CC BY
23УДК 623.454.862 DOI 10.25257/FE.2020.4.66-72
Janos PETRANYI
National University of Public Service, Budapest, Hungary E-mail: [email protected]
Lajos KATAI-URBAN
PhD, Associate Professor, National University of Public Service, Institute of Disaster Management, Budapest, Hungary E-mail: [email protected]
Attila ZSITNYANYI
National University of Public Service,
Budapest, Hungary
E-mail: [email protected]
INVESTIGATION OF THE ARCHITECTURE OF EARLY WARNING RADIATION MONITORING SYSTEMS
Incorrectly chosen technology and architecture used in radiation monitoring systems prevent collecting high-quality information in the right place and at the right time. The authors set a goal to find new architectures, protocols and requirements for radiation monitoring systems.
The efficiency of a radiation monitoring system can be increased, and new capabilities can be implemented by using optimised architectures, specific protocols and intelligent radiation detectors.
An intelligent radiation detector is intelligent because it processes, analyses, interprets the analogue signal coming out from a sensor, runs various algorithms on it, then generates information relevant to the user, performs tasks, stores data and communicates with the outside world.
A combination of different intelligent detectors can build a modern monitoring system, which can supervise borders, facilities and entire countries. This article focuses on what components and how they should be implemented to get the most suitable system for disaster management purposes.
Key words: radiation protection, radiation measurement, monitoring systems, protocol, data exchange.
The components of a radiation monitoring system are monitoring stations, communication infrastructure, data centers, information broadcast subsystem, reconnaissance units.
The simplest monitoring system requires one radiation monitoring station to measure and one data center to visualise the data. If the task of the radiation monitoring system is to continuously monitor the site and the radiation measurement and data visualisation must be performed at two distant locations, data communication infrastructure must be established between the station and the data center. The information available at the data center should be shared with professionals, and the public should also be informed in case any action like evacuation is required. For this purpose an information broadcast subsystem should be implemented. After an alarm signal, a reconnaissance unit should be sent on-site with the capability to analyze the threat with extended measuring capacity.
data as a National Radiation Monitoring, Signaling and Control System (OSJER) [1].
Since 1962, the Hungarian Defense Forces (MH) has been responsible for establishing a unified radiation monitoring and evaluation system. In 1993 MH established a telemetry network with 50 automatic fixed measuring stations. The monitoring stations in the MH Automated Measuring and Data Collection System (AMAR) have been installed since 2000 and have since been renewed several times. The AMAR automatic environment monitoring system is capable of continuously and independently measuring gamma background radiation of a given area and supplying information into the OSJER system.
The other monitoring system in the OSJER system called the Automated Radiological Industrial Safety Telemetry Network (RTH) is operated and maintained by the National Directorate General for Disaster Management (BM OKF) [2].
RADIATION MONITORING SYSTEM IN HUNGARY
n Hungary, there is Government Regulation about the National Nuclear Emergency Response System. This regulation gives the task to install, operate monitoring systems and use the collected
MONITORING STATIONS
The most essential component of a country size monitoring system is the monitoring station which has the measuring, data collecting and data transmitting capability to support any data center with relevant information.
66
© Petrányi J., Kátai-Urbán L., Zsitnyányi A., 2020
1. Sensing environmental parameters. The first generations of monitoring stations were capable of measuring only a few parameters. Today's stations have intelligent detectors with a wide range of measurable environmental parameters. Essential parameters like temperature, wind direction, wind speed, gas concentration, radiation level have been expanded with such parameters as humidity, air pressure, air activity concentration. In the future, the capabilities of the stations will include nuclide identification, for example.
The most common capabilities of monitoring stations used for disaster management purposes are:
- measurement of toxic industrial chemical and chemical warfare agents;
- measurement of gamma dose rate;
- determination of the activity concentration of airborne radioactive alpha, beta, gamma active particles;
- measurement of meteorological parameters;
- collection and transmission of measured data.
Basic measuring devices transform the
environmental parameter into a human-readable value. The mercury thermometer is a good example of it. As the temperature changes, the mercury level will change too, after the stabilization of the mercury level the actual temperature can be read on the scale. In the early times sensors converted environmental parameters into a measurable analogue electrical signal, which was processed by a multi-channel measuring electronics and showed all the measured values on different analogue scales [3]. The disadvantage of this technology is that analogue signals are transmitted via wires, exposing the measured value to significant noise. Another disadvantage of this architecture is that the multichannel measuring electronics had to deal with each of the connected analogue sensors separately, compensate their values and provide data to the user, resulting in overcomplicated multi-channel measuring electronics. Nowadays, the analogue output of the sensor is digitized as near as possible to the sensor. Moreover, a dedicated microcontroller is directly connected to the sensor to establish a standard interface. The advantage of this technology is that it lowers the noise level and reduces the role of the central unit, and one detector malfunction will not cause an overall system shutdown.
An intelligent detector consists of:
- a sensor, which provides the analogue signal;
- an analogue/digital converter, which makes possible for the processor to read the signal;
- a processor, which can do compensation, processing tasks and create average values;
- a communication unit, which can send the digitalized measured value to a higher informatics level.
The parts of a general-purpose sensor can be seen in figure 1.
Most sensors (the unit that converts an environmental parameter to a measurable raw analogue electrical signal) have temperature dependency. An intelligent detector can compensate temperature dependence with built-in temperature sensors and calibration factors recorded previously in the climate chamber.
Many intelligent detectors can automatically calibrate themselves before or during measurement. A micro-controller needs to interrupt the measurement at a given moment by placing a built-in etalon (reproducing known environmental parameter quantity) in front of the sensor; with that the detector is able to adjust itself. An example of temperature compensation is when LED is used for scintillation detector spectrum stabilisation. The temperature should be compensated because LEDs modify their light output as temperature changes [4].
2. Processing the measured data. In addition to directly measured data, the determination of post-processed parameters has been added to the monitoring station capability list. Post-processed or indirect calculated values like comfort temperature, dew point, the cloud base can be determined with the help of directly measured parameters, and specific formulas. Nowadays, the post-processes can be set up in most monitoring stations, in some cases, even user-configurable.
The first large-scale use of chemicals occurred during World War I; since then wind speed and wind direction parameters have been measured in the field of CBRN to determine dispersion [5]. For dispersion calculation vertical air stability is also a necessary parameter. In order to calculate vertical air stability, temperature parameters should be measured at several
Environmental parameter
A.4
Analogue signal
Digital signal
Data package
c Sensor ч
4 J
A/D converter
( N
Processor
4 )
Ж
Communication unit
T
с \
Intelligent detector __
Figure 1. Structure of the intelligent detector. Source: compiled by the author
heights [6]. A dispersion calculation can use local meteorological data, physical, chemical form, and the amount of a released hazardous substance [7]. There are various calculation models for the dispersion of both chemical and radioactive materials, which have become more and more usable with the advancement of computer technology. The results of the models, when integrated with mapping systems, can give an accurate picture of the areas that may be affected by a disaster. If the model is linked to the population database, it is possible to filter out the number of people needed to be evacuated from the affected area.
With technology improvements new detection technologies have become widespread, enabling online measurement of previously unmeasurable parameters. Such an ability is to determine the activity concentration of airborne radioactive alpha, beta, and gamma active particles. Previously, these abilities could only be part of an offline, after event monitoring system.
In many cases, it took the laboratory method days to run, and the system significantly slowed down the response time. Implementation of the measurement at the monitoring station was often hindered by the fact that it was not possible to perform the required measurement in the field outside the laboratory [8]. There were several reasons for this: the measuring detector could not withstand extreme environmental conditions. Typically, in Europe, a temperature range of -30 °C to +50 °C causes problems for an analyser designed to operate at +25 °C ±5 °C ambient temperature. Efforts were made to build a separate container around laboratory detectors, so that it was possible to maintain a constant climatic condition for the detectors, which significantly increased the cost of monitoring stations.
3. Alarm handling processes. In a normal operation mode, the station sends measurement data at specific intervals, usually every 10 minutes. After a communication failure stored data is sent to the higher IT system on a first come first served (FIFO) based logic. During this process, an alarm event will come up after all stored data had has been sent, which can cause delay during the most important time period on handling disasters.
In more advanced systems, the alarm data packages are sent before the standard, background measurement data packages. If the station does not favour the alarm packages, the alarm message will not be able to reach the data center for up to several hours, as only stored standard historical data is transmitted instead of sending alarm message first. Another alarm management option is that the alarm message is transmitted to the data center via a route independent of the standard data traffic channel. For example, the data is sent over GPRS packages, while the alert message is sent by SMS. For this solution, both data channels must be available between the station and the data center. An additional alarm event management process can be the deleting of all the previously measured data stored before the event. The disadvantage of this is that the
pre-alarm data will be missing from the archive, which may cause decision uncertainty.
Reliability of the system. The reliability and availability level of a monitoring system can be increased by using redundant detectors. If two detectors are used, in the event of a failure of one, the station will still be able to provide data with the help of the other detector. The best solution is for the secondary detector to measure a given environmental parameter with a completely different technology. An example of this is the simultaneous use of scintillation and GM detectors. While the GM detector is capable of operating at high dose rates, the scintillation detector is capable of performing energy selective measurements. The two detectors complement each other's capabilities as well as replace one another if a failure came up. There is a problem with redundant detectors when the two measurement results are contradictory and it is not possible to decide which detector has failed.
Along with the increase of measurement capabilities, the issue of physical protection of the station has also come to the surface, as high-value measuring equipment has been placed in external locations without direct human supervision.
Physical protection of the monitoring stations currently in use generally security solutions that meet today's requirements. Opening, tamper sensors, remote video surveillance can be mounted on the stations.
Positioning systems (GPS) can be integrated into monitoring stations. This technology allows tracking the position of a given measuring station and associating the measurement data with an accurate time stamp and coordinates. Continuous sending of coordinates is unnecessary for a station that is permanently installed, but essential for a mobile station.
Monitoring stations equipped with intelligent detectors are able to do self-diagnosis, allowing errors to be reported in time or even before the actual failure occurs.
COMMUNICATION INFRASTRUCTURE
There are several ways of transmitting the measured data from the monitoring station to the data center. In the past, an analog signal from a radiation detector was routed via a cable to the display unit. The analog signal is disturbance sensitive at greater distances. Therefore, instead of analog signals, information is now transmitted via digital data transmission. The method of data connection between the display unit and the detector is determined by pre-established rules (hereinafter: protocols).
Monitoring stations are equipped with communication data transmission means that ensure the secure exchange of data between the station and the data centers. For on-site evaluation of the data, a handheld readout and display unit can be used, which can even download data and perform maintenance activities. Detectors installed on the monitoring stations measure
the given environmental parameters. Then a signal proportional to the measured value is stored, sometimes processed and transmitted to the data center. The data center can receive data from all monitoring stations, displaying signals information and alerts the operators if the user action is required. The use of multiple data centers requires that the system communicates over various communication channels and that each data center can be operated in parallel.
Mobile and fixed monitoring stations can be part of a monitoring system. Stations can communicate simultaneously across multiple communication channels.
Protocols used for data exchanges. Some countries share the data, that their system has measured. Within the European Union, the data exchange of background measurement networks is in most cases based on the IRIX protocol [9]. This protocol defines the format in which data is to be saved in a file and transmitted to another country. The format of the file is XML [10] based, which means that the data is included in so-called "tags". The title of the "tag" clearly identifies the data it contains, so when someone opens the file, they can interpret it accurately. The XML file is human-readable, editable with a simple text editor, so data security is not provided at the file level. Information protection can be made by means of a secure data transmission channel or encryption of the entire file. Some part of the data in these shared files is public and available on international websites [11].
In addition to background dose rate level data, other information is shared between countries. The data exchange standard initiated by the US authorities is very popular in many parts of the world. The N42 data format is intended to facilitate the sharing of data from radiation measuring instruments between law enforcement agencies [12]. In this format, in addition to dose rate, other measurement results, such as the measured spectrum or measurement settings, can be transmitted. XML is also the basic structure of this standard.
In addition to the US standard for data exchange, the European Union's own standards have also been published [13].
The main features of the IEC 63047 standard are that it is binary, unreadable for humans, optimized primarily for machines, so it can be used with good efficiency for data exchange between machines. It also includes encryption and authentication features.
DATA CENTERS
The data center must be able to receive and evaluate the data sent by monitoring stations. The data center and display units can be installed in a building, integrated into a vehicle or even in a bag-sized box, so that the operator can access the measurement data in all circumstances.
In addition to data collection, data centers can be linked to other existing systems through their databases, enabling them to manage action plans, geospatial and
dispersion calculation systems, supporting a mission control system.
For general monitoring systems, it is common for stations to communicate with only one data center at a time. However, it is nowadays required that multiple data centers should be able to operate simultaneously, increasing the availability of the whole system. In an emergency, the entire system is rendered inoperative if the data center fails. This situation can happen for several reasons. Data center power is lost, data communications are interrupted, or only the hardware, software components in it are malfunctioning or corrupted. As a result, data centers nowadays are not just a fix installed server in a building but can be either onboard or installed in a suitcase.
The data center may be damaged due to an earthquake, it loses its ability to collect data from the monitoring stations, and so the information needed by decision-makers will be not available.
This situation can be solved if a secondary data center takes over the role of a failed data center. By launching a mobile data center, the information service can be restored or even moved to a place where it can be safely operated.
An ideal solution is to have at least two fixed data centers, as far apart as possible, with independent infrastructure support. A natural or man-made disaster rarely extends to a nationwide scale, leaving one data center with a good chance of survival. In addition to the two fixed data centers, a mobile data center also provides significant system security, as it is more comfortable to move and more durable in emergencies with portable power generating units.
The possibility of parallel data centers raises several issues. The first problem is that sending data generated by the stations to the two data centers causes unneeded duplication of data traffic, especially since the stations in most cases have wireless data transmission, where the data transmission fees are already higher. Therefore, it is advisable to configure the secondary data center to operate so that communication with the stations begins only when the primary data center has failed. Direct data synchronization between primary and secondary data centers can also be useful, provide up-to-date data at both locations.
The other problem with parallel data centers is the difficulty of organizing the different data transmissions. The stations usually have local data storage capacity. This feature allows storing the locally measured data in the event of a communication error and sending the stored data once the communication channel has been established again.
There are two major types of data transmission between stations and a data center:
- polling type data transmission (Master-slave);
- multiple sender type of data transmission (Multi-Master).
Polling type data transmission (Master-slave). The data center will periodically ask each station to see if there is
□DD
ODD
DDD .....................................................................
Data center 1
ODD
DDD
DDD
Factory
Data center 2
Figure 2. Polling architecture of a monitoring system. Source: compiled by the author
any new data available, and if so, it will collect it. Figure 2 shows the polling architecture of a monitoring system. The advantage of this solution lies in its simplicity. The data center can allocate its data channel resources and communicate at the appropriate synchronous time. The disadvantage is that an alarm event can only arrive at the data center with a delay. In a multi-station system, in the worst case, the alarm triggering event occurs at a station just after the data transfer from this station to the data center is finished, so the system will only be notified of the alarm event after all other stations have been handled and the alarming station again gets in the row. If a system uses 2G data calls, the time of data transfer per station can take up to 3 minutes. In case of a few dozen stations, the alarm event can be unnoticed for hours.
Much better results can be achieved if the data center can handle multiple stations simultaneously, or if the monitoring system uses a faster data transfer method. Data transfer using TCP/IP packets based on 3G/4G/5G technologies allows further acceleration. The full response time of the system depends on the speed of the data channel and the response time of the endpoints devices. In this polling mode, the data center cannot proceed to query the next device until the current device responds.
If the device fails, there is no response, so the data center must start a timer after the request was sent to the station. If there is no answer from the station during this period, the data center should continue the data collection with another station. The waiting time should be picked carefully, as a short waiting time will cause a system to declare a slow device to be defective,
and a long waiting time may unnecessarily increase the overall query time.
Multiple sender type of data transmission (MultiMaster). Every station has the right to send data to the data center asynchronously, without losing information. Multiple implementations of this data communication method are possible. Figure 3 shows the Multi-master architecture of a monitoring system.
One organization architecture example is based in the MQTT protocol [14], designed for querying detectors, which allows managing the mass of detectors over the Internet. The detector subscribes to a channel in the data center where it can send a message at the time it chooses. This technology is ubiquitous in civilian life, using the Internet of Things (IoT) technology, which is the common name of how smart detector is connected to the Internet. The advantage of this technology is easy implementation, and there are many providers in the cloud to collect detector data. The disadvantage is that it is less secure due to its simplicity and challenging to integrate into multiple data center systems.
Another Multi-Master organization architecture example is using data transmission via TCP/IP ports. The stations open a TCP/IP channel at the data center and in case of an alarm, first the alarm packet and then the typical measurement results are sent to the data center. The data center returns a confirmation message, after which the station can send the next data package until all the stored data is sent. This solution provides increased availability, reduces data loss, and provides data security with encryption. This working method can serve multiple data centers if the station does not
i
2
3
4
Data center 2
Figure 3. Multi-master architecture of a monitoring system. Source: compiled by the author
receive any confirmation; it can send data to a secondary data center. In addition to the asynchronous auto-send mode, the station can also be requested from the data center, and it is possible to allocate time slots to each station so that data transmission can be distributed evenly according to the limited bandwidth.
Bottom-up data transmission also allows solar-powered, energy-efficient operation. The detectors are turned off by default. For each measurement cycle, it will wake up for one measurement task, then turn on the data transfer unit and send the data to the data center. This technology is a great temptation because no power supply or data transmission cabling is required during installation, significantly reducing installation costs. Unfortunately, this solution is not applicable in early warning systems because the detectors must be
continuously on and should send an alarm as soon as the alarm level is exceeded.
Monitoring systems have two other important components: an information broadcast subsystem and a reconnaissance unit, but their detailed analysis is expected in a future paper.
SUMMARY
Different algorithms, architectures and protocols have been implemented into radiation monitoring systems. With the help of such systems, the time between the first sign of a disaster and the first response to prevent can be reduced. Already existing systems can be updated to increase accuracy, reliability, to widen the measuring ranges and add new capabilities.
REFERENCES
1. Magyarorszag Kormanya 167/2010 (V. 11) Korm. rendelet az orszagos nukleârisbaleset-elhârftâsi rendszerrol [Government of Hungary. 167/2010. Government Decree on the National Nuclear Accident Prevention System] Network library. 2010, vol. 167, pp. 1-13. Available at: https://net.jogtar.hu/ jogszabaly?docid=a1000167.kor (accessed September 18, 2020) (in Hung.).
2. Katai-Urban L., Vass G., Zellei G. 25 éve mûkodik hazankban a radiologiai tâvmérô halozat. Hadtudomany. Hadtudomânyi Târsasâg Folyôirata (Military science. Journal of the Military Science Society). 2018, vol. 28, no. 2, pp. 140-146 (in Hung.). DOI: 10.17047/HADTUD.2018.28.2.140
3. Zheng B., Chen X. Design of Multi-channel data acquisition system based on single Chip computer and LabVIEW. Journal of Physics: Conference Series. 2019, vol. 1345, no. 6. DOI: 10.1088/1742-6596/1345/6/062050
4. Saucke K., Pausch G., Stein J., Ortlepp H.G., Schotanus P. Stabilizing scintillation detector systems with pulsed LEDs: a method to derive the LED temperature from pulse height spectra. IEEE Transactions on Nuclear Science. 2005, vol. 52, no. 6, pp. 3160-3165. DOI: 10.1109/TNS.2005.862929
5. Fitzgerald G.J. Chemical Warfare and Medical Response During World War I. American Journal of Public Health. 2008, vol. 98, no. 4, pp. 611-625. DOI: 10.2105/ajph.2007.11930.
6. Dabberdt W.F., Frederick G.L., Hardesty R.M., Lee W.C., Underwood K. Advances in meteorological instrumentation for air quality and emergency response. Meteorology and Atmospheric Physics. 2004, vol. 87, no. 1-3, pp. 57-88. DOI: 10.1007/s00703-003-0061-8
7. Casal J. Chapter 6 Atmospheric dispersion of toxic or flammable clouds. Industrial Safety Series. 2008, vol. 8, pp. 195248. DOI: 10.1016/S0921-9110(08)80008-0
8. Honkamaa T., Toivonen H., Nikkinen M. Monitoring of Airborne Contamination Using Mobile Equipment. U.S. Department of Energy Office of Scientific and Technical Information, 1996.
9. International Radiological Information Exchange (IRIX) Format. International Atomic Energy Agency. Vienna, 2020. Available at: https://www-pub.iaea.org/MTCD/Publications/PDF/ EPR_IRIX_web.pdf (accessed September 18, 2020)
10. Bray T., Paoli J., Sperberg-McQueen C.M., Maler E., François Y., Cowan J., Extensible Markup Language (XML) 1.1 (2nd Ed.). World Wide Web Consortium, 2006, pp. 1-38. Available at:
https://www.w3.org/TR/2006/REC-xml11-20060816/ (accessed September 18, 2020).
11. EURDEP Gamma Dose Rates Simple Map. Joint Research Centre, Radioactivity Environmental Monitoring, 2020. Available at: https://remap. jrc.ec.europa.eu/Simple.aspx (accessed September 18, 2020).
12. ANSI/IEEE N42.42 Standard. National Institute of Standards and Technology. U.S. Department of Commerce, 2018. Available at: https://www.nist.gov/programs-projects/ansiieee-n4242-standard (accessed September 18, 2020).
13. IEC 63047:2018 Nuclear instrumentation - Data format for list mode digital data acquisition used in radiation detection and measurement International Electrotechnical Commission. 2006, pp. 1-13. Available at: https://www.sis.se/api/document/ preview/80009626/ (accessed September 18, 2020).
14. MQTT Version 3.1.1. Edited by A. Banks and R. Gupta. 29 October 2014. OASIS Standard. Available at: http://docs.oasis-open.org/mqtt/mqtt/v3.1.1/os/mqtt-v3.1.1-os.html. Latest version: http://docs.oasis-open.org/mqtt/mqtt/v3.1.1/mqtt-v3.1.1.html
Материал поступил в редакцию 6 октября 2020 года.
