INTERNATIONAL JOURNAL OF ENVIRONEMNTAL SCIENCES Volume 5, No 2, 2014 © Copyright by the authors - Licensee IPA- Under Creative Commons license 3.0
Review article
ISSN 0976 – 4399
Development of inexpensive multi-parameter sensors based network system for water environment monitoring Kushantha Lakesh S.H.P. 1, Dahanayaka D.D.G.L. 1, Anne Nisha G. 1, Nalin Warnajith1, Hideyuki Tonooka2, Atsushi Minato1, Satoru Ozawa1 1- Graduate School of Science & Engineering, Ibaraki University, Japan 2- Department of Computer and Information Sciences, Ibaraki University, Japan
[email protected] doi:10.6088/ijes.2014050100030 ABSTRACT In the conservation of water environments, it is necessary to monitor the physico-chemical parameters, such as color of water, spectroscopic data of water surface, CO2 density, temperature and pressure around the shore. However, the designing of a flexible multi parameter sensing system using a computer is generally considered expensive. In addition, programming and circuitry of a multi-parameter sensing system is complex. A computer also requires electrical power. In this paper, we propose a new technique to modularize sensors. Applying this technique, a multi-parameter sensing system can be constructed with relative ease, and a controlling program of computer and an electrical circuit can be simplified. Using a small computer, a prototype measurement system was constructed and fundamental measurement was carried out. The power consumption and data acquisition process are also discussed. Keywords: CO2 sensor, small computer, spectrometer, water environment, water color. 1. Introduction Monitoring of water environment parameters such as color of water, spectroscopic data of water surface and CO2 density around the shore is critical for the conservation and sustainable utilization of water bodies such as estuaries, lagoons and lakes. Previously, we have proposed water environment monitoring methodologies using in-situ and satellite data in Sri Lankan lagoons and Japanese inland lakes (Dahanayaka et al., 2012, 2013). During those studies, we used commercially available instruments for water quality measurements, such as KRK CHL-30 handheld Chlorophyll meter for Chlorophyll -a monitoring and BSR112E spectrometer produced by B&W Tek Inc., USA, for water reflectance measurements etc. However, because the costs of these instruments were great for our limited budget and only one set of them was used in the studies, it was difficult for us to conduct ideal in-situ measurements for water quality monitoring. In order to understand water body productivity further, it is necessary to conduct simultaneous measurements over large sampling areas of the water body (Lovett et al., 2007; Vaughan et al., 2001). Therefore the development of a low-cost portable multiple-sensor system is necessary for the use in large scale, concurrent monitoring programs. As most of our present study sites are located in developing countries in the Asian region, such as Sri Lanka, Vietnam, Thailand, Philippines and Indonesia, the development of a low-cost multiple-sensor system for the continuous monitoring of water environments will be distinctly advantageous. Application of sensor technology is a solution for this issue and it provides biologists with the means of acquiring these synoptic data and also offers a cost-effective tool for complementing regular monitoring
Received on August 2014 Published on September 2014
341
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
programmes (Torrent, 2004, Yick et al., 2008; OECD, 2009; Postolache et al., 2011; Sahota et al., 2011, Lakesh et al., 2013; Postolache et al., 2013; Satyanarayana and Mazaruddine, 2013). Currently various sensors are commercially available, which can measure multiple physical properties including temperature, CO2 density, humidity, air pressure (Light, 1983; Tans, 2002; Pereira et al., 2009; Postolache et al., 2009; Wang et al., 2009; Thammita et al., 2013). The aim of this work is to present a new concept for developing an easy-to-build, low-cost system for in-situ monitoring of water environments. The system combines various sensors which are modularized. It’s controlled by a computer and selects only a sensor module and gets data. 2. Concept of System Design Sensors are generally connected to a computer or a microcontroller. There are various interfaces between a sensor and a computer. They are anolog to digital conversion, serial communication, Serial Peripheral Interface (SPI), Inter-IC (I2C) bus interface and serial communication interface (RS-232) (Zhou and Mason, 2002). It is confusing for a computer to communicate with some sensors with different interfaces. Especially as circuit design and controlling program become complicated. Our novel technique is to modularize each sensor. Each sensor module has a sensor, small microcontroller and USB serial interface. The microcontroller plays the part of interface between the sensor and the computer. The sensor modules are connected to the controlling computer via a USB port. The controlling computer selects a communication port and sends request commands such as data acquisition. One microcontroller receives a specific signal and aquires measurement data from the sensor. The microcontroller sends data to the controlling computer as serial data. To reduce power consumption, a small computer is preferable, compared to standard Personal Computer (PC). There are various types of small computers, such as Raspberry Pi, BeagleBone, and Intel Galileo. These computers consume less energy. They are also less expensive than standard PCs. They can be used in various Operating Systems (OS) such as Linux, FreeBSD, RISC OS, etc. For this experiment, the small and inexpensive Raspberry Pi was used as the controlling computer. Sensors were modularized and connected to the computer. The Raspberry Pi is a single board, computer, 5cm x 4cm in size (Upton and Halfacree, 2010). It runs on Linux distributions. Rasbian OS(Raspberry Pi,2014b) is used in this development. Raspberry Pi is available with relatively low price (around US$35). The Raspberry Pi (Model B) has an Ethernet port which allows to connect to Internet and also acts as a data server. While this model needs 5 V power supply, the Model A without any Ethernet port can work with 2.5 W power consumption. In this study, we used the Model A in order to reduce power consumption (Membrey and Hows, 2013). A brief overview of the complete system is shown in Figure 1. Three sensor modules and a camera are controlled by the computer: (1) Mini spectrometer (10988MA) to aquire water spectrum, (2) CO2 sensor (K30 engine), (3) Pressure and temperature sensor (SCP1000), and (4) Camera, which is directly controlled by Raspberry Pi. (1) Mini spectrometer (10988MA): The Hamamatsu (C10988MA) MS series sensor was used for the system. The internal optical system is comprised of a convex lens on which a grating is formed by Nano-imprint. In addition, it has a spectral range of 340 to 750 nm, 14 nm resolution, 5 V supply voltage and a low power consumption of 30 mW (Hamamatsu. 2014). The output of MS series is a video signal. Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
342
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
Figure 1: The system architecture (2) CO2 sensor (K30 engine): The K30 engine is used for CO2 monitoring. K30 engine uses NDIR (Non dispersive type inferred rays abortion) and measures range 0-10,000 ppm. It has a UART serial interface, I2C interface and analog output (Sensor air, 2014). The UART serial interface was used for our system. (3) Pressure and Temperature Sensor Module (SCP1000): SCP1000 is an atmospheric pressure and temperature sensor. The sensor consists of a silicon bulk micro machined sensing element chip. Two different interfaces are available: SPI and TWI. The pressure output word-length is 19 bits and the temperature output word-length is 14 bits. Supply voltage is 2.4 to 3.3 V, and the measurement range is between 30 and 120 kPa, and the resolution is 1.5 Pa (Sparkfun, 2014). The SPI interface was used for our system. (4) Raspberry Pi Camera: The Raspberry Pi Camera module is a custom designed add-on, which can be attached to the Raspberry Pi. This interface uses the dedicated CSI interface, which was designed especially for interfacing to a camera. The camera has a native resolution of 5 megapixels, and has a fixed focus lens onboard. The camera is capable of 2592 x 1944 pixel static images, and also Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
343
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
supports video formats (Raspberrypi, 2014a). The camera is directly controlled by Raspberry Pi and captures image data. The above three sensor modules have R8C/M11A or M12A microcontrollers and USB serial module using FT232. R8C microcontroller is 16-bit CISC microcontroller of the R8C Renesas family platform (Renesas, 2014a; Renesas, 2014b). With 1 Mbyte of address space, the CPU core is capable of executing the instruction at high speed with 20 MHz, and consists of 2 Kb EEPROM, AD converter, and UART serial interface. A programme was prepared to support SPI communication on R8C microcontrollers. The data streaming rate of UART interface between the microcontroller and the USB serial interface was set to 9600 bps. When a microcontroller receives a specific command, such as‘s’ acquire data from USB serial interface, it downloads data from one of the sensors. The microcontroller returns data through the USB serial interface. The power for the microcontroller and the sensor is also supplied from the USB serial interface.
Figure 2: System flow of python program. Linux-based Python program is used to control the sensor modules. Pi-serial extension was added to the Raspberry Pi to enable serial communication through USB serial interface. An example of the measurement flow using the python program is shown in Figure 2. First, an image is captured. Then the computer selects the communication port of the CO2 sensor, and request data by sending the serial command‘s’, so that the microcontroller for the CO2 sensor returns CO2 density data. Next, the computer selects the communication port of spectrometer, and requests data by sending the serial command‘s’, so that the microcontroller for the spectrometer returns data. Finally, the computer selects the communication port of pressure and temperature sensor, and sends the command ‘p’ or‘t’ for acquiring pressure or temperature data, respectively, from the microcontroller. The received data are stored in an Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
344
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
SD card of the computer. The Python program can be write in short lines and it is easy to add or remove sensor modules.
Figure 3: The system setup: (1) Small Computer Module, (2) Pi Camera Module, (3) Spectrometer Module, (4) CO2 Module, and (5) Pressure and Temperature Module.
Figure 4: System testing in Lake Senba 3. Experiment and Result: Three field tests were conducted from May 2014 to August 2014: Lake Senba (lat 36.368655, long 140.461436) at 11:00 am on 1 June 2014, Kujigawa River (lat 36.482971, long 140.604311) at 11:40 am on 28 May 2014, and Ayukawa River (lat 36.57092, long 140.651736) at 11:30 am on 23 May 2014 (all the sites are located in Ibaraki Pref., Japan). During the period of each experiment, image, water spectrum, CO2, temperature and pressure data were collected with a constant time interval. Figures 3 and 4 show photos of the original system setup in our laboratory and the field testing at Lake Senba, respectively. Figures 5, 6 and 7 show the water spectral values obtained from 10:00 am to 2:00 pm at Lake Senba, Kujigawa River, and Ayukawa River, respectively, which were used for getting the maximum reflection. The spectrum range from 678 to 705 nm is useful for estimating
Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
345
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
concentration of Chl-a which has a distinct absorption in the red and near infrared wavelength ranges (Gitelson, 1991; Kirk, 1994).
Figure 5: Mean of the spectral intensity measured by the developed system at Lake Senba.
Figure 6: Mean of the spectral intensity measured by the developed system at Kujigawa River.
Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
346
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
Figure 7: Mean of the spectral intensity measured by the developed system at Ayukawa River. The reflectance spectra by a commercial spectrometer (BSR112E) and those acquired by the system developed by ourselves measured in the same bodies of water, are generally similar in shape. Figure 8 shows an example of the water reflectance spectra measured by the BSR112E Spectrometer and Figure 9 shows an example of the developed system at Senba Lake on 6 June 2014. The both spectra indicate the presence of algal chlorophyll in the water column by the following features: (1) low reflectivity between 400 and 500 nm due to absorption of blue light, (2) maximum green reflectivity between 500 and 600 nm, (3) a minor inflection at about 640 nm presumably due to either backscattering from dissolved organic substances (Gitelson and Kondratyev, 1992) or accessory pigments, (4) classic red absorption near 678 nm, (5) prominent NIR reflectivity at about 705 nm, and (6) a minor NIR reflectance feature at about 810 nm probably caused by backscattering from organic matter (e.g., algal cells). High reflectance between 500 to 600 nm can be explained by Chl-a absorption at these wavelengths coupled with high backscattering caused by algal cells (Dekker, 1993). The prominent feature at (5) is explained as the fluorescence of phytoplankton pigments (Hoge and Swift, 1987), or anomalous scattering caused by absorption minimum at 675 to 680 nm (Morel and Prieur, 1977). At Ayukawa River, a long-hour filed experiment was also conducted from 8 am to 5 pm (eight hours) on 20 June and 15 July 2014 in order to investigate the diurnal changes of the water environment, CO2, temperature, and atmospheric pressure. Since pressure and temperature show larger values than the actual level due to the sun’s direct light, the system was placed as to avoid it. In addition, in order to measure both temperature and pressure data from 4 am to 9 pm (the sunrise time in Japan is from 4 to 5 am in summer), we continued to run the system from the previous midnight to the evening of that day. Figures 7, 10, 11 and 12 show the obtained values of the above parameters.
Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
347
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
Figure 8: Spectral data recorded in Senba Lake on 6 June 2014 using the BSR112E Spectrometer.
Figure 9: Spectral intensity measured by the developed system at Senba Lake on 6 June 2014.
Figure 10: CO2 changes throughout the eight hour experimental period at Ayukawa River on 15 July 2014.
Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
348
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
Figure 11: Temperature changes throughout the eighteen hour experimental period at Ayukawa River on 5st of August 2014.
Figure 12: Pressure changes throughout the eighteen hour experimental period at Ayukawa River on 5th of August 2014. 4. Summary and Conclusions A new concept design for a multi-parameter sensor modules system is presented in this paper. A fundamental experimental system was developed and applied for water environment monitoring. Spectroscopic data of water surface, CO2 density, air temperature, pressure and water color were measured at various locations. A system design based on a sensor module method is generally easy to use―the controlling computer only selects one of the sensor module ports, sends commands, and receives data exclusively from the selected sensor. A system designer does not need to pay attention to the interface of sensors, such as UART serial interface, SPI or I2C. Using a small computer like Raspberry Pi, we can build a measurement system with low cost and low power consumption. For example, the total cost of our system was around 150 US$, and 12 V rechargeable battery (12 Ah) of our system Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
349
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
allows us to measure more than 66 hours continuously (the average power consumption estimated is 2.40 W). The Raspberry Pi, which we used for our system, has an Ethernet port, making it relatively easy to build a network sensing system, and enabling it to work as a data server in the network. Increasing the number of sensors and the ability of covering a wide area are considered as issues to address during actual application. A possible solution is to connect small computers with sensors by a wireless module such as Zig-Bee. Furthermore, our system provides the capability to communicate between the microcontroller and the USB serial interface through the UART serial interface. Therefore, if a wireless module is used, sensor modules can be separated from the computer, which allows the coverage of a wide area, using only a single computer. If a GPS sensor module is attached to the present system, it will make it easier to acquire information from a given time and location. When considering Chl-a measurements at water bodies in tropical countries such as Sri Lanka, one of the challenges is sudden rainfall, which could lead to a significant change in Chl-a. If a continuous monitoring system is in place, as proposed in the present study, such anomalies in data can be successfully monitored. Over the last few decades, water environments have been poorly monitored in Sri Lanka, due to a lack of technological and financial resources, as well as the relevant expertise (Dahanayaka et al. 2012). The present study can be expected to significantly contribute to such monitoring in future. 4. References 1.
Anuruddha, T. A. S., Ozawa, S., Minato, A., (2013), Development of GPS Assisted Online CO2-Temperature Mapping System, Journal of Emerging Technology and Advanced Engineering, 3(7), pp 22-25.
2.
Dahanayaka, D.D.G.L., (2012), Circumstances of Remote Sensing Applications and Educations in Sri Lanka, Journal of the Remote Sensing Society of Japan, 32(3), pp 160-163.
3.
Dahanayaka, D.D.G.L., Tonooka, H., Wijeyaratne, M.J.S., Minato, A., and Ozawa, S., (2012), Monitoring Land Use Changes and their Impacts on the Productivity of Negombo Estuary, Sri Lanka Using Time Series Satellite Data, Asian Fisheries Science, 25, pp 97-212.
4.
Dahanayaka, D.D.G.L., (2013), Environmental assessment of coastal and inland water bodies using satellite remote sensing data, PhD thesis, Ibaraki University, Japan, pp 177.
5.
Dahanayaka, D.D.G.L., Tonooka, H., Wijeyaratne, M.J.S., Minato, A., and Ozawa, S., (2013), Two decadal trends of surface chlorophyll-a concentrations in tropical lagoon environments in Sri Lanka using satellite and in-situ data, Asian Journal of Geo informatics, 13(3), pp 15-25.
6.
Dekker, A.G. (1993), Detection of optical water quality parameters for eutrophic watersby high resolution remote sensing, PhD Thesis, Institute of Earth Sciences, Vrije Universiteit, Amesterdam, The Nederlands, 222 pp.
7.
Gitelson, A. A., and Kondratyev, K.Y, (1991), Optical models of mesotrophic and eutrophic water bodies. International Journal of Remote Sensing, 12, pp 373–385.
Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
350
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
8.
Hamamatsu, (2014), Mini-spectrometer MS series, available at https://www.hamamatsu.com/ esources/pdf/ssd/c10988ma-01_etc_kacc1169e09.pdf, accesssed on 15-04-2014.pp.1-10.
9.
Hoge, F., and Swift, R., (1987), Ocean color spectral variability studies using solar induced chlorophyll fluorescence, Applied Optics, 26, pp 18-21.
10.
Kirk J.T.O., 1994, Light and Photosynthesis in Aquatic Ecosystems 2 Eds, Cambridge Univ. Press, 509.
11.
Lakesh S.H.P.K., Dahanayaka D.D.G.L., Atsushi Minato., Hideyuki Tonooka., Satoru Ozawa, (2013), Spectral signatures identifying Instrument on Spectroscopic Measurement technique for Water quality monitoring, Advanced Materials Research, 838-841 , pp 2014-2018.
12.
Lovett G.M., Burns D.A., Driscoll C.T., Jenkins J.C., Mitchells M.J.,RustadL.,Shanley, J.B., Likens G.E., Haeuber, R., (2007), Who needs environmental monitoring? Frontiers in Ecology and the Environment, 5(5), pp 253-260.
13.
Matt Richardson, Shawn Wallace, Getting Started with Raspberry Pi, Published by O’Reilly Media, Inc., 1005 Gravenstein Highway North, Sebastopol, CA 95472., 2013, p.viii.
14.
Membrey, P., and Hows, D. (2013), Learn Raspberry Pi with Linux, A press Publishers, p.xxv.
15.
Morel A. and Prieur, L., 1977, Analysis of variation in ocean color, Limnology and Oceanography, 22, pp 709-722.
16.
OECD, (2009), Smart Sensor Networks:technologies and applications for green growth, OECD Conference on, ICTs, the environment and climate change, Helsingør, Denmark, available at www.oecd.org/sti/ict/green-ict, accessed on 31 July 2014.
17.
Pereira, J.M.D, Postolache, O., Girão, P., (2009), Spread Spectrum Techniques in Wireless Communication, IEEE Instrumentation and Measurement Magazine, 12(6), pp. 21- 24.
18.
Pieter P. Tans., Inez Y., Fung., Taro Takahashi., (2002),Observational Contrains on the Global Atmospheric Co2 Budget, Science, 247(49), pp 1431-1438.
19.
Postolache. O., Dias Pereira J.M., P. Girão P.,(2009), Smart Sensors Network for Air Quality Monitoring Applications, IEEE Trans. on Instrumentation and Measurement, 58(9), pp. 3253 – 3261.
20.
Postolache, O.; Girão, P.M.; Dias Pereira, J. M.; (2011)Water Quality Assessment through Smart Sensing and Computational Intelligence, - Chapter in New Developments and Applications in Sensing Technology, Subhas C. Mukhopadhyay, Aime Lay-Ekuakille, Anton Fuchs, Springer, Berlin Heidelberg. 83, pp 191-206.
21.
Postolache O., Dias Pereira J.M., Girão P.M,(2012) Geenhouses Microclimate RealTime Monitoring based on a Wireless Sensor Network and GIS, Proc IMEKO World Congress, Busan, Korea, I, pp 1- 5.
Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
351
Development of inexpensive multi-parameter sensors based network system for water environment monitoring
22.
Postolache.O., Pereira.M., Girão.p.,(2013), Sensor Network for Environment Monitoring: Water Quality Case Study`4th Symposium on Environmental Instrumentation and Measurements, June 3-4, Lecce, Italy, pp 30-34.
23.
Raspberry Pi, (2014a), available at http://www.raspberrypi.org/wp-content/uploads/ 2013/07/RasiCam-Documentation.pdf, accesssed on 18-04-2014.
24.
Raspberry Pi, (2014b), Downloads noobs RaspberryPi, http://www.raspberrypi.org/help/noobs-setup, accessed 18 march 2014.
25.
Renesa,(2014b), available at, R8C/12,renesas.com,http://am. renesas.com/ products/ mpumcu/r8c/r8c1x/r8c12/index.jsp, accessed on 12-03-2014.
26.
Satyanarayana. G. V., Mazruddin. S.D., (2013), Wireless Sensor Based Remote Monitoring System for Agriculture Using Zig-Bee and GPS, Conference on Advances in Communication and Control Systems, (CAC2S 2013).
27.
Sahota, H., Kumar, R., and Ahmed Kamal, A, (2011), Wireless sensor network for precision agriculture, wireless Communications and Mobile Mobile Computing, Wirel. Commun. Mob. Comput, 11, pp 1628–1645.
28.
Sensor air, (2014), productspecification,http://www.senseair.se/products/oemmodules/k30/,Accesssed on 17-04-2014,pp 1-11.
29.
Sparkfun, (2014), SCP1000D01, https://www.sparkfun.com/datasheets/Components/ SCP100-D01.pdf ,Accesssed on 20-05-2014.
30.
Torrent J., (2004), Low-cost Conductivity Cells for Water measurement Purposes, IMTC 2004 Instrumentation and Measurement Technology Conference, Como, Italy.
31.
Truman S., Light, (1983), Temperature Dependence and Measurement of Resistivity of Pure Water, Anal. Chem., 56, p 1138.
32.
Upton, E., and Halfacree, G., Raspberry Pi User Guide, Wiley Publishers, (2010), p.1, http://www.raspberrypi.org/downloads, Accessed on 18-March-2014.
33.
Vaughan, H., Brydges, T., Fenech, A, Lumb, A, (2001), Monitoring long-term ecological changes through the Ecological Monitoring and Assessment Network: Science-based and policy relevant. Environmental Monitoring and Assessment 67, 3– 28.
34.
Wang, Y. X., Murray, D., Horie, T., and Yoshimura, T., (2009), CO2 Sequestration Monitoring in a Low Salinity Formation Water Environment, aapg hedberg conference, Geological Carbon Sequestration: Prediction and Verification, pp. 1-4.
35.
Yick, J., Mukherjee, B., Ghosal, D.,(2008) Wireless sensor network survey, Computer Networks, 52, pp 2292–2330.
36.
Zhou, J., and Mason, A.,(2002), Communication Buses and Protocols for Sensor Networks, Department of Electrical and Computer Engineering, Michigan State University, 2120 Engineering Building, East Lansing, Michigan, USA 488241226,Sensors 2002, 2, pp. 244-257.
Kushantha Lakesh S.H.P. et al International Journal of Environmental Sciences Volume 5 No.2, 2014
available
at
352
