Back to the programme printer.gif Print




Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'Advanced operation & maintenance' taking place on Thursday, 13 March 2014 at 11:15-12:45. The meet-the-authors will take place in the poster area.

Jaime Laviada Universidad de Oviedo, Spain
Co-authors:
Jaime Laviada (1) F P Marcos Rodríguez-Pino (1) Fernando Las-Heras (1)
(1) Universidad de Oviedo, Gijon, Spain

Printer friendly version: printer.gif Print

Abstract

Wireless sensor network to monitor internal safety in a multi-use open-sea platform harvesting wind and wave power

Introduction

The aim of the H2OCEAN project [1] is the design of multi-use offshore platform to exploit renewable energy. The platform includes a set of hybrid floating units to harvest wave and wind energy. This energy is (partially) transformed into hydrogen so it can be easily transport to the cost by means of ships rather than conventional submarine cables.
In this paper, a flexible strategy to monitor the H2OCEAN safety, which is mainly related to the hydrogen storage, is described. Special emphasis is devoted to the design of autonomous long-range wireless nodes that can be fed by solar panels.


Approach

The H2OCEAN platform involves a large number of processes running in parallel. Abnormal working in some of them can yield risk situations. In particular, hydrogen leakages and hydrogen flames are identified as the most critical variables to monitor. A wireless sensor network has been identified as the most successful technology to monitor the safety of the platform due to the high flexibility to meet a large number of platform layouts as the ones that enable the modular structure of the H2OCEAN platform.
In addition to the risks associated to hydrogen, the nodes have to be able to handle other variables that are relevant for the correct working of the platform such as sea water temperature, nitrogen in the aquaculture cages, osmosis tanks level, etc. Moreover, the platform spans along several squared kilometres and, consequently, the distances to cover are in the order of several kilometres. Low-power communications systems covering large distances have usually strong bandwidth limitation. Hence, a careful choice of the number of nodes and the data to be sent has to be done.
In summary, the communication technology must comprise the following features: i) capability to handle a large number of nodes; ii) cover large outdoor distances; iii) autonomy to avoid or simplify the platform wiring; iv) the use of directional antennas must be avoided since they are expected to exhibit a poor behaviour in a floating platform.
The used of low-level radiofrequency modules working at low ISM frequency bands equipped with a set of standard connections has been identified as the best solution. In particular, the bands of 868 MHz (Europe) and 900 MHz (USA) provide good coverage in outdoor environment even if non-directional antennas are used. Hence, the use of directional antennas (e.g., Yagi antennas), which would require a stabilisation system if they are placed in a floating platform, is avoided. This technology is implemented by the XBee-Pro 868® and XBee-Pro 900® manufactured by Digi® [2].
The design of the network involves taking into account the following points: i) node location; ii) number of nodes; iii) total bandwidth; iv) autonomy of the nodes; v) Maximum distance.


Main body of abstract

The H2OEAN project aims the design of an offshore multi-use platform. The consortium is composed of seventeen partners from 5 different countries. In this platform, several floating units are placed to harvest energy from wind and waves. The platform is also equipped with units that are in charge of the purification and desalination of ocean salt water so that the filtered water can be used in an electrolysis process. This electrolysis is also powered by the harvested energy producing hydrogen and oxygen. The resulting hydrogen is stored and periodically shipped to the coast where it can be turned into electrically power again. Similarly, the produced fresh water is also periodically ship. In addition, the platform will host a multi-trophic fish farm powered by part of the harvested energy.
The platform was originally expected to be unmanned and, therefore, any potential risk has to be detected as soon as possible in order to carry out the corresponding actions. Major risks are associated to hydrogen which can easily escape due to the molecular properties of this extremely light gas. Moreover, these leakages can easily result in deflagrations or detonations due to the very low hydrogen LFL (lower flammability limit). Furthermore, these leakages can also result in flames which are not visible for the human eye with the corresponding hazard.
In order to monitor the safety of the platform, a wireless sensor network has been designed. Previous approaches for wireless hydrogen detection have mainly focused on onshore facilities wherein the communications do not require the autonomy and distance features previously discussed (e.g, [3]).The designed network consists on a set of peripheral nodes as well as a central node. The peripheral nodes are based on one RF module equipped with the circuitry to connect the sensor with standard outputs. In particular, RS232, 0-5V and 4-20mA outputs can be directly connected. Commercial solutions [4] as well as ad-hoc designs have been considered with similar results. The RF modules enable communications at distances of approximately 10 km and, therefore, it suffixes the maximum dimensions of the platform. The nodes are configured so that peripheral nodes are powered by a battery of 2300mAh together with a solar panel. Power consumption calculations show that the system can work without the need of external energy for 16 days. This limit has been overcome by including a solar panel with each peripheral node.
The network is configured as a star topology. Thus, only transmissions from the peripheral nodes to a single central node are considered. Peripheral nodes carry out a sampling of the variables which are monitored with a period between 30s and 1 minute depending on the variable to monitor. In case of abnormal values, the nodes can send an alert signal without the need of waiting for the periodic sampling. First estimations of the number of nodes reveal that this number is in the order of 400 resulting in a bandwidth of 3 kbps, which meets the specifications of the RF modules.
Commercial hydrogen sensors are chosen to detect concentration of 10000 ppm in any point where hydrogen could be accumulated due to a leakage. The sensor KHS-200 MEMS distributed by Kebaili has been chosen to detect these leakages. On the other hand, several flame detectors based on infrared cameras are placed to monitor the storage sensor. The model X3302 manufactured by DET-TRONICS has been chosen for this purpose.
All the information is gathered by a central node which is connected to a computer equipped with redundant systems. The data are stored in a database and they are periodically sent to the ground segment by means of a satellite connection so that the safety, in case of unmanned platform, can be easily monitored from ground.


Conclusion

In this paper, a strategy for risk monitoring in a large multi-use offshore platform is presented. The approach involves a wireless sensor network with a star topology. Each node is equipped with a set of standard connections to guarantee the connectivity between the nodes and the sensors. All the information is gathered by a central node that stores the information in a database whose content is periodically sent to the ground segment. In addition, nodes can raise an alert in case of anomalous values that could be associated to a risk.
Major constraints for the network design are distances which are in the order of 10 km, the bandwidth of the network and the autonomy of the nodes. A trade-off solution has been achieved by using large range XBee modules working at 868 MHz (Europe) or 900 MHz (USA) together with the corresponding circuit to level the incoming signals from most of commercial sensors. Special emphasis is paid to risks associated with hydrogen leakages and flames.
In case of significant dimension increment of the platform, the network can be scaled with a reasonable low effort by replacing the star topology of the network by a tree topology in order to reduce the hop distance. In this case, several router nodes must be placed in the network. These nodes cannot be move to sleep state and, therefore, they require continuous connection to the network. Short range modules working in the 2.4 GHz ISM band with support for tree topologies must also replace the long-range modules.



Learning objectives
The learning objectives of this paper are focused on offshore platform monitoring with special emphasis in hydrogen leakage and flames detection. The main features of the network are the autonomy of the nodes, the large distances and the reduced bandwidth.


References
[1] www.h2ocean-project.eu/ . Funded by EC, grant agreement no: 288145.

[2] http://www.digi.com/xbee/

[3] X. Yu, C. Li, Z.N. Low, J. Lin, T.J. Anderson, H.T. Wang, F. Ren, Y.L. Wang, C.Y. Chang, S.J. Pearton, C.H. Hsu, A. Osinsky, A. Dabiran, P. Chow, C. Balaban, J. Painter, Wireless hydrogen sensor network using AlGaN/GaN high electron mobility transistor differential diode sensors, Sensors and Actuators B: Chemical, Volume 135, Issue 1, 10 December 2008, Pages 188-194.

[4] http://www.libelium.com