This research focuses on automatic data acquisition and monitoring system for RAS ... Keywords:IoT, Monitoring system, Automation Data Ac-quisition, RAS, Fog ...
Journal of Computer Hardware Engineering (2018) Volume 1 doi:10.63019/jche.v1i2.610
IoT Monitoring and Automation Data Acquisition for Recirculating Aquaculture System Using Fog Computing Khalid Al-Hussaini1＊, Siti Maryam Zainol2, R. Badlishah Ahmed3, Shuhaizar Daud4 Department Of Information Technology, Faculty of CSIS, University of Thamar , Dhamar, Yemen Faculty of Informatics Computing, Univ. of Sultan ZainalAbidin (UniSZA), Kuala Terengganu, Malaysia 3 School of Computer and Communication, Universiti Malaysia Perlis (UniMAP), Perlis, Malaysia 1 2
ABSTRACT A monitoring system is the most important part of Recirculating Aquaculture System (RAS) to provide a healthy environment for the fish living because it will affect the quality of the fish growth and life in indoor tanks. It is more challenging in a ”controlled” environment like RAS and also in fish production especially when involving with profit, cost, and human resources. For that reason, an advanced technology has to be developed in fish production or fish catching and also to increase the awareness on benefits of fish. In contrast, the aquaculture men still monitor the data of RAS manually nowadays. This research focuses on automatic data acquisition and monitoring system for RAS using fog computing technology and low-cost system using Raspberry Pi to overcome the existing problems. The fog computing technology is applied to overcome all problems and acts as an advanced data acquisition system to keep data safely by sharing the processed data in fog computing for every tank which will be extended to the cloud and analyse the data to make an accurate control/decision in the real time. Besides, open source technology plus embedded system based will be integrated for this research because of its benefits such as small size, low cost, lightweight, portable, high efficiency and low power consumption. Finally, the efficiency of data acquisition process has been improved from manual to fog computing technology successfully. Keywords:IoT, Monitoring system, Automation Data Ac-quisition, RAS, Fog Computing.
1. Introduction High-density fish farming with Recirculation Aquaculture System (RAS) is said as the key to technology which allows the world aquaculture community to supply the world per capita needs for aquatic species over the coming decades with environmentally and friendly manner. RAS also sustainable, infinitely expandable, environmentally compatible and has the ability to guarantee both safety and the quality of the fish produced throughout the year. On other hands, RAS offers the controlled environment, the permitting controlled product growth rates and predictable harvesting schedule. The heat and water can be converted to the reused water after reconditioning by biological filtration[1]. Nevertheless, there are some parameters need to be mon-itored in RAS that can be divided into two aspects. The first aspect is the parameters of environment control such as temperature, Dissolve Oxygen (DO), PH and Ammonia (NH3) level to control the healthy environment for fish especially DO level which is the first limiting factor in an intensive aquaculture system. It can immediately cause the fish fatality if DO is not at the required level. The second aspect is the parameters to be calculated to determine the fish growth and Feed Conversion Rate (FCR) such as the weight of fish, the weight of feed and the frequency of feeding[1,2]. A real-time and efficient monitoring system is really desired for some critical parameters which can improve the quality and determine future projection by using the stored data previously. Additionally, a very low latency is required to response the output device and to alert the stakeholder and aquaculture men if something unwanted condition happens such as temperature going higher. The secure and reliable system also highly demanded due to high cost and profit business such as RAS[1,3]. Copyright © 2018 Khalid Al-Hussaini et al. doi: 10.63019/jche.v1i2.610 EnPress Publisher LLC.This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). http://creativecommons.org/licenses/ by/4.0/
1
Consequently, fog computing technology is suggested in this research to make RAS monitoring system is more efficient. Fog computing offers a reduction of network traffic by pro-viding a platform to filter and analyze the data generated by the devices which close to the edge and for local data views. As result, it will automatically reduce the traffic being sent to the cloud[4,6]. Fog computing also suitable for the Internet of Things (IoT) tasks and queries which most of the smart devices need to capture events only about a hundred meters from it, no need to access global data from the cloud. Low latency requirement also provided by fog computing which critical parameter can be reflected with a high-speed real-time response. Furthermore, it can reduce the scalability issue since fog computing aims the incoming data get closer to the data source itself and reduces the burden of that processing in the cloud and without concern of the increasing number of endpoints[1,5,7]. This research aims to design and develop an automatic data acquisition system for recirculation aquaculture system (RAS) using fog computing. It focuses on a system contains data collection, data analysis, data sharing on the server by using fog computing technology and decision making to control RAS. Two Raspberry Pi units are used to collect data from RAS.
2. System Overview Automatic data acquisition for recirculation aquaculture system (RAS) using fog computing consists of data collectors which act as client known as RaspDAQ and one server called as RaspFog as shown in Fig 1. In this research, both collectors and server use Raspberry Pi 3 as the platform. It is because it is very cheap, has friendly general purpose input output (GPIO) pin and uses the free operating system, Linux. In addition, it has the most important and special feature in this research which is 802.11n Wireless LAN.
Fig. 1. Automatic data acquisition for RAS using Fog computing overview.
Fig. 2. Block diagram of automatic data acquisition for RAS using Fog computing.
2
RaspDAQ collects the temperature and water level data and displays a graphical user interface (GUI) of RaspDAQ by real time. At the same time, the data and the camera input is sent to RaspFog using WiFi and store in MySQL server. WiFi has its own strength compare which suitable for this research to others. RaspFog generates graphs for the temperature vs. time and water level vs. time. The GUI of the server can be displayed by browsing the server website from any devices as they are connected to the same network. Fig 2 shows the block diagram of automatic data acquisition system for recirculation aquaculture system (RAS) using fog computing. Every RaspDAQ consists of one temperature sen-sor and one water level sensor. Every RaspDAQ also connects to camera module to view the condition of every fish tank which attaches to RaspDAQ. The monitor is connected to RaspDAQ to view the real-time data which are current temper-ature, temperature average, minimum temperature, maximum temperature, current water level, water level average, minimum water level and maximum water level. If there are warnings from the sensors, the temperature status LED, water level status LED and the buzzer is activated. RaspFog receives the data from RaspDAQ and plots the graphs for each data. Monitor, keyboard, and mouse are connected to RaspFog to allow administrators and users to print, save and view the website of RaspFog easily. RaspFog also has a response system which consists of email and Short Message Service (SMS). The email and SMS are sent to notify the admin and the user if there are limited values from the received data. Research methodology is conducted in three phases, which starts with data collecting, data processing, system integration and whole system testing Phase 1: Design a Data Collecting System for RAS Fig 3 shows the process flow to design a data collecting system for RAS which known as RaspDAQ. This research consists of two RaspDAQs acting as a smart device which they can: 1) Collect data from temperature sensor and water level sensor. 2) Display data using Graphical User Interface - Tkinter: Current date, time and system name and version. Current temperature and water level values. Average temperature and water level values. Minimum temperature and water level values. Maximum temperature and water level values.
Uptime for the system: The duration of system operation.
3) Detect and alert the limited value using LED and buzzer. 4) Send the current data and camera view to RaspFog for every 5 seconds. These process flow details are explained in the next subtopics.
3
Fig. 3. Phase 1: Design a data collecting system for RAS process flow.
Phase 2: Design A Data Server and Processing System using Fog Computing. Fig 4 shows the process flow of designing a data server and processing system using fog computing. Raspberry Pi 3 is used only with WiFi technology and no connection with any component to setup as fog computing server, RaspFog. The next subtopics explain the steps to: 1) Install Apache, PHP, and Mysql server. 2) Create tables and run the required command in MySQL server. 3) Write a program for Graphical User Interface (GUI) using web-based: Home, log in and welcome page. Graph for continuous real-time data. Alert system using email and SMS. 4) Integrate data collecting system and data processing system. 5) Test: Log in and welcome page. Graph plotting. Email and SMS sending.
4
Fig. 4. Phase 2: Design a data server and processing system using fog computing
Phase 3: Integrate, Test and Validate the Automatic Data Collection and Processing Strategy for RAS Fig 5 shows the process flow of Phase 3 which integrate, test and validate the automatic data collection and processing strategy for recirculation aquaculture system (RAS). Firstly, RaspDAQ1, RaspDAQ2, and RaspFog are connected to execute data collecting, processing and sharing. Then, Python programs for RaspDAQ1 and RaspDAQ2 are executed to send data to RaspFog.
3. Result And Discussion
This section is divided into three subsections which are, Data collecting system for RAS, RaspDAQ, Data process-ing system and server using fog computing, RaspFog, and
5
Fig. 5. Phase 3: Integrate, test and validate the automatic data collection and processing strategy for RAS.
Integrate, test and validate automatic data collection and processing strategy for RAS. Data collecting system for RAS, RaspDAQ Fig 6 shows two data collecting systems for RAS, Rasp-DAQ1, and RaspDAQ2 are setup successfully. Each RaspDAQ has temperature sensor connects to Raspberry Pi through analogue digit converter, MCP3002, camera module, LEDs and water level sensor, HC-SR04.
Fig. 6. Data collecting system for RAS, RaspDAQ.
Next, Fig 7 shows GUI for RaspDAQ using Tkinter library as designed. It displays the system name and version, cur-rent date and time, current temperature, temperature average,
6
minimum temperature, maximum temperature, current water level, water level average, minimum water level, maximum water level and system uptime. The temperature values are in unit
and
. The system uptime starts
counting when the system is power on.
Fig. 7. Graphical user interface (GUI) for RaspDAQ.
Data Processing System and Server using Fog Computing, RaspFog Results, Test and Validation Data processing system and server using fog computing, RaspFog results, test and validation is divided into following part: 1) Tables and required command in MySQL server. 2) Home, log in and welcome page. 3) Graph plot. 4) Email and SMS. There are tables and required command has been done in RaspFog MySQL server as shown in Fig 8, which shows the tables list in RaspFog database. Table temperature storage while age. Table
is for current is for water level stor-
,
,
and
are created for
average, minimum and maximum value for each sensor. Table user is for authorised RaspFog system users. Next, Fig 9 shows the content of table
.
Every data stored is real time transferred from RaspDAQ. The same procedures are applied to table and
,
,
,
,
.
It can be seen that each data received every
second. The welcome page offers many access options which
are displaying a graph of water level sensor, RaspDAQ condition view
7
real-time video, and RaspDAQ temperature
sensor as shown in Fig 10, Fig 11 and Fig 12. There are four main graphs for every RaspDAQ which are current, average, minimum and maximum of every RaspDAQ data sensors. They are displayed in RaspFog system by click-ing RaspDAQ option button. Fig 13, Fig 14 and Fig 15. are the
Fig. 8. Tables in RaspFog database.
Fig. 9. Table
content.
8
Fig. 10. RaspDAQ water level sensor page.
graph examples of current, average, maximum and minimum RaspDAQ temperature sensor.
Fig. 11. RaspDAQ condition view page.
Fig. 12.
Current temperature vs. date, time.
9
Fig. 13. Temperature average vs. date, time.
Integrate, Test and Validate the Automatic Data Collection and Processing Strategy for RAS Fig 16 shows Automatic Data Acquisition for Recirculation Aquaculture System (RAS) using Fog Computing is setup successfully. The system integrates two data collecting systems for RAS, RaspDAQ1 and RaspDAQ2 and data server and processor, RaspFog. The set values are detected when sensor
Fig. 14.
Maximum temperature vs. date, time.
Fig. 15.
Minimum temperature vs. date, time.
page is entered or refreshed. The sensor page also mentions that message has been sent when the condition of
10
sensor is not good as shown in Fig 17.Every phase of automatic data acquisition system for recirculation aquaculture system (RAS) using fog computing is tested and validated successfully.
4. Conclusion
Fig. 16.
Automatic data acquisition for RAS using fog computing.
Previously, there is no Fog computing implementation in RAS. By implementing Fog, no dependency to internet access
Fig. 17. Alert message when sensor value is not good.
because the server is localized
intranet. Furthermore, Fog is low latency, the IoT devices are very near to the
server. This research has achieved the objectives which are design and develop data collecting system, data processing system using fog computing for RAS and validate the system. The data collecting system for RAS (RaspDAQ) is developed by connecting Raspberry Pi 3 to temperature sensor (LM35DT) using ADC MCP3002, water level sensor (HC-SR04), Rpi camera module, LEDs and buzzer. Software and program are built using Python and Apache server to run every functions of RaspDAQ. Two RaspDAQ are used in this research which are RaspDAQ1 and RaspDAQ2. While third Raspberry Pi 3 is setup as data processing and server system (RaspFog). Raspfog uses PHP, Apache and MySQL database. Both RaspDAQ and RaspFog are based on a Raspbian operating system. After that, RaspDAQ1 and RaspDAQ2 are connected to RaspFog using WiFi technology to send sensors data in real time. The received data are stored and plotted using Highcharts.com graph. Both RaspDAQ, RaspFog have been tested and validated. At the same time, users can see the graph output in the real time for temperature, water level sensor and real condition using Rpi camera module of RaspDAQ1 and RaspDAQ2 by browsing RaspFog website.
11
References 1. AV Dastjerdi, H Gupta, RN Calheirosand, et al. Fog Computing: Principles, Architectures, and Applications, CoRR 2016; arXiv: 1601.02752v2; 1–26, . 2. F J Espinosa-Faller, GE Rendon-Rodriguez. A zigBee wireless sensor network for monitoring an aquaculture recirculating system,Journal of Applied Research and Technology 2012; 10; 380–387. . 3. J Bregnballe. A guide to recirculation aquaculture: an introduction to the new environmentally friendly and highly productive closed fish farming systems. Food and Agriculture Organization of the United Nations (FAO) and EUROFISH International Organisation, 2016; 10; 1–93. 4. Fog vs. Cloud (2015), https://blogs.cisco.com/perspectives/iot-from-cloud-to-fog-computing, retrieved on 3 April 2016 5. N Vijayakumar, R Ramya. The real time monitoring of water quality in IOT environment, IEEE Sponsored 2nd International Con-ference on Innovations in Information,Embedded and Communication systems (ICIIECS) 2015. 6. SS Lagu, SB Deshmukh. Raspberry pi for automation of water treatment plant. International Conference on Computing Communica-tion Control and Automation 2015; 532–536. 7. IoT Growth , https://www.bbvaopenmind.com/en/iot-a-fog-cloud-computing-model/, retrieved on 3 April 2016
12
