting
contamination of the fluid distribution system involves deploying contamination sensors in
the fluid distribution system. The contamination sensors sense whether the fluid, say water or
air, has a contaminant, and provides the sensed input to an operator of the fluid distribution
network, who can take further appropriate steps, say to contain the contamination from
spreading. As will be understood, the contamination sensors are configured in such a way that
the contaminant is sensed when the quantity of the contaminant is greater than a threshold
prescribed value. In an example, the threshold prescribed value can be determined based on
the quantity of the contaminant considered innocuous for humans.
[0004] The deployment of the contamination sensors in the fluid distribution network can
be done in a variety of ways. Conventionally, one way is to provide one contamination sensor
at every node of the network. However, the cost associated with sensor placement restricts the
number of sensors that can be used in the fluid distribution network. Therefore, a limited
number of contamination sensors have to be positioned for optimized detection and
containment in a large distribution system.
3
SUMMARY
[0005] System and computer implemented method for contamination sensor positioning
in a fluid distribution system are described herein. In an implementation, the computer
implemented method includes defining an objective function for optimized positioning of a
plurality of contamination sensors in the fluid distribution system. Based on flow properties of
a fluid circulating in the fluid distribution system, at least one impact parameter is ascertained.
The impact parameter is indicative of an impact of positioning at least one sensor from the
plurality of contamination sensors at a predetermined potential location in the fluid
distribution system. Further, the impact parameter takes into account a time of introduction of
a contaminant into the fluid distribution system. Additionally, a final location for positioning
each contamination sensor from among the plurality of contamination sensors is determined,
based on the objective function and the at least one impact parameter.
[0006] This summary is provided to introduce concepts related to contamination sensor
positioning in a fluid distribution system, which are further described below in the detailed
description. This summary is not intended to identify essential features of the claimed subject
matter nor is it intended for use in determining or limiting the scope of the claimed subject
matter.
BRIEF DESCRIPTION OF THE DRAWINGS
[0007] The detailed description is described with reference to the accompanying figure(s).
In the figure(s), the left-most digit(s) of a reference number identifies the figure in which the
reference number first appears. The use of the same reference number in different figure(s)
indicates similar or identical items. The features, aspects, and advantages of the subject matter
will be better understood with regard to the following description, and the accompanying
drawings.
[0008] Fig. 1 illustrates a network implementation of a sensor-location design system for
contamination sensor positioning in a fluid distribution system, according to an
implementation of the present subject matter.
[0009] Fig. 2a, 2b, 2c, 2d, 3a, 3b, 3c, and 3d illustrate various graphs of impact of the
positioning of contamination sensors, in accordance with an embodiment of the present
subject matter.
4
[0010] Fig. 4 illustrates a method for contamination sensor positioning in fluid
distribution systems, in accordance with an embodiment of the present subject matter.
DETAILED DESCRIPTION
[0011] The subject matter described herein relates to systems and methods for positioning
of a plurality of contamination sensors in a fluid distribution system. Examples of fluid
distribution system can include, but are not limited to, a water distribution system or an air
distribution system, such as a heating, ventilation, and air-conditioning (HVAC) system.
[0012] Generally, fluid distribution systems being susceptible to contamination, efforts
are made in developing effective techniques for detecting contamination in the fluid
distribution system. One of such techniques includes deployment of contamination sensors in
the fluid distribution system. However, positioning the sensors in the fluid distribution system
poses a challenge in using such techniques for effective positioning of sensors in the fluid
distribution system.
[0013] More so, such a situation is seen to occur in cases in which the contamination
sensors are to be deployed in a large fluid distribution system, spanning to thousands of
nodes. A node of the fluid distribution system can be understood as such a junction in the
network of conduits and passages at which a change in the flow pattern in the entire flow
distribution system can be effected. For example, in case of a water distribution system, the
nodes can include a water source, a water outlet, a pipe junction, a reservoir, and a pump,
deployed in the water distribution network. In the aforementioned case, deployment of one
contamination sensor at each node of the fluid distribution system may exponentially affect
the cost of instrumentation in the fluid distribution system.
[0014] Further, certain conventional techniques of sensor positioning use a collection of
design objectives for determining the location of sensors in the fluid distribution systems.
Such design objectives include contaminant transport in the fluid distribution system,
response of the contamination sensors, contamination event detection, emergency response of
the contamination sensors, and installation and maintenance costs of the fluid distribution
system. In addition, various other parameters influence the position of the contamination
sensors and play a role in effective detection of contamination in the fluid distribution system.
5
[0015] However, taking into account all such design objectives for determining the sensor
positions can involve a large amount of computational resources and may also take a large
amount of time for computation and obtaining the locations for positioning the contamination
sensors in the fluid distribution system. On the other hand, considering only a few of the
design objectives as described above, can adversely affect the operational effectiveness of the
various contamination sensors, when positioned based on such design objectives. As a result,
the purpose of having contamination sensors for contamination detection may not be fully
served.
[0016] The present subject matter describes aspects related to positioning of a plurality of
contamination sensors in a fluid distribution system. According to an implementation, a
layout of a fluid distribution system, for which the sensor positioning is to be achieved, is
obtained. Based on the layout of the fluid distribution system, potential locations of a plurality
of contamination sensors in the fluid distribution system are identified. Further, the impact of
placement of the contamination sensors at the respective potential locations is taken into
account for determining a final location of each of the contamination sensors in the fluid
distribution system. The impact of the placement of the contamination sensors, in turn, is
captured by considering a timing of introduction of contaminants into the fluid distribution
system.
[0017] In one implementation of the present subject matter, an objective function is
defined, the objective function being indicative of the design objectives on the basis of which
the sensor-location design is to be achieved. According to an implementation, a plurality of
design objectives can be identified for defining the objective function. In one example, the
design objectives can be identified from among a plurality of given design objectives, the
identified design objectives being the ones which take into account a holistic approximation
of as many of the other design objectives as possible. For instance, in one case, the effect of
the placement of various sensors on each design objective from the plurality of given design
objectives can be determined. Based on the determined effects, the design objectives to be
used in the objective function can be identified and selected. For example, if there is a high
correlation between a first design objective and two other design objectives amongst the
plurality of given design objectives, the first design objectives can be used as a proxy for the
6
two other design objectives and hence all three design objectives can be encompassed into a
single identified design objective.
[0018] In one implementation, the present subject matter can identify expected volume of
contaminated fluid consumed before detection of contamination as a proxy for population
affected and time to detection and a contamination detection possibility as the design
objectives. As will be understood, in case of an air distribution system, the consumption of
fluid can refer to intake or inhaling of the fluid, i.e., air. Further, the expected volume of
contaminated fluid consumed can be an appropriate representative of two other design
objectives, namely, expected time to detection of contamination and expected population
affected due to contamination. As a result, multiple design objectives are folded into a single
design objective and, accordingly, an accurate objective function is obtained based on which
the locations in the fluid distribution system for positioning the sensors are determined.
[0019] Further, for defining the objective function, a plurality of objective parameters can
be determined, and on the basis of the objective parameters, the objective function can be
defined. In one example, while defining the objective function, the expected volume of
contaminated fluid consumed before detection of contamination is denoted by a volume
objective parameter. And, in said example, the contamination detection possibility or, in other
words, the likelihood of contamination being detected, is denoted by a detection-possibility
objective parameter. Further, in said example, the volume objective parameter can, in turn, be
a normalized value of decrease in the total contaminated fluid consumed, represented by a
fraction of a decrease in the total volume of contaminated fluid consumed due to placement of
one sensor in the fluid distribution system with respect to the maximum possible
contaminated volume that can be consumed in the fluid distribution system. Additionally, the
detection-possibility objective parameter can take into account a marginal increase in the
contamination detection possibility due to placement of one sensor in the fluid distribution
system, and can be depicted by a fraction of the total number of detections achieved to a total
number of scenarios taken into account for studying the impact of placement of sensors in the
fluid distribution system. In addition, according to an implementation, a weighing factor can
also be introduced in the objective function, which provides a trade-off between the two
parameters in the objective function.
7
[0020] Additionally, as mentioned previously, the time of introduction of the contaminant
into the fluid distribution system to capture the dynamics of transport of the contaminant in
the fluid distribution system, is taken into consideration. In an implementation, in order to
consider the time of introduction of the contaminant, the determination of the impact of
positioning of the contamination sensors begins substantially at the instance when a change in
demand pattern occurs in the fluid distribution system. In one example, the demand pattern in
a fluid distribution system can be understood as a behavioral consumption of the fluid in the
system, which can affect the transport of the fluid inside the network.
[0021] Accordingly, based on duration of operation of the fluid distribution system and
demand patterns at various nodes of the fluid distribution system, the time instances, at which
the demand pattern at one or more node changes, are identified. In an implementation, such
identification can be done on the basis of a simulation of the layout of the fluid distribution
system having the fluid circulating within the network. Further, a time period between two
successive time instances of change of demand pattern is identified as an epoch during which
the demand pattern, and hence, the flow pattern in the fluid distribution system is static.
[0022] Subsequently, the total duration of operation of the fluid distribution system is
divided into numerous epochs, and the flow properties of the fluid in the fluid distribution
system are determined for each epoch. In an example, the flow properties can include flow
values for fluid in the fluid distribution system. According to an implementation, such flow
properties can be determined for each epoch in the total duration of operation of the fluid
distribution system. Considering the transportation properties and dispersion of the
contamination in the fluid distribution system, based on the time of introduction of the
contaminant, for the purpose of determining the locations for positioning the contamination
sensors allows for substantially less expenditure of computational resources to be used. At the
same time, the accuracy and the effectiveness of contamination detection, by positioning the
contamination sensors at the determined locations, is not compromised.
[0023] Further, once the flow properties for the fluid in the fluid distribution system have
been determined, the impact of placement of the contamination sensors is determined for
obtaining the sensor location design for the fluid distribution system. In an implementation,
for determining the impact of placement of the contamination sensors, a scenario impact
mapping is generated. The scenario impact mapping can take into consideration various
8
scenarios influencing the effective detection of contamination in the fluid distribution system.
In an example, each scenario can indicate a combination of each respective potential location
of the contamination sensors and each node of the fluid distribution system. Further, in an
example, the potential locations for positioning the contamination sensors can be determined
based on fixed locations for positioning the contamination sensors in the fluid distribution
system, say the nodes of the fluid distribution system. Such potential locations can be
previously defined, say by a designer of the layout of the fluid distribution system.
[0024] Accordingly, for each scenario in the scenario impact mapping, one or more
impact parameters can be determined, based on the flow properties of the fluid. The impact
parameters can include a volume impact parameter, a time impact parameter, and a detection
impact parameter. In an example, the volume impact parameter indicates the total volume of
contaminated fluid consumed in the entire network, when for the given scenario, the
contamination occurs at the node included in that scenario. Further, the time impact parameter
can indicate a time taken for detection of contamination by the contamination sensor for that
scenario, and the detection impact parameter can indicate whether the detection has occurred
or not. Hence, for all the scenarios, the impact parameters can be determined and populated in
the scenario impact mapping.
[0025] As will be understood, since the flow properties for the fluid are determined for the
individual epochs that the entire operational duration of the fluid distribution system is
divided into, the impact parameters determined for each scenario is also determined for each
epoch in the entire operational duration.
[0026] Subsequently, the final location for positioning of each contamination sensor can
be determined, based on the impact parameters. In an implementation, the final location of
one contamination sensor from among the plurality of contamination sensors is determined,
and subsequently, the final location of the rest of the contamination sensors is determined
based on the already determined final locations of the previous contamination sensor(s).
[0027] According to an implementation, an overall impact parameter is determined for
one or more of the volume impact parameter, time impact parameter, and the detection impact
parameter, for determining the final location for the contamination sensor. The overall impact
parameter can be understood as the value of the impact parameter for the entire duration of
operation of the fluid distribution system and for all the scenarios considered for that sensor in
9
the scenario impact mapping. For example, for the latter case, the impact on volume of
contaminated fluid consumed and the detection possibility at various nodes of the fluid
distribution system with reference to the placement of the sensor at one potential location can
be used. For the former case, the impact parameter for all the epochs are obtained based on
the impact parameter for each individual epoch in the operational duration, say by integrating
the values of impact parameters for all the individual epochs in the operational duration.
[0028] In an implementation, an overall volume impact parameter and an overall
detection impact parameter can be determined for obtaining the final location. In one
example, the overall volume impact parameter can indicate the total volume of contaminated
fluid consumed in the fluid distribution system when the contamination sensor is positioned at
one potential location and the contamination occurs at all the nodes, whereas the overall
detection impact parameter can indicate the total number of contamination events that can be
detected by positioning the contamination sensor at that potential location, when
contamination occurs at all the nodes of the fluid distribution system.
[0029] Further, according to an aspect, based on the overall impact parameter and the
objective function, the final location for positioning the contamination sensor under
consideration is obtained. In an implementation, a location, from among the various locations
covered in the different scenarios, for which a predefined optimal value of the objective
function is achieved, is selected as the final location for that contamination sensor. For
example, the normalized value of decrease in the total volume of contaminated fluid
consumed and the marginal increase in the contamination detection possibility due to the
placement of the sensor are determined. In said example, the location of the contamination
sensor for which the value of a sum of these two factors is maximized, can be selected as the
final location for that contamination sensor.
[0030] Further, based on the location determined for one sensor, the location for
positioning the subsequent contamination sensor is determined. In an implementation, the
impact of the final location of the contamination sensor is taken into account for determining
the final location of the rest of the contamination sensors. Accordingly, an updated impact
parameter is determined again for the scenarios in the scenario impact mapping, i.e., at rest of
the potential sensor locations, with the contamination sensor positioned at the determined
10
final location in the fluid distribution system. The updated impact parameter is determined
from the existing impact parameter values and the sensors placed.
[0031] In one implementation, an updated time impact parameter is determined for all the
scenarios in the scenario impact mapping, based on the positioning of the contamination
sensor in the final location. Since the effect of placement of the contamination sensor at the
final location would reduce the detection time of contamination in the system, the updated
time impact parameter would indicate an overall reduction in time taken for the detection of
contamination with the sensor positioned at the final location. Additionally, in an example,
based on the updated time impact parameter for all the scenarios, the updated volume
parameter can be determined for all the scenarios. As will be understood, in other
implementations, the updated impact parameters can be determined in various combinations
and in various sequences.
[0032] Subsequently, the scenario impact mapping is updated by including the updated
impact parameter therein, and an updated scenario impact mapping is obtained. Further, the
updated scenario impact mapping is employed to determine the impact of positioning of
another contamination sensor. Accordingly, the final location for positioning the other sensor
in the fluid distribution system can be determined based on the updated scenario impact
mapping and the objective function, in the same manner as described above, for effective
detection of contamination in the fluid distribution system. Further, the same technique can be
iteratively employed for obtaining the respective final locations for all the contamination
sensors available for deployment.
[0033] While aspects of described systems and methods for contamination sensor
positioning in the fluid distribution systems can be implemented in any number of different
computing systems, environments, and/or configurations, the embodiments are described in
the context of the following system architecture(s).
[0034] Fig. 1 illustrates a network 100 implementing a sensor-location design system 102
for contamination sensor positioning in a fluid distribution system, according to an
embodiment of the present subject matter. The fluid distribution system can be, for example, a
water distribution system or an air distribution system. In an example, the water distribution
systems can be a water supply network catering to a locality or a city. Further, the air
distribution system can include a heating, ventilation, and air-conditioning (HVAC) system of
11
a building or a cooling system of a server centre. The sensor-location design system 102,
hereinafter referred to as the system 102, can be configured to determine the locations in the
fluid distribution system for positioning a plurality of contamination sensors, to effectively
detect contamination in the fluid distribution system.
[0035] Further, in said embodiment, the system 102 is connected to and interacts with an
auxiliary system 104 over a communication network 106. The auxiliary system 104 can be
configured to provide a layout of the fluid distribution system for which the sensor position is
to be achieved. In an implementation, the auxiliary system 104 can deploy a simulation tool,
such as EPANETTM, which can be used to simulate the layout of the fluid distribution system
with the fluid circuiting in conduits and passages of the simulated fluid distribution system. In
turn, the system 102 can be employ the simulated layout of the fluid distribution system and
determine the sensor locations based on the layout.
[0036] In an implementation, the system 102 and the auxiliary system 104 can be
implemented as a variety of computing devices, including, for example, servers, desktop PCs,
notebooks or portable computers, workstations, mainframe computers, mobile computing
devices, cellular phones, entertainment devices, PDAs, and internet appliances.
[0037] The system 102 is connected to the auxiliary system 104 over the communication
network 106 through one or more communication links. The communication links can be
enabled through a desired form of communication, for example, via dial-up modem
connections, cable links, digital subscriber lines (DSL), wireless or satellite links, or any other
suitable form of communication.
[0038] Further, the communication network 106 may be a wireless network, wired
network or a combination thereof. The communication network 106 can be implemented as
one of the different types of networks, such as intranet, local area network (LAN), wide area
network (WAN), the internet, and such. The communication network 106 may either be a
dedicated network or a shared network, which represents an association of the different types
of networks that use a variety of protocols, for example, Hypertext Transfer Protocol (HTTP),
Transmission Control Protocol/Internet Protocol (TCP/IP), Wireless Application Protocol
(WAP), to communicate with each other. Further, the communication network 106 may
include a variety of network devices, including routers, bridges, servers, computing devices,
storage devices.
12
[0039] Further, in one implementation, the system 102 includes processor(s) 108 coupled
to modules 110. The system 102 can further include memory 112, and interface(s) 114. The
processor(s) 108 can be a single processing unit or a number of units, all of which could
include multiple computing units. The processor(s) 108 may be implemented as one or more
microprocessors, microcomputers, microcontrollers, digital signal processors, central
processing units, state machines, logic circuitries, and/or any devices that manipulate signals
based on operational instructions. Among other capabilities, the processor(s) 108 can be
configured to fetch and execute computer-readable instructions and data stored in the memory
112.
[0040] The module(s) 110 can include routines, programs, objects, components, data
structures, etc., which perform particular tasks or implement particular abstract data types. In
another implementation, the module(s) 110 may be implemented as, signal processor(s), state
machines, logic circuitries, and/or any devices or components that manipulate signals based
on operational instructions. In an implementation, the module(s) 110 include, for example, an
objective definition module 116, a flow determination module 118, an impact determination
module 120, a location determination module 122, and other module(s) 124. The other
module(s) 124 can supplement applications or functions performed by the system 102.
[0041] The memory 112 may include any non-transitory computer-readable medium
known in the art including, for example, volatile memory, such as Static Random Access
Memory (SRAM), Dynamic Random Access Memory (DRAM), etc., and/or non-volatile
memory, such as Erasable Program Read Only Memory (EPROM), and flash memory. The
non-transitory computer-readable medium, however, excludes a transitory, propagating
signal. The memory 112 can further include objective function data 126, fluid distribution
system (FDS) data 128, scenario impact data 130, location data 132, and other data 134. The
other data 134 can include data generated as a result of the execution of one or more modules
in the other modules 124. The data 126, 128, 130, 132, and 134 in the memory 112, amongst
other things, serves as a repository for storing data processed, received, and generated by one
or more of the module(s) 110.
[0042] Further, the interface(s) 114 may include a variety of software and hardware
interfaces, for example, interfaces for peripheral device(s), such as a keyboard, a mouse, an
external memory, and a printer. Further, the interface(s) 114 enables the system 102 to
13
communicate with other devices, such as web servers and external repositories. The
interface(s) 114 can also facilitate multiple communications within a wide variety of networks
and protocol types, including wireless networks, such as WLAN, cellular, or satellite. For the
purpose, the interface(s) 114 may include one or more ports.
[0043] As mentioned previously, in operation, the system 102 is configured to achieve
positioning of a plurality of contamination sensors in a fluid distribution system. According to
an implementation, the system 102 takes into account the impact of placement of the
contamination sensors at the respective potential locations for determining a final location of
each of the contamination sensors in the fluid distribution system. In turn, the system 102 can
consider a timing of introduction of contaminants into the fluid distribution system to capture
the impact of the placement of the contamination sensors. In an implementation, the sensorlocation
design or layout achieved by the system 102 provides optimized performance of the
contamination sensors in detecting contamination in the fluid distribution system.
[0044] In one implementation of the present subject matter, in order to initiate the process
of determining the locations for the contamination sensors, the objective definition module
116 defines an objective function. In an example, the objective function can be indicative of
the design objectives on the basis of which the sensor-location design is to be achieved.
According to said implementation, the objective definition module 116 can be further
configured to identify a plurality of design objectives for defining the objective function. In
one example, the objective definition module 116 can identify the design objectives from
among a plurality of given design objectives, the objective definition module 116 being
configured to identify those design objectives for defining the objective function which take
into account a holistic approximation of as many of the other design objectives as possible.
For instance, in one case, the objective definition module 116 can determine the correlation of
the effect of the placement of various sensors on each design objective from the plurality of
design objectives. Based on the correlation, the objective definition module 116 can identify
and select the design objectives for the objective function.
[0045] In one implementation, the objective definition module 116 can identify expected
volume of contaminated fluid consumed before detection of contamination and a
contamination detection possibility as the design objectives. As will be understood, in case of
14
an air distribution system, the consumption of fluid can refer to intake or inhaling of the fluid,
i.e., air.
[0046] Further, as will be understood, the expected volume of contaminated fluid
consumed can be an appropriate representative of two other design objectives, namely,
expected time to detection of contamination and expected population affected due to
contamination. As a result, the objective definition module 116 can encompass multiple
design objectives into a single design objective and, accordingly, obtain an accurate objective
function, based on which the system 102 can determine the locations in the fluid distribution
system for positioning the sensors.
[0047] Further, in an implementation, the objective definition module 116 can be
configured to determine a plurality of objective parameters for defining the objective function.
And on the basis of the objective parameters, the objective definition module 116 can define
the objective function. In one example, the objective definition module 116 can represent the
first design objective, namely, expected volume of contaminated fluid consumed to before
detection of contamination by a volume objective parameter, and the second design objective
of contamination detection possibility by a detection-possibility objective parameter.
[0048] Further, in turn, the objective definition module 116 can be configured to select a
relation for the volume objective parameter and the detection-possibility objective parameter.
In an example, the objective definition module 116 can select the following relation for the
volume objective parameter:
[0049] In the above example, the relation depicts a normalized value of decrease in the
total contaminated fluid consumed, represented by a fraction of a decrease in the total volume
of contaminated fluid consumed due to placement of one sensor in the fluid distribution
system (Vn) with respect to the maximum possible contaminated volume that can be
consumed in the fluid distribution system (Vmax).
[0050] Additionally, in an example, the objective definition module 116 can select the
following relation for the detection-possibility objective parameter:
15
[0051] In the above mentioned example, the relation for the detection-possibility objective
parameter can represent a marginal increase in the contamination detection possibility due to
placement of one sensor in the fluid distribution system, and can be depicted by a fraction of
the total number of detections achieved to a total number of scenarios taken into account for
studying the impact of placement of sensors in the fluid distribution system.
[0052] Accordingly, in one example, the objective definition module 116 can define the
following normalized objective function, taking into account the above mentioned objective
parameters:
[0053] As will be understood, the objective function defined by the objective definition
module 116 considers the decrease in the contaminated volume of fluid consumed and an
increase in the contamination detection possibility. In addition, the objective definition
module 116 can include a weighing factor (α) in the objective function, for providing a tradeoff
between the two objective parameters. Further, the objective definition module 116 can
store the defined objective function in the objective function data 126.
[0054] Additionally, the flow determination module 118 can take into account the time of
introduction of the contaminant into the fluid distribution system to capture the dynamics of
transport of the contaminant in the fluid distribution system. In an implementation, in order to
take the time of introduction of the contaminant, the flow determination module 118 can
begin the determination of the impact of positioning of the contamination sensors
substantially at the instance when a change in demand pattern occurs in the fluid distribution
system.
[0055] Accordingly, the flow determination module 118 can obtain the simulation of the
fluid distribution system from the auxiliary system 104, the simulation including the layout of
the fluid distribution system and also providing information regarding the fluid circulating in
the fluid distribution system. Further, the flow determination module 118 can employ the
simulation of the fluid distribution system, and the based on duration of operation of the fluid
16
distribution system and demand patterns at various nodes of the fluid distribution system, the
flow determination module can determine the time instances, at which the demand pattern at
one or more node changes, are identified.
[0056] Further, for the purpose of simplifying the process, the flow determination module
118 can determine the time periods during which the contamination has a fixed concentration.
Accordingly, in an implementation, the flow determination module 118 can determine a time
period between two successive time instances at which change of demand pattern occurs, and
such a time period is referred to as an epoch during which the demand pattern, and hence, the
flow pattern in the fluid distribution system is static. In an example, the flow determination
module 118 can store the determined epoch in the FDS data 128. Further, to take into account
the effect of the transportation and dispersion of the contaminant in the fluid distribution
system, the flow determination module 118 can determine the flow properties of the fluid in
the fluid distribution system for such an epoch during which the flow pattern in the fluid
distribution system remains static. In an example, the flow properties can include flow values
for fluid in the fluid distribution system, and can be determined based on the simulation of the
fluid distribution system retrieved from the auxiliary system 104. In another example, the
system 102 can connect to an actual fluid distribution system for which the sensor-location
design is to be achieved, and obtain the flow properties from various flow sensors deployed in
that fluid distribution system. Further, in one example, the flow determination module 118
can store the determined flow properties the FDS data 128.
[0057] Accordingly, the flow determination module 118 can be configured to segregate
the total duration of operation of the fluid distribution system into a plurality of epochs, and
determine the flow properties of the fluid for each epoch. In one example, as mentioned
above, the flow determination module 118 can store such information regarding the epochs
and the flow properties in the FDS data 128. By considering the transportation properties and
dispersion of the contamination in the fluid distribution system, based on the time of
introduction of the contaminant, the flow determination module 118 allows for substantially
less expenditure of computational resources to be used. At the same time, the accuracy and
the effectiveness of contamination detection, by positioning the contamination sensors at the
determined locations, is not compromised.
17
[0058] Further, once the flow determination module 118 has determined the flow
properties for the fluid to account for the transportation and dispersion of the fluid in the fluid
distribution system, the impact determination module 120 can subsequently determine the
impact of potentially placing the contamination sensors in the fluid distribution system. In an
implementation, the impact determination module 120 can be configured to generate a
scenario impact mapping for determining the impact of potential placement of the
contamination sensors. During such determination, the impact determination module 120 can
consider various scenarios involving impact of positioning each contamination sensor at a
potential location in the fluid distribution system and then determining the effect of such
positioning on the overall contamination detection by the other contamination sensors.
[0059] Hence, while generating the scenario impact mapping, the impact determination
module 120 can take into consideration various scenarios influencing the effective detection
of contamination in the fluid distribution system, and as described, each scenario can indicate
a combination of each respective potential location of the contamination sensors and each
node of the fluid distribution system. In an example, the scenario impact mapping can be in
the form of a matrix, with each scenario, considered for determining the final locations, being
represented by an entry M(i, j) of the matrix.
[0060] Further, in an example, the potential locations for positioning the contamination
sensors can be determined based on fixed locations for positioning the contamination sensors
in the fluid distribution system, say the nodes of the fluid distribution system. Such potential
locations can be previously defined, say by a designer of the layout of the fluid distribution
system. Additionally, the impact determination module 120 can also identify the potential
locations for the contamination sensors based on the number of contamination sensors
available for deployment and the number of locations in the fluid distribution system where
the contamination sensors can be deployed.
[0061] Further, as part of determining the impact of the potentially positioning the
contamination sensors, the impact determination module 120 can select one or more impact
parameters to be determined for each scenario in the scenario impact mapping, the impact
parameters being determined based on the flow properties of the fluid. In an implementation,
the impact determination module 120 can include one or more of a volume impact parameter,
a time impact parameter, and a detection impact parameter. In an example, the volume impact
18
parameter is represented as the total volume of contaminated fluid consumed in the entire
network (Vn), when for the given scenario, the contamination occurs at the node included in
that scenario. Further, the time impact parameter can be represented as a time taken for
detection of contamination by the contamination sensor for that scenario (Td), and the
detection impact parameter (D) can be represented as whether the detection has occurred or
not.
[0062] In an implementation, the impact determination module 120 can determine the
impact parameters for all the scenarios in the scenario impact mapping and populate the
values in the scenario impact mapping. As will be understood, since the flow properties for
the fluid are determined for the individual epochs that the entire operational duration of the
fluid distribution system is divided into, the impact determination module 120 can determine
the impact parameters for each scenario, for each epoch in the entire operational duration. In
an implementation, the impact determination module 120 can store the information relating to
the potential locations of the sensors, available sensors for deployment, the scenario impact
mapping, and the impact parameters in the scenario impact data 130.
[0063] Subsequently, the location determination module 122 can determine the final
location for positioning each contamination sensor, based on the scenario impact mapping and
the impact parameters populated therein. In an implementation, the location determination
module 122 can be configured to determine the final location of one contamination sensor
from among the plurality of contamination sensors, and subsequently, determine the final
location of the rest of the contamination sensors based on the already determined final
locations for the previous contamination sensor(s).
[0064] According to an aspect, for determining the final location of one sensor, the
location determination module 122 can determine an overall impact parameter is for one or
more of the impact parameters, in one case, the volume impact parameter, the time impact
parameter, and the detection impact parameter. The overall impact parameter can be
understood as the value of the impact parameter for the entire duration of operation of the
fluid distribution system and for all the scenarios considered for that sensor in the scenario
impact mapping. For example, for the latter case, the impact on volume of contaminated fluid
consumed and the detection possibility at various nodes of the fluid distribution system with
reference to the placement of the sensor at one potential location can be used. For the former
19
case, the impact parameter for all the epochs are obtained based on the impact parameter for
each individual epoch in the operational duration, say by integrating the values of impact
parameters for all the individual epochs in the operational duration.
[0065] In one implementation of the present subject matter, the location determination
module 122 can be configured to determine an overall volume impact parameter and an
overall detection impact parameter for obtaining the final location. In one example, the overall
volume impact parameter can indicate the total volume of contaminated fluid consumed in the
fluid distribution system when the contamination sensor is positioned at one potential location
and the contamination occurs at all the nodes, whereas the overall detection impact parameter
can indicate the total number of contamination events that can be detected by positioning the
contamination sensor at that potential location, when contamination occurs at all the nodes of
the fluid distribution system.
[0066] Further, according to an aspect, based on the overall impact parameter and the
objective function, the location determination module 122 can determine the final location for
positioning the contamination sensor under consideration. In an implementation, for each
location covered in the various scenarios in the scenario impact mapping for the
contamination sensor, the location determination module 122 determines the value of the
objective function. In said implementation, the location determination module 122 determines
that location from among the plurality of locations, for which a predefined optimal value of
the objective function is achieved. Further, the location determination module 122 selects that
location as the final location for that contamination sensor.
[0067] As discussed earlier, in one example, the location determination module 122
determines the final location for the sensor on the basis of the maximization of the objective
function, i.e., maximization of the cumulative value of the objective parameters. In said
example, the location determination module 122 determines the normalized value of decrease
in the total volume of contaminated fluid consumed and the marginal increase in the
contamination detection possibility due to the placement of the sensor for all the various
locations covered in the scenario impact mapping. In said example, the location determination
module selects that location as the final location for the contamination sensor for which the
value of a sum of these two factors is maximized.
20
[0068] Further, according to an implementation, the location determination module 122
can determine the location for positioning the rest of the sensors based on the final location
determined for one sensor. In said implementation, the location determination module 122 can
take into consideration the impact of the final location of the contamination sensor is taken
into account for determining the final location of the rest of the contamination sensors.
Accordingly, considering that the contamination sensor is positioned at the determined final
location in the fluid distribution system, the impact determination module 120 can determine
an updated impact parameter for the various scenarios in the scenario impact mapping, i.e., at
rest of the potential sensor locations, to capture the response of positioning of the
contamination sensor in the determined final location. As will be understood, the updated
impact parameter is determined in the same manner as described earlier, for example, based
on the simulation of the fluid distribution system.
[0069] In one implementation, the impact determination module 120 can determine an
updated time impact parameter for all the scenarios in the scenario impact mapping. As will
be understood, the effect of placement of the contamination sensor at the final location would
reduce the detection time of contamination at other potential locations in the fluid distribution
system where the contamination sensors can be placed. Accordingly, the updated time impact
parameter determined by the impact determination module 120 can indicate such reduction in
time taken for the detection of contamination at the other potential locations. Additionally,
according to said implementation, to obtain the updated scenario impact mapping, the location
determination module 122 determines the updated volume parameter based on the updated
time impact parameter for all the scenarios. As will be understood, the impact determination
module 120 can be configured to determine the updated impact parameters in various
combinations and in various sequences, in other implementations.
[0070] Further, the impact determination module 120 can obtain an updated scenario
impact mapping by updating the values of the impact parameters in the scenario impact
mapping based on the values of updated impact parameters determined above. In an example,
the impact determination module 120 can store the updated impact parameters and the
updated scenario impact mapping in the scenario impact data 130.
[0071] Subsequently, the location determination module 122 can use the updated scenario
impact mapping to determine the impact of positioning of another contamination sensor, in
21
the same manner as described above. Accordingly, the final location for positioning the other
sensor in the fluid distribution system can be determined based on the updated scenario
impact mapping and the objective function, as described above, for effective detection of
contamination in the fluid distribution system. As will be understood, the location
determination module 122 can iteratively determine the final locations for positioning the
contamination sensors in the manner as described above, until the final locations for
positioning all the contamination sensors available for deployment have been identified. Once
completed, the location determination module 122 can generate a sensor-location map of the
fluid distribution system, based on the simulated layout of the fluid distribution system, and
provide the same to a user who is designing the fluid distribution system for positioning the
contamination sensors.
[0072] The aforementioned features of the present subject matter are described with
reference to the following examples. Considering an example in which the scenario impact
mapping is in the form of a matrix. In such an example, as mentioned previously, each entry
of the matrix M(i, j) can indicate a scenario representing a combination of a potential location
of one sensor and a node of the fluid distribution system. In an example, each entry in the
matrix includes values of the impact parameters, namely, the volume impact parameter, the
time impact parameter, and the detection impact parameter, when a contamination event
occurs at node i and is detected by a sensor having a potential location at j.
[0073] As mentioned previously, the volume impact parameter can be the total volume of
contaminated fluid consumed in the entire network (Vn) when the contamination is at the
source node i and the contamination sensor is placed at j; the time impact parameter can be
the time to detection by the contamination sensor at node j (Td); and the detection impact
parameter can be a flag (D) that indicates whether the contamination has been detected or not.
Hence, in said example, each entry M(i, j) of the matrix is a tuple of the form (Vn, Td, D).
[0074] Further, in said example, if the contamination sensor placed at j is unable to detect
the contamination, then the impact determination module 120 can set the entry Vn to Vmax
which indicated the maximum volume of contaminated fluid that can be consumed in the
network. Further, if no detection occurs, then the impact determination module 120 sets the
entry Td to Tmax which represents the total duration of operation of the fluid distribution
22
system, say the total duration for which the fluid distribution system is studied. Additionally,
if contamination is detected at the location j of the contamination sensor when contamination
occurs at the node i, then the impact determination module 120 sets the flag D to 1, otherwise
the flag D is set to 0. As also described previously, the impact determination module 120
determines the impact parameters for all the epochs that the entire operational duration of the
fluid distribution system can be divided into.
[0075] Subsequently, in said example, to obtain the final location of the first sensor, the
location determination module 122 determines the overall impact parameter, say the overall
volume impact parameter (Vj) and the overall detection impact parameter (Dj). In an example,
the location determination module 122 can compute the sums of the columns of the matrix for
determining the overall impact parameters, which is represented as an example by the
following relations:
[0076] Further, based on the values of the overall volume impact parameter Vj and the
overall detection impact parameter Dj, the location determination module 122 determines that
location from the various potential locations j of the contamination sensor, for which the value
of the objective function is optimized. Accordingly, such location for the contamination
sensor is determined as the final location and assigned to a location variable l. Thereby,
location l in the fluid distribution system acts the final location for the first sensor.
[0077] Further, once the final location for the first sensor has been determined, the impact
determination module 120 updates the matrix representing the scenario impact mapping to
obtain the updated scenario impact mapping, for reflecting the impact of positioning of the
first sensor at the determined final location l. In an implementation, the impact determination
module 120 determines the time impact parameter, i.e., the time to detection, for each
scenario in the matrix. For example, with the first sensor already positioned at the final
23
location in the fluid distribution system, the impact determination module 120 determines the
time to detection for each combination of the contamination node i and the potential location j
of other sensors.
[0078] Accordingly, the updated values of the time impact parameter are populated in the
matrix. Further, based on the updated values of the time impact parameters, the impact
determination module 120 determines the updated values of the volume impact parameter for
each scenario and updated in the matrix. The matrix is iteratively updated as described, until
all the final locations for all the contamination sensors have been determined.
[0079] Figures 2a to 2d illustrate comparison of the performance of the contamination
sensors in the fluid distribution system positioned according to the present subject matter,
with the performance of contamination sensors positioned in the fluid distribution system
according to conventional methodologies.
[0080] In order to compare different sensor-location designs, i.e., those based on the
present subject matter, and those based on the conventional techniques, the performance of
contamination sensor location design is evaluated based on four testing parameters, namely,
time taken to detect contamination (Z1), volume of contaminated water consumed (Z2),
population affected due to contamination (Z3), and contamination detection possibility (Z4).
[0081] Further, the sensor-location designs according to the different techniques were
tested on two benchmark water distribution networks. The first benchmark water distribution
network includes 126 nodes, 1 water source, two tanks, 168 pipes, 2 pumps, and 8 valves. For
the purpose of studying the behavior of the sensor-location design, the first benchmark water
distribution network was subjected to four variable demand patterns, each of 24 hour duration,
and the behavior was studied for 96 hours.
[0082] The second benchmark water distribution network includes 12523 nodes, 2
constant head sources, 2 tanks, 14822 pipes, 4 pumps, and 5 valves. For the purpose of
studying the behavior of the sensor-location design, the second benchmark water distribution
network was subjected to five variable demand patterns, and the network behavior was
studied for 48 hours.
[0083] Further, the conventional techniques which are used for comparison of the sensorlocation
design achieved according to the present subject matter are referred to as Krause and
TEVA-SPOT. The conventional technique referred to as Krause is known to produce a large
24
number of non-dominated solutions to the problem of identifying the sensor locations for
optimized operation of the contamination sensors, however, uses greater computational
resources. Further, the conventional technique referred to as TEVA-SPOT is based on a
Greedy Randomized Adaptive Search Procedure (GRASP) heuristic, which is known to
provide accurate solutions over a wide spectrum of contamination scenarios, but such
technique also uses large amount of computational resources.
[0084] As shown, fig.2a, fig. 2b, fig. 2c, and fig. 2d illustrate the comparison of the
performance of the sensor-location design with reference to a range of contaminant injection
concentrations and durations, and illustrate the comparative performance for the four testing
parameters, namely Z1, Z2, Z3, and Z4, respectively. As mentioned previously, the sensor
locations as determined in the manner as described with reference to fig. 1 are achieved for
the assumption that the contamination takes place in fixed concentration and that the duration
of contamination event is also fixed. However, in reality, the concentration and duration,
along with the source of contamination, of contamination injection are unknown. Therefore,
the performance of the sensor-location design is achieved for a range of contamination
injection concentrations and durations.
[0085] Further, two factors are considered as part of test scenarios in the performance
comparison. The first factor refers to the test concentration of the contaminant normalized
with respect to the concentration of the contaminant assumed during the determination of the
scenario impact mapping, whereas the second factor refers to the test injection duration
normalized with respect to the injection duration assumed during determination of the
scenario impact mapping. As can be seen from the figures, for all the testing parameters, the
sensor-location design achieves a relatively equivalent performance with reference to the
conventional techniques across a range of test contaminant concentrations and injection
durations. Hence, the sensor-location design achieved according to the present subject matter
provides a robust technique for determining locations for positioning contamination sensors in
the fluid distribution system, and at the same time, involves use of substantially less
computational resources and time.
[0086] Further, fig. 3a, 3b, 3c, and 3d illustrate the comparative performance of sensorlocation
design according to the present subject matter with the conventional techniques
against the four testing parameters, namely Z1, Z2, Z3, and Z4, respectively, with reference to
25
the number of contamination events. As understood from the foregoing description, the sensor
locations as determined in the manner as described with reference to fig. 1 are achieved for
the assumption that the contamination takes place at one node at one time in the fluid
distribution system. However, in reality contaminants may be simultaneously introduced from
more than one point in the system. As is seen in fig. 3a, 3b ,3c, and 3d, the relative
performance of the sensor-location design according to the present subject matter with respect
to Krause and TEVA-SPOT is similar to that discussed with reference to fig. 2a, 2b, 2c. In
addition, it can be seen that across the different testing parameters, as the number of
simultaneous contamination increases, the detection possibility Z4 for the present subject
matter increases and the detection time Z1 decreases. However, the volume of contaminated
water consumed Z2 and the populated affected Z3 increase. This is because the presence of
several contaminant introduction points in the system reduces the average temporal distance
in the between the point of contamination and the sensor location, thereby decreasing the
detection time Z1. However, having multiple sources in the system increase the volume of
water contaminated. Therefore, even though the detection time Z1 is reduced, within this
short detection time, more volume of contaminated water gets consumed as a result of which
the volume of contaminated water consumed Z2 and the populated affected Z3 increase.
Further, having more contaminated water in the network increases the chance of it getting
detected by one of the contamination sensors, and as a result, the contamination detection
possibility Z4 increases.
[0087] Additionally, the following table illustrates the impact of the weighing factor α on
the four testing parameters, and also compares the overall performance of the sensor-location
design according to the present subject matter, with the conventional techniques. As will be
understood, for different values of the weighing factor α, the sensor locations and hence the
performance of the sensor placement will be different. The values reported in the following
table are the average of those observed when the sensor placement was tested by injecting
each possible network location with a contaminant concentration half that used for the
purpose of sensor-location design. Further, the values shown for Krause and TEVA-SPOT are
also average values, determined using the same set of boundary conditions as those for the
values for the technique according to the present subject matter.
26
Weighing factor
(α)
Detection Time,
Z1 (minutes)
Contaminated
water consumed,
Z2 (gallons)
Population
affected, Z3 (#)
Detection
possibility,Z4
0.000 618 3311.7 169.46 0.676
0.002 605.98 3332.76 173.92 0.700
0.004 608.07 3408.72 175.94 0.707
0.006 701.98 3589.76 174.65 0.723
0.008 725.18 3903.47 182.47 0.739
0.010 998.85 4354.7 188.69 0.784
Krause 818.4 4394.51 192.68 0.77
TEVA SPOT 737.73 10123.33 399.18 0.78
[0088] From the table, the correlation between the time taken to detect contamination Z1,
the volume of contaminated water consumed Z2, and the population affected due to
contamination Z3 can be seen. Accordingly, as mentioned earlier, such correlation allows the
use of the volume of contaminated water consumed Z2 to represent the other two design
parameters, without compromising on the accuracy of the sensor-location design for effective
contamination detection. Further, the results shown in the table above also corroborate that the
volume of contaminated water consumed Z2 can represent the time to detection Z1 and the
population affected Z3. As can be understood from the table, when the volume consumed Z2
increases, the time to detection Z1 and the population affected Z3 seem to increase as well.
Further, as seen, the reversal in trend in certain cases is due to the fact that the detection
threshold of contaminant (≈0.1 mg/l) and hazard threshold (≈0.3 mg/l) may not be equal in the
test scenarios, which causes a reduction in the correlation between the three parameters.
[0089] Further, it can be seen from the results that as the weighing factor α increases, the
performance of the contamination sensors located according to the present subject matter is
enhanced in terms of the contamination detection possibility Z4, while compromising on the
other three. Varying the weighing factor α allows to determine a series of non-dominating
solutions for the sensor-location design, from which a user or an administrative personnel
responsible for a fluid distribution system can select the sensor-location design based on
based on their own set of regulations, requirements, and preferences.
[0090] Further, it can also be seen from the table that for a range of the values of the
weighing factor α, the contamination sensors perform better than both Krause and TEVA27
SPOT in terms of the time to detection Z1, the volume consumed Z2, and the population
affected Z3, but the performance with respect to the detection possibility Z4 is slightly poor.
However, the overall enhancement in performance is substantially greater than the
degradation witnessed. For instance, when the weighing factor α = 0:004, when compared
with Krause, the performance of the contamination sensors according to the present subject
matter improves Z1, Z2, and Z3 by around 25.67%, 22.43%, and 8.69% respectively while
worsening Z4 by 8.18%. When compared with TEVA-SPOT, the improvement in the
performance with regard to Z1, Z2, and Z3 is of around 17.58%, 66.32%, and 55.92%
respectively, while worsening Z4 by 9.12%.
[0091] Fig. 4 illustrates a method 400 for contamination sensor positioning in fluid
distribution systems, according to an implementation of the present subject matter. In one
example, the method 400 is carried out by the sensor-location design system 102. The method
400 may be described in the general context of computer executable instructions. Generally,
computer executable instructions can include routines, programs, objects, components, data
structures, procedures, modules, functions, etc., that perform particular functions or
implement particular abstract data types. The method may also be practiced in a distributed
computing environment where functions are performed by remote processing devices that are
linked through a communications network. In a distributed computing environment, computer
executable instructions may be located in both local and remote computer storage media,
including memory storage devices.
[0092] The order in which the method 400 is described is not intended to be construed as
a limitation, and any number of the described method blocks can be combined in any order to
implement the method 400, or an alternative method. Additionally, individual blocks may be
deleted from the method 400 without departing from the spirit and scope of the subject matter
described herein. Furthermore, the method 400 can be implemented in any suitable hardware,
software, firmware, or combination thereof.
[0093] With reference to the description of fig. 4, for the sake of brevity, the details of the
components of the various devices, such as the sensor-location design system 102, for
achieving positioning of contamination sensors in the fluid distribution systems, are not
discussed here. Such details can be understood as provided in the description provided with
reference to figure 1.
28
[0094] Referring to fig. 4, at block 402, a plurality of objective parameters affecting
location of positioning a plurality of contamination sensors in a fluid distribution system. The
objective parameters can in turn be identified based on one or more design objectives based
on which the sensor-location design is to be achieved. In one implementation, the objective
definition module 116 can identify expected volume of contaminated fluid consumed before
detection of contamination and a contamination detection possibility as the design objectives,
based on which the objective parameters are identified. Accordingly, in said implementation,
based on the design objectives, a volume objective parameter, and a detection-possibility
objective parameter are identified as the objective parameters, the volume objective parameter
indicative of a normalized value of decrease in the total contaminated fluid consumed,
represented by a fraction of a decrease in the total volume of contaminated fluid consumed
due to placement of one sensor in the fluid distribution system with respect to the maximum
possible contaminated volume that can be consumed in the fluid distribution system. On the
other hand, the detection-possibility objective parameter is indicative of a marginal increase in
the contamination detection possibility due to placement of one sensor in the fluid distribution
system, and can be represented by a fraction of the total number of detections achieved to a
total number of scenarios taken into account for studying the impact of placement of sensors
in the fluid distribution system.
[0095] Further, at block 404, based on the objective parameters, an objective function is
defined. In one example, the objective function is defined by obtaining a sum of the volume
objective parameter and the detection-possibility objective parameter. Further, a weighing
factor may also be taken into account while defining the objective function, to provide a
trade-off between the two parameters in the objective function.
[0096] At block 406, flow properties of the fluid circulating in the fluid distribution
system are determined. In an implementation, the flow properties are determined to take into
account a time of introduction of the contaminant into the fluid distribution system for
capturing the dynamics of transport of the contaminant in the fluid distribution system. In an
implementation, in order to take into account the time of introduction of the contaminant, the
determination of the impact of positioning of the contamination sensors can be commenced
substantially at the instance when a change in demand pattern occurs in the fluid distribution
system. As will be understood, the instance at which the demand pattern changes is indicative
29
of dispersal of the contaminant in the fluid distribution system, and therefore, of the time of
introduction of contaminant into the fluid distribution system.
[0097] Accordingly, a simulation of the fluid distribution system can be obtained, the
simulation including the layout of the fluid distribution system and also providing information
regarding the fluid circulating in the fluid distribution system. In an example, the flow
properties can include flow values for fluid in the fluid distribution system, and can be
determined based on the simulation of the fluid distribution system. In another example, the
flow properties can be obtained from an actual fluid distribution system for which the sensorlocation
design is to be achieved, from various flow sensors deployed in that fluid distribution
system. Further, in one example, based on the simulation, the time instances, at which the
demand pattern at one or more node changes, can be determined.
[0098] Further, for the purpose of simplifying the process, the time periods during which
the contamination has a fixed concentration are identified. In an example, such time periods
can be determined based on duration of operation of the fluid distribution system and demand
patterns at various nodes of the fluid distribution system. Accordingly, in an implementation,
a time period, referred to as epoch, between two successive time instances at which change of
demand pattern occurs. Hence, an epoch can be understood as a time period during which the
demand pattern and the flow pattern in the fluid distribution system is static.
[0099] Further, to take into account the effect of the transportation and dispersion of the
contaminant in the fluid distribution system, the flow properties of the fluid in the fluid
distribution system can be determined for each epoch during which the flow pattern in the
fluid distribution system remains static.
[0100] Further, the total duration of operation of the fluid distribution system can be
divided into a plurality of epochs, and the flow properties of the fluid can be determined for
each epoch. Considering the transportation properties and dispersion of the contamination in
the fluid distribution system, based on the time of introduction of the contaminant, allows for
substantially less expenditure of computational resources to be used for sensor-location
design. At the same time, the accuracy and the effectiveness of contamination detection, by
positioning the contamination sensors at the determined locations, is not compromised.
[0101] Further, at block 408, a scenario impact mapping is generated, and each scenario
in the scenario impact mapping can depict a combination of a potential location from a
30
plurality of potential locations of a contamination sensor in the fluid distribution system with
each node of the fluid distribution system. In one example, the scenario impact mapping can
be generated in the form of a matrix of the order M(i, j), each i in the matrix depicting one
node of the fluid distribution system, and each j depicting the potential location of a
contamination sensor in the fluid distribution system.
[0102] In an implementation, the scenario impact mapping is generated to determine the
impact of potentially placing the contamination sensors in the fluid distribution system. In an
example, the potential locations for positioning the contamination sensors can be determined
based on fixed locations for positioning the contamination sensors in the fluid distribution
system, say the nodes of the fluid distribution system. Such potential locations can be
previously defined, say by a designer of the layout of the fluid distribution system.
[0103] In an implementation, in order to generate the scenario impact mapping, one or
more impact parameters can be determined for each scenario in the scenario impact mapping,
the impact parameters being determined based on the flow properties of the fluid. In an
implementation, the impact parameters can include one or more of a volume impact
parameter, a time impact parameter, and a detection impact parameter. In an example, the
volume impact parameter is represented as the total volume of contaminated fluid consumed
in the entire network, when for the given scenario, the contamination occurs at the node
included in that scenario. Further, the time impact parameter can be represented as a time
taken for detection of contamination by the contamination sensor for that scenario, and the
detection impact parameter can be represented as whether the detection has occurred or not.
[0104] In an implementation, the impact parameters can be determined for all the
scenarios in the scenario impact mapping and the values can be populated in the scenario
impact mapping. As will be understood, since the flow properties for the fluid are determined
for the individual epochs that the entire operational duration of the fluid distribution system is
divided into, the impact parameters can also be determined for each scenario, for each epoch
in the entire operational duration.
[0105] Further, at block 410, a final location of one of the plurality of contamination
sensors is determined, based on the scenario impact mapping and the objective function. In an
implementation, for determining the final location of one sensor, an overall impact parameter
is for one or more of the impact parameters can be determined. The overall impact parameter
31
can be understood as the value of the impact parameter for the entire duration of operation of
the fluid distribution system and for all the scenarios considered for that sensor in the scenario
impact mapping.
[0106] In one example, an overall volume impact parameter and an overall detection
impact parameter can be determined and considered for determining the final location of the
contamination sensor. In said example, the overall volume impact parameter can indicate the
total volume of contaminated fluid consumed in the fluid distribution system when the
contamination sensor is positioned at one potential location and the contamination occurs at
all the nodes, whereas the overall detection impact parameter can indicate the total number of
contamination events that can be detected by positioning the contamination sensor at that
potential location, when contamination occurs at all the nodes of the fluid distribution system.
[0107] Further, for each location in the various scenarios in the scenario impact mapping
for the contamination sensor, the value of the objective function is determined. In said
implementation, that location from among the plurality of locations, for which a predefined
optimal value of the objective function is achieved is determined and selected as the final
location of that contamination sensor. In one example, the final location for the contamination
sensor is determined on the basis of the maximization of the objective function, i.e.,
maximization of the cumulative value of the objective parameters.
[0108] Further, the impact of the final location of the contamination sensor is taken into
account for determining the final location of the rest of the contamination sensors.
Accordingly, at block 412, an updated impact parameter for other contamination sensors
positioned in respective potential locations, with one contamination sensor positioned at the
final location is determined. In one example, an updated time impact parameter for all the
other contamination sensors at their respective potential locations are determined, the
potential locations being based on various scenarios in the scenario impact mapping.
[0109] Additionally, according to said implementation, at block 414, based on the updated
time impact parameter for all potential locations of the contamination sensors, another
updated impact parameter included in the scenario impact mapping is determined for each
potential location of the rest of the contamination sensors. In an implementation, based on the
updated time impact parameter, an updated volume impact parameter is determined.
32
[0110] Further, at block 416, the updated time impact parameters and the updated volume
impact parameters can be included in the scenario impact mapping to obtain an updated
scenario impact mapping. The updated scenario mapping can be indicative of the impact of
positioning the contamination sensor at the final location on the performance of the other
contamination sensor at their potential locations, and for determining the final locations of the
other contamination sensors.
[0111] Further, at block 418, it is determined whether the final locations for position of all
the available contamination sensors have been determined or not. In case the final locations of
all the contamination sensors have been identified (‘Yes” path from block 418), then at block
420 a sensor-location design can be provided for the fluid distribution system. In an example,
the sensor-location design can be based on the simulated layout of the fluid distribution
system and can depict the final locations of the various contamination sensors in the fluid
distribution system.
[0112] On the other hand, in case the final location for all the contamination sensors have
not been identified (‘No’ path from block 418), then based on the updated scenario impact
mapping, the final location of a subsequent contamination sensor can be determined in the
same manner above. Further, the method 400 can be iteratively executed until the respective
final locations of all the contamination sensors have been determined.
[0113] Although implementations for contamination sensor positioning in fluid
distribution systems have been described in a language specific to structural features and/or
methods, it is to be understood that the present subject matter (and not appended claims) is not
necessarily limited to the specific features or methods described. Rather, the specific features
and methods as described herein are disclosed as implementations of the present invention.
33
I/we claim:
1. A computer implemented method for contamination sensor positioning in a fluid
distribution system, the method comprising:
defining an objective function influencing positioning of a plurality of
contamination sensors in the fluid distribution system;
ascertaining at least one impact parameter, based on flow properties of a fluid, the
at least one impact parameter being indicative of an impact of positioning at least one
contamination sensor from the plurality of contamination sensors at a predetermined
potential location in the fluid distribution system, and wherein the impact parameter
takes into account a time of introduction of a contaminant into the fluid distribution
system; and
determining a final location for positioning each contamination sensor from among
the plurality of contamination sensors, based on the objective function and the at least
one impact parameter.
2. The method as claimed in claim 1, wherein the defining the objective function is based
on an expected volume of contaminated fluid consumed before detection of
contamination in the fluid distribution system by the at least one contamination sensor
and a contamination detection possibility.
3. The method as claimed in claim 1, wherein the ascertaining comprises computing the
impact parameter for each combination of the respective predetermined potential
locations of each of the plurality of contamination sensors with each of a plurality of
contamination-occurrence nodes in the fluid distribution system.
4. The method as claimed in claim 1, wherein the impact parameter comprises at least one
of a volume impact parameter indicative of a total volume of contaminated fluid
expected to be consumed in the fluid distribution system, a time impact parameter
indicative of time taken for detection of contamination in the fluid distribution system,
and a detection impact parameter indicative of whether the contamination is detected or
not.
5. The method as claimed in claim 1, wherein the ascertaining comprises:
identifying an epoch for the fluid distribution system, wherein a flow pattern of
fluid in the fluid distribution system is substantially static in the identified epoch; and
34
determining the at least one impact parameter for the identified epoch.
6. The method as claimed in claim 1, wherein the determining comprises ascertaining
updated values of the at least one impact parameter, based on the final location
determined for one contamination sensor from among the plurality of contamination
sensors, wherein the final location of each subsequent contamination sensor is based on
the updated values of the at least one impact parameter.
7. A sensor-location design system (102) for contamination sensor positioning in a fluid
distribution system, the sensor-location design system (102) comprising:
a processor (108);
an objective definition module (116) coupled to the processor (108), the objective
definition module (116) configured to define an objective function, wherein positioning
of a plurality of contamination sensor in the fluid distribution system is achieved based
on the objective function;
an impact determination module (120) coupled to the processor (108), the impact
determination module (120) configured to determine an impact parameter indicative of
an effect of positioning one or more of the plurality of contamination sensors in the fluid
distribution system, wherein the impact determination module (120) is further
configured to determine the impact parameter based on a time of introduction of a
contaminant in the fluid distribution system; and
a location determination module (122) coupled to the processor (108), the location
determination module (122) configured to determine a final location for positioning
each of the plurality of contamination sensors, based on the impact parameter and the
objective function.
8. The sensor-location design system (102) as claimed in claim 7, wherein the impact
determination module (120) is configured to compute the impact parameter for each
combination of the respective predetermined potential locations of each of the plurality
of contamination sensors with each of a plurality of contamination-occurrence nodes in
the fluid distribution system.
9. The sensor-location design system (102) as claimed in claim 7, wherein the location
determination module (122) is further configured to ascertain the final location for
positioning one contamination sensor from among the plurality of contamination
35
sensors, in the fluid distribution system, based on an optimized value of the objective
function for a potential location for positioning the contamination sensor.
10. The sensor-location design system (102) as claimed in claim 9, wherein the impact
determination module (120) is further configured to determine an updated impact
parameter based on the final location of the contamination sensor, wherein the final
location of each subsequent contamination sensor is based on the updated impact
parameter.
11. The sensor-location design system (102) as claimed in claim 7, further comprising a
flow determination module (118) coupled to the processor (108), the flow determination
module (118) configured to:
identify an epoch for the fluid distribution system, wherein the epoch represents a
time period of substantially static flow pattern of the fluid in the fluid distribution
system; and
determine flow properties of a fluid circulating in the fluid distribution system for
the epoch, wherein the flow properties are indicative of the time of introduction of the
contaminant.
12. The sensor-location design system (102) as claimed in claim 11, wherein the flow
determination module (118) is further configured divide a total duration of operation of
the fluid distribution system into a plurality of epochs.
13. The sensor-location design system (102) as claimed in claim 11, wherein the impact
determination module (120) is configured to determine the impact parameter for at least
one epoch from the plurality of epochs.
14. A computer-readable medium having embodied thereon a computer program for
executing a method for contamination sensor positioning in a fluid distribution system,
the method comprising:
defining an objective function affecting positioning of a plurality of contamination
sensors in the fluid distribution system;
determining flow properties of a fluid circulating in the fluid distribution;
ascertaining at least one impact parameter, based on flow properties of the fluid,
the at least one impact parameter being indicative of an impact of positioning one or
more of the plurality of contamination sensors at respective predetermined potential
36
locations in the fluid distribution system, and wherein the impact parameter takes into
account a time of introduction of a contaminant into the fluid distribution system; and
determining a final location for positioning each contamination sensor from
among the plurality of contamination sensors, based on the objective function and the
at least one impact parameer.