Customised wireless sensor node to detect hazardous gas pipeline leakage


Published on

For more projects visit @

Published in: Technology, Business
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Customised wireless sensor node to detect hazardous gas pipeline leakage

  1. 1. PIC18LF4620 Based Customizable Wireless SensorNode to Detect Hazardous Gas Pipeline LeakageS. Edward Jero1, A. Balaji Ganesh2Opto-Electronic Sensor Research Laboratory,Velammal Engineering College, Chennai -66, INDIAedwardjero@gmail.com1, abganesh@velammal.org2Abstract— The paper describes the performance andfunctional characteristics of PIC18LF4620 based wireless sensornode in monitoring the parameters such as CO2, Oxygen,temperature, humidity and light around the pipeline structure.The system is deployed to monitor any deviations in theseparameters with the standard atmospheric values eventually alertthe user even to a remote location. The proposed system is abattery operated wireless sensor node which is interfaced withthe external sensors to measure the parameters listed above. Thedistance range between sensor node and coordinator node is alsotested. The signal conditioning module associated with detailedcalibration procedure for the individual sensor is also described.Zigbee protocol stack is implemented to enable wirelesstransmission and performance of the same is evaluated.Keywords—PIC18LF4620, Wireless Sensor Node, Gas PipelineLeakage Detection, Zigbee moduleI. INTRODUCTIONObviously, the atmospheric air is composed of variousgases including nitrogen, methane, oxygen and carbon dioxideetc in different percentage [1]. Industries are using gases likeCO2, Oxygen, Nitrogen, Methane etc and commercializing itbased on the requirements [2]. Gas pipelines take part of vitalrole in transmission and distribution of gases in industrial anddomestic purposes [3]. Leakage of hazardous gas pipelinesmodifies the percentage of gas contents in the atmospheric airand causes undesirable effects in the environment. Theproposed system used to monitor the concentration of CO2 andthe percentage of oxygen, humidity and temperature in theatmosphere in the randomly selected locations in the pipeline.The proposed system is a PIC18LF4620 based wireless sensornode and it has the features of continuous monitoring ofspecified parameters with easy deployment procedures andincrease the battery lifetime. Wireless connectivity is obtainedby MRF24J40 zigbee module which enables data transmissionof the system to the remote location [4].II. PROPOSED SYSTEM OF COMPONENTSThe proposed system consists of,Sensor module consists of sensor for each parameter andits respective power circuitry.Output voltages of the procured sensors are not found to beinsufficient to have a better interpretation with Analog-to-Digital Converter (ADC) and micro computer. Signalconditioning module amplifies the sensor output and the sameis fed to ADC [5].The chosen PIC18LF4620 is a low voltage, nano wattMicro controller unit. The microcontroller is interfaced withsensor modules, communication module. The requiredcomputations for the analog signals received from sensors,data transmission are done within micro controller unit.Wireless communication module consists of 2.4GHz IEEE802.15.4 based MRF24J40 RF Transceiver zigbee unit whichis used to establish the wireless connectivity of the system.The 9V Ni-MH rechargeable battery is used to power upthe system [6]. In the mere future, a solar battery charging unitwill be adapted to charge the battery. Voltage regulation isdesigned to provide the rated input voltage supply for eachmodule.The block diagram representation of the proposed systemis shown in the Figure 1.Figure 1. Block Diagram Representation of Proposed SystemIII. SENSOR NODE – HARDWARE COMPONENTSDESCRIPTIONThe sensor node consists of various sensors to monitor thevalues of CO2 concentration, percentage of oxygen,temperature and relative humidity. The sensors to be embeddedwith the system are listed in Table 1TABLE I. SENSOR COMPONENTS DESCRIPTIONSensor Type & Its Details MeasurandTGS4161 (M/s Figaro, USA) CO2KE-50 (M/s Figaro, USA) OxygenChipCap-L (M/s GE Sensing, USA)Temperature andHumidityPROCEEDINGS OF ICETECT 2011978-1-4244-7926-9/11/$26.00 ©2011 IEEE 563
  2. 2. Each sensor outputs are amplified by the gain amplifier [7].The maximum gain of an amplifier is selected to obtain thereference voltage of the analog to digital converter (ADC). Thefilter circuit is added with the output of the gain amplifier toreduce the noise level of the amplified output. The filteredanalog signals are given as inputs to the 10Bit ADC modules ofthe micro controller. The block diagram of signal conditioningcircuit is shown in Figure 2.Figure 2. Block diagram of sensor modulePIC18LF4620 Low power micro controller is contributedfor computations, conversion and data transmission of theproposed system. The input power supply required to operatethe micro computer is 3.3V. The current drawn by the microcontroller is 11µA at run mode. Thirteen number of 10-bitADC modules support the extension of the existing design;SPI module and ZigBee compatibility are the key featureswhich are used by the proposed system. The 10bit ADC of themicro controller converts the analog input signal from thesensor module into digital value within the range between 0and 1024. Computations are used for the conversion of digitalvalues to its respective engineering units. The block diagramof microcontroller unit is shown in the Figure 3.The PICDEMZ Evaluation board is used for designverification of the proposed sensor node in programming andfunctionality. The Sensor modules are connected with the fullfunction device node. The coordinator node is connected withthe computer for monitoring the measured parameters from thesensor node. The ICD3 debugger provides hardware breakpoints and supports hardware debugging while the execution ofprogram.Figure 3. Block diagram of microcontroller unitMRF24J40MA zigbee module used to establish thewireless connectivity of the designed system. Serial-ParallelInterface (SPI) communication protocol supports the interfacebetween zigbee module and micro controller unit. Theinformation required to be transmitted are loaded in the dataframe and the payload is placed in the transmission buffer andready to transmit. Wireless communication takes place whenthe requested node is founded as idle. The block diagram ofinterfacing zigbee module with micro controller is shown inFigure 4.Figure 4. Wireless Communication moduleIV. SOFTWARE PROCESS MODEL OF SENSORNODEThe MPLAB IDE with PIC C18 compiler is used tocompose the programming part of the proposed system. Thecomputed values of each sensor output parameters inPIC18LF4620 are obtained by the program written inEmbedded C using PIC C18 compiler and loaded in the datafields D0, D1, D2 and D3 of the data frame. Data field D4 isloaded with the cyclic redundancy check value (CRC) which iscalculated for all the computed sensor data. The transmissionbuffer of the zigbee stack is loaded with that data frame andready to transfer the data on request as shown in the blockdiagram of Figure 5. The CRC value is used to verify thetransmitted data of the sensor node at the receiving end.Figure 5. Block diagram of software architecture564
  3. 3. V. RESULTS AND DISCUSSIONSA four pin carbon dioxide sensor, which contains a heaterwith 70 ohm resistance. This sensor requires 5V voltage to theheater circuit and the response time of the sensor is calculatedapproximately 90 seconds. The EMF generated by the sensoris inversely proportional to the concentration of CO2 andmeasured through a high impedance operational amplifierTLC271 [8]. The proposed system is designed to measure theconcentration of CO2 in the range between 350PPM and3500PPM. EMF generated at 350 PPM of CO2 is one of theranges between 220 mV and 490 mV. The EMF generated at3500 PPM of CO2 by the range between 44 mV and 72 mV.The test environment for CO2 sensor to observe the variationin the concentration is created as the sensor is enclosed by thefiber box which contains an air vent; an exhalation throughthat air vent increases the CO2 concentration inside theenclosed area of the box result in changes in EMF.Figaro KE-50 Oxygen sensor provides a linear voltageoutput signal for the rate of oxygen present in the atmosphere.Test setup is obtained by the percentage of oxygen present inthe plastic pot is reduced by exhalation and the sensor isclosed by that pot. The test procedure is found that the outputof the sensor at 21% of oxygen is within the range between 47mV and 65 mV. The maximum output of the sensor is 260 mVat the maximum (100%) of oxygen. Accordingly the oxygenvalue is calculated from the linear relationship with oxygen ofthe sensor’s output. After each testing, oxygen sensor output isbring back to the approximately 20% of oxygen by kept it inthe normal atmospheric air and then the next test is conducted.The Humidity and Temperature sensor ChipCap-L providesthe linear output for temperature and humidity of theatmosphere in the range between 0 and 1 volt [9]. All thecalibrated sensors are cross checked with the same typereference instruments.Several test case scenerios are performed to validate thefunctional charatestics of developed sensor node. Tests areconducted in both laboratory and outdoor environments. Theresults are obtained and also the parameters are tested bykeeping the sensor node at a distance of 100 fts from thecoordinator node without any dataloss. The results obtainedare plotted and are shown in figures 6 and 7. The coordinatornode comprises with a computer and a wireless access unit.The database is maintaind in computer for the offline futureanalysis. The hardwares which are selected to build thissystem to obtain higher performance with minimum hardwarecomponents and requirement of minimal cost. The proposedsensor node has the features of low power consumption anddeal with minimal number of components with larger distancecovarage. In the mere future, it is planned to construct acustomizable sensor node with the advent of same identifiedhardware components such as microcontroller, signalconditioning circuits and communication module.Figure 6. Carbon-di-Oxide Sensor OutputFigure 7. Sensors (O2, Temperature & Humidity) OutputVI. CONCLUSIONSThe functionaliites and performance of PIC18LF4620 basedwireless sensor node has been evalutead by interfacing thesame with various sensors. The sensor node is successfullytested for the distance of 100 ft witout any dataloss. Theindividiual sensors such as Carbon-di-Oxide, Oxygen,Temperature and Humidity are calibrated and deployed for theestimation of hazarourdous gas pipe line leakage. The signalconditioning circuit for the individual sensor has beendesigned eventually tested. A ten byte of data frame is formedand its data fields are allocated to each measured parameter.ACKNOWLEDGMENTThe authors gratefully acknowledge the financial support ofTIFAC-DST through the centre ‘Pervasive ComputingTechnologies’ to Velammal Engineering College, Chennai.REFERENCES[1] Atmospheric Composition;[2] Industrial Application Gases;[3] Supply of industrial gases;[4] Node Architecture ;[5] Peng Jiang, Hongbo Xia, Zhiye He and Zheming Wang Article: Designof a Water Environment Monitoring System Based on Wireless SensorNetworks Sensors 2009, 9(8), 6411-6434;[6] Barrenetxea, G.; Ingelrest, F.; Schaefer, G.; Vetterli, M.; Couach, O.;Parlange, M. ; SensorScope: Out-of-the-Box Environmental Monitoring, International Conference on Information Processing in SensorNetworks, 2008. IPSN 2008. 332 – 343565
  4. 4. [7] Low power operational amplifier;[8] Carbon dioxide;[9] Humidity and Temperature;[10] Atmospheric Composition;[11] Stoianov, I.; Nachman, L.; Madden, S.; Tokmouline, T.; Csail, M.;PIPENET: A Wireless Sensor Network for Pipeline Monitoring, 6thInternational Symposium on Information Processing in SensorNetworks, 2007. IPSN 2007., 264 – 273.[12] Volgyesi, P.; Nadas, A.; Koutsoukos, X.; Ledeczi, A.; Air QualityMonitoring with SensorMap, International Conference on InformationProcessing in Sensor Networks, 2008. IPSN 08.,529 – 530[13] Murty, R.N.; Mainland, G.; Rose, I.; Chowdhury, A.R.; Gosain, A.;Bers, J.; Welsh, M.; CitySense: An UrbanScale Wireless SensorNetwork and Testbed, IEEE Conference on Technologies for HomelandSecurity 2008.,583 – 588[14] S. Koushanfar, F. Kosterev, A. Tittel, F. Rice Univ., HoustonLaserSPECks: Laser SPECtroscopic TraceGas Sensor Networks SensorIntegration and Applications, International Conference on InformationProcessing in Sensor Networks, 2007.IPSN 2007., 226 – 235[15] Barrenetxea, G.; Ingelrest, F.; Schaefer, G.; Vetterli, M.; Couach, O.;Parlange, M. ; SensorScope: Out-of-the-Box Environmental Monitoring, International Conference on Information Processing in SensorNetworks, 2008. IPSN 2008. 332 – 343[16] SWATS: Wireless Sensor Networks for Steamflood and WaterfloodPipeline Monitoring[17] Merrett, G.V.; Weddell, A.S.; Harris, N.R.; Al-Hashimi, B.M.; White,N.M.; A Structured Hardware/Software Architecture for EmbeddedSensor Nodes, 17th International Conference on ComputerCommunications and Networks, 2008. ICCCN 08. 1 – 6[18] Node Architecture ;[19] Sukwon Choi, Nakyoung Kim, Hojung Cha and Rhan Ha ; Micro SensorNode for Air Pollutant Monitoring: Hardware and Software IssuesSensors 2009, 9(10), 7970-7987; doi:10.3390/s91007970[20] Peng Jiang, Hongbo Xia, Zhiye He and Zheming Wang ; Design of aWater Environment Monitoring System Based on Wireless SensorNetworks Sensors 2009, 9(8), 6411-6434; doi:10.3390/s90806411[21] MTS/MDA Sensor Board Users Manual;[22] Waspmote ; Feng Xia ;Wireless Sensor Technologies and Applications Sensors 2009, 9(11),8824-8830; doi:10.3390/s91108824566