Microsoft word - ginseng-petrobras.docx

SEVENTH FRAMEWORK PROGRAMME
ICT Call 2
FP7-ICT-2007-2

Grant agreement for: Small or medium-scale focused research project (STREP1)
proposal

Annex I - “Description of Work”
Project acronym: GINSENG Project full title: Performance Control in Wireless Sensor Networks Grant agreement no.: 224282 Date of preparation of Annex I (latest version): 5.2.2008 Date of approval of Annex I by Commission: (to be completed by Commission) List of Beneficiaries
Beneficiary
Beneficiary name
Beneficiary
Date enter Date
short name
project**
project**
University College Cork Department of Informatics Computer Science AB Technische Universitaet Carolo- Wilhelmina zu Braunschweig University of Cyprus 8 * Please use the same beneficiary numbering as that used in the Grant Agreement Preparation Forms ** Normally insert “month 1 (start of project)” and “month n (end of project)” 1 Specific International Cooperation Actions – SICAs –use the standard STREP Part B template TABLE OF CONTENTS
BUDGET BREAKDOWN AND PROJECT SUMMARY
B 1. CONCEPT AND OBJECTIVES, PROGRESS BEYOND STATE-OF-THE-
ART, S/T METHODOLOGY AND WORK PLAN
Concept and objectives
1.1.3. The GINSENG Approach for Performance Control in Wireless Sensor Networks 1.1.6. Relationship to the Call for Proposals Beneficiary
Beneficiary name
Beneficiary
short name
Computer Science AB Technische Universitaet Carolo- Wilhelmina zu Braunschweig University of Cyprus B 1.
Concept and objectives, progress beyond state-of-the-art, S/T
methodology and work plan
1.1.
Concept and objectives
1.1.1. Background and Motivation
A node in a wireless sensor network (WSN) is a small embedded computing device that interfaces with sensors/actuators and communicates using short-range wireless. Such nodes act autonomously but cooperatively to form a logical network in which data packets are routed hop-by-hop towards management nodes, typically called sinks or base stations. A WSN comprises a potentially large set of nodes that may be distributed over a wide geographical area, indoor or outdoor. Wireless sensor networks enable numerous sensing and monitoring services in areas of vital importance such as efficient industry production, safety and security at home, and in traffic and environmental monitoring. In 1999, Business Week chose wireless sensor networks as one of the 21 most important technologies for the 21st century, while Technology Review ranked it among 10 emerging technologies that will change the world. Also, the market potential of this technology is enormous. The analysis company ON World estimates that by 2010 168 million nodes could be deployed. They also predict that the market for wireless sensor networks will increase from ca. 400 million Euros in 2006 to about 9000 million Euros in 2010. Another ON World study from July 2005, based on a survey with OEMs and platform providers, ranked reliability (or rather the lack of reliability) as the major adoption inhibitor for wireless sensor networks. Despite the utmost importance of reliability and performance control, these types of topics have not received enough attention from the research community. One of the reasons might be that most US-based research on WSNs to date has been heavily influenced by military application scenarios in which dense networks of sensors are expected to be deployed in a random manner, primarily with a view to tracking objects. However in most other application domains it is the case that sensor networks will need to be deployed in a planned manner, close to selected locations/objects of interest. For example, in industrial environments is it clear that there will be selected machinery, pipes, doorways, etc that will need to be monitored and controlled. A planned WSN raises end-user expectations in terms of the service level to be provided, both in terms of the reliability of the network and also its latency in responding to events. It corresponds well to the practical needs of industry, which currently rely on wired technologies with well-established performance levels to achieve plant monitoring and control. For such industries to adopt a solution using WSN technology should result in significant savings in deployment and maintenance costs, and can be more easily reconfigured and rapidly deployed to adapt to changing business needs. The GINSENG proposal had its origins in the E-NEXT Network of Excellence, to which most of the partners were active contributors. 1.1.2. Vision and Goal
Our vision is based on wireless sensor networks in which deployment is planned rather than random, and in which performance can be controlled to ensure operation within specific application service bounds for latency and reliability. In adopting this position we reject the more widely accepted assumption that WSNs are entirely self-configuring and can achieve robustness by virtue of very high levels of redundancy. Such unplanned networks are unsuitable for most industrial and environmental applications, and are unable to offer performance assurances. Our goal is a wireless sensor network that will meet application-specific performance targets, that will integrate with industry resource management systems, and that will be proven in a real industry setting where performance is critical. 1.1.3. The GINSENG Approach for Performance Control in Wireless Sensor
Networks
Our vision of performance controlled wireless sensor networks demands a new and cohesive approach to the design and management of WSNs, with solutions to a specific set of new scientific research challenges in three areas: network design, network operation, and system integration. Figure 1 depicts the GINSING approach and its main components. In contrast to applications that are based on random deployment, GINSING assumes a planned and careful deployment of the sensor nodes as a basis to achieve performance control. The second basis of GINSENG are software components with assured performance including operating systems that execute tasks within a given time and predictable access to the radio medium by means of a MAC layer that enables access to the radio medium within a certain time bound. The third basis of GINSENG is a set of algorithms that ensure control with respect to network topology and traffic. These three components enable the possibility to deploy sensor networks with assured performance. Due to the inherent uncertainties in e.g. node availability and the radio medium, it is possible for undesired changes in the operating environment, motivating the need to monitor and potentially debug the performance of the deployed system. GINSENG will provide mechanisms and tools to perform performance debugging of deployed systems and reconfigure when given performance metrics can no longer be achieved. Industry does not accept additional systems that do not interface with existing equipment and therefore one of the objectives of GINSENG is the integration with industry IT systems, a proposition strengthened by the participation of SAP as a partner. The applicability of the technology developed in GINSENG will be proven by developing a wireless sensor network for a real-world application where performance is critical – in the context of an oil refinery run by GINSENG partner GALP. 1.1.4. Scientific Objectives
The overall scientific objectives of GINSENG are to provide WSN technology that: 1. meets application-specific performance targets, 2. integrates with industry resource management systems, 3. is proven it in a real industrial setting where performance is critical. Therefore, the following specific objectives are targeted: 1. Identification of the constraints and factors affecting the provision of assured 2. Investigation of different approaches for providing controlled performance, such as topology control, overload control and resource allocation mechanisms aimed to keeping the WSN operational within the specified bounds 3. Devising topology control and traffic control mechanisms to keep the WSN operational within the specified bounds. 4. Development of software components offering assured performance while optimizing energy consumption. These components include: a. real-time node operating systems. b. MAC techniques providing assured access to the radio medium. c. Congestion control and traffic management. 5. Design WSN planning and deployment methods as well as mechanisms and tools for performance debugging and reconfiguration of deployed systems. 6. Integration of such performance controlled WSN with industry resource management systems by development of a middleware to bridge the gap between low level monitoring and backend application systems. 7. Realization of prototypes allowing for demonstration and evaluation in a refinery 1.1.5. Application Context
The selected application domain for the proposal is devoted to monitoring and control of industrial processes, safety and pollution supervision in the oil industry. In that context we propose to explore the need for performance-controlled wireless sensor networks in industry, and to demonstrate the efficacy of the research solutions that we derive. In pursuing our scientific objectives the choice of an application domain will serve mainly to guide and influence the research assumptions, but ultimately the fruits of our research can be extended to other industrial settings where performance–controlled monitoring and safety are particularly important. The Petroleum Industry in the European Economic Context
The petroleum industry is undoubtedly a key component of the European economy. It not
only provides fuels for transportation and heating, but supplies raw materials, such as paints
or plastics, for the chemical industry. According to the BP Statistical Review of World
Energy of July 2006, European Union consumption accounts for 18.2% of the world
consumption, which is roughly 84 million barrels a day. This makes the petroleum industry a
major contributor to the EU-25 gross product. However, Europe’s high structural dependence
on imported fossils puts at risk not just the security of supply but also prices, given global consumption levels. In the next decades, the petroleum industry must prepare to address and materialize a number of critical challenges. Environmental concerns, EU regulations and policies, competitive forces, process improvements or energy efficiency, just to name a few, are regarded as major forces for change. To meet these challenges and maintain the profitability of the petroleum industry Research and Development will be needed. Petroleum refining has grown increasingly complex in the last two decades, mostly due to lower-quality crude oil, crude oil price volatility and more recently to environmental regulations that require manufacturing processes and higher performance products. In terms of energy consumption in the refinery, atmospheric and vacuum distillation account for around 40% of the total process energy consumed. Another high energy consumer is hydrotreating which accounts for 20%. In this process sulphur, nitrogen and metal contaminants are removed from feeds. In the last few decades, refineries have been engaged in developing energy efficiency policies and practices, such as plant heat integration, recovery of waste heat and implementing effective energy efficiency oriented programs. Given the volubility of oil prices, environmental costs and decreasing margins, refineries will look to improvements in overall energy efficiency to increase the profitability, by lowering operating costs. Advances in technology are a compelling choice for improving the way energy is used, particularly in high energy-intensive processes. Safety and Environmental Performance
Concerning safety and environmental conservation, the refining industry is heavily regulated.
The manufacturing processes used to produce petroleum products generate a variety of air
emissions and other residuals, being some of them hazardous and/or toxic chemicals.
Additionally, refineries also produce process waste-water which consists of surface water
runoff, cooling water and process water. Waste-waters are treated in water treatment facilities
and subsequently discharged to public water treatment plants or, under permit, to surface
waters. The main sources of air emissions in refineries are the combustion emissions related
to the burning of fuels, including the emissions associated with electricity generation,
equipment leak, process vent, storage tank and waste-water. The main goal is to integrate into
the production side environmental concerns, such as balancing sulphur in the refinery from
crude to products. However, to support continuous improvements in terms of environmental
performance, better sensing and monitoring systems are required to enable the optimization of
process variables, monitor emissions as they arrive, and activate effective fault tolerant
optimal control policies to correct the situation.
Inspection methodologies and routines play a critical role in the overall energy, economic,
safety, reliability and environmental performance of refining industries. Effective inspection
of equipment is vital to the construction and safe operation of distillation equipment, furnaces,
heat exchangers, piping systems and other unit operations. Testing and robust monitoring of
equipment integrity is essential to plant safety and optimum reliability. The tendency is to rely
on online non-invasive inspection techniques, in order to allow immediate detection of loss
containment and provide early warnings for corrosion and potential flaws regarding the
structural integrity of equipment. Additionally, inspections should be conducted
automatically, without human intervention, enabling complete knowledge of equipment conditions in real-time. The Refining Unit
Petroleum refineries are complex systems specifically designed based on the desired products
and the properties of the crude oil feedstock. Refineries may range from medium integrated
refineries to fully integrated refineries (or total conversion refineries), based on the use of
different processing units.
The refinery feedstock is crude oil, which is a mixture of hydrocarbon compounds. The
hydrocarbons in crude oil are a mixture of three chemical groups including paraffins (normal
and Bitumen Production Unit isoparaffins), naphthenes, and aromatics.
Typical processing units in refineries include: Desalting, Primary Distillation, Bitumen
Production, Hydrogen Consuming Processes, Pretreating and Catalytic Reformer, Catalytic
Cracking, Gas Plant, Etherification, Alkylation, Polymerization, Coking, Visbreaking, Lube
Oil Production, Gas Treatment and Sulfur Recovery Units.
There are significant health and safety issues regarding many different components of the
refinery. These concern procedures, infrastructure and risks for which monitoring and
following safety procedures and standards are crucial. Some of the most relevant safety
concerns are listed next:
1. Fire Protection - fire protection issues and steps, including monitoring, are needed to promote the safe storage, handling and processing of petroleum and petroleum products in refineries; 2. Protection Against Ignitions Arising Out of Static, Lightning, and Stray Currents – methods are necessary for assessing, monitoring and minimizing the potential ignition hazards associated with static electricity, lightning and stray currents in petroleum industry operations, including marketing terminals, transportation and storage of flammable liquids; 3. Safe Welding, Cutting, and Hot Work Practices in the Petroleum and Petrochemical Industries – safety procedures are necessary for use in and around petroleum and petrochemical operations to help prevent injuries and property damage as a result of fire during gas and electric cutting and welding activities; 4. Monitoring and Management of Atmospheric Storage Tanks for Fire Prevention and Protection – it is necessary to monitor and plan for fire hazards in atmospheric storage tanks containing flammable and combustible materials; 5. Safe Storage and Handling of Heated Petroleum-Derived Asphalt Products and Crude- Oil Residua – potential hazards related to these products and residua should be understood, and this knowledge should be applied in order to help reduce incident probability and severity; 6. Flame Arresters in Piping and other Systems – Flame arresters including Water Spray Systems should be in place and include monitoring and immediate action for Fire Protection; 7. Ignition Risk of Hydrocarbon Vapours by Hot Surfaces in the Open Air – given the ignition risks of these vapours, measures should be taken to sense, monitor and plan for the risks arising from these. Robust Monitoring Needs in the Petroleum Industry
Continuous combustible and toxic gas monitoring is a critical facet of refineries. There are
many processes and special production units in a refinery that pose specific safety hazards.
These include:
1. Crude Desalting - The potential exists for a fire due to a leak or release of crude from heaters in the crude desalting unit. Low boiling point components of crude may also be released if a leak exists. These are closed processes, however, heaters and exchangers in the atmospheric and vacuum distillation units could provide a source of ignition, and the potential for a fire exists should a leak occur. 2. Thermal cracking, coking, and catalytic cracking - These are some other closed process with the potential for fire coming from the leakage of liquids, gases, or vapors that come into contact with an ignition source. 3. Catalytic dust - Explosive concentrations of catalyst dust can accumulate during its recharge or disposal. The handling of coked catalyst creates the possibility for iron sulphide fires, which can occur when iron sulphide ignites spontaneously in air. 4. Hydrogen generation - Hydrogen generation is required to provide for a continuous supply. This creates a hazard in the event of a leak or release of product or hydrogen gas. 5. Hydrogen Sulphide (H2S) - The hydrogen sulphide content of the feedstock must be continuously monitored to prevent personnel exposure to toxic concentrations, reduce corrosion, and prevent environmental pollution. 6. Isomerization - Isomerization processes convert n-butane, n-pentane and n-hexane into their respective isoparaffins of substantially higher octane number. This is another closed process with hazardous implications in the form of leaks coming into contact with an ignition source. 7. Sweetening - Air or oxygen is used in sweetening processes. If too much oxygen enters these processes, it is possible for a fire to ignite in the settler due to the generation of static electricity. Benefits From Wireless Sensor Networks
Large industrial sites, such as oil refineries, already have complex process-control systems in
place, but there are many additional points that could provide additional valuable data to
optimize processes and enforce safety. Monitoring involves checking the status of key motors,
valves, pumps, and supporting process variables. Currently, signals acquisition and control
actions are automatically delivered/provided using wired data communication networks. In
many senses this hardware solution has proven to work reasonably well, particularly when all
the system is planned from the beginning. However, when new actuators and sensors or
surveillance systems are required for a given task and there are no wired data communication
networks available then wireless solutions will be a valuable choice, reducing costs and time of implementation. Wireless sensor networks allow industry to collect information with more monitoring points, providing awareness into the environmental conditions that affect overall uptime, safety, or compliance in industrial environments and enabling agile and flexible monitoring and control systems. Wireless sensor networks connect critical processes or assets with the systems or experts that can interpret the data or take immediate action. Wireless Sensor Network in the Field
An area where wireless networks are very welcome is the monitoring and surveillance in long
range pipelines where leak diagnosis and intrusion detection are critical, being in this case
wireless technology more reliable and with lower installation costs. Furthermore, by their
particular nature, refineries make use of inflammable materials that could be exposed to direct
sunlight during the daytime, putting in danger the safety of plant, as is the case of sulphur
storage area and the related recovery unit. In this application, wireless sensor networks could
provide adaptive monitoring capabilities with the flexibility needed in this field. Additionally,
wireless sensor networks are adaptable to changes in both the configuration of equipment on
the plant floor and in the layout of the network itself, which is a key value for hazardous
monitoring in sulphur storage environments. This same utility can be exploited to support
remote monitoring in dangerous environments (e.g. fuel and gas storage tanks) under
maintenance, using a set of sensors to measure butane or propane gas levels.
Finally, wireless sensor networks can be applied in monitoring of environmental impacts such
as in waste-waters treatment facilities or air emissions allowing a proactive implementation of
a social responsibility culture.
1.1.6. Relationship to the Call for Proposals
This proposal relates to all three parts of the Target Outcomes for Objective ICT-2007.3.7, but primarily to the outcome entitled “Cooperating Objects and Wireless Sensor Networks”. In that respect what we propose is a set of cooperating objects (sensor nodes) that work together to ensure that application-specific tasks are achieved within certain performance bounds. These nodes are in spatial proximity by virtue of being in the same factory or industrial plant. Such a system enables development and deployment of ambitious future applications in which reacting to the physical world is performance-critical for safety and other reasons. Our objectives include sophisticated algorithms for resource control and management, software components in terms of operating system support as well as communication and media access protocols for optimal/predictable execution, and programming abstractions embedded in an integrated middleware system – clearly matching with most of the items listed for this outcome. Middleware is the main focus of target outcome a) with the emphasis on enabling predictability and QoS awareness – exactly the target of GINSENG. Important aspects in GINSENG are dynamic reconfiguration when performance goals are not met as well as minimal power consumption of the sensor nodes’ software components. Target outcome c) entitled “Control of large-scale complex distributed systems” refers explicitly to predictable and safe infrastructures for energy production and distribution – an exact match with our selected application domain of process, safety and pollution monitoring and control in the oil industry. The objective of GINSENG’s performance control is to master uncertainties in delay and bandwidth by means such as controlled deployments and dedicated research into software components with assured performance. Our proposal relates to all four of the stated Expected Impacts for this Objective. We propose to use wireless sensor networks to monitor and control the environmental impact of oil refinery plants, improving safety, maintenance and effectiveness (through reconfiguration and rapid deployment) of such large industrial processing plants. By solving the scientific questions that are necessary to produce a performance controlled wireless sensor network we will enable new applications and open up new market opportunities for this technology. In particular, the fact that our vision for planned WSNs contrasts with that dominating most US-based research inspires us to believe we will operate in a domain that can provide European industry with a leading edge in an important emerging market.

Source: http://saloon.inf.ufrgs.br/twiki-data/Projetos/MW4R/WebHome/GINSENG-PetroBras.pdf