Managing Development, Construction and O&M for Offshore Wind Farm Projects
As offshore wind farm development, construction and O&M take place in increasingly challenging environments, it becomes ever more essential to harness new trends in IT to improve the safety and efficiency of the project.
By Simon Gulliver, Technical Manager, SeaPlanner, UK
Business Process Management
Because of the variety of consumers, there is a need to present data in different ways for different audiences. It is desirable to be able to present technical data in a way that makes it useful to a non-technical audience. Business processes can now be improved by working with this ever-increasing flow of data; there can be a tighter control of process, more effective use of planning tools, and ‘just-in-time’ strategies become more pervasive. This is particularly relevant offshore where costly plant and metocean constraints demand the optimisation of operations. The ability to monitor service delivery for reliability and cost-effectiveness is also desirable. Indeed, if a developer can provide the tools, their enhanced business processes will enable their suppliers to become more efficient, helping to keep the project on time and to budget.
Lessons Learnt Approach
As offshore work becomes more tightly managed, so too do Health and Safety processes. There is now more information available about project risks and how to employ best practice. This information should be available to everyone who might be affected to ensure effective active safety management. Availability of data also facilitates a ‘lessons learnt’ approach through trending and analysis of areas for concern. With increased availability of information corporate Health and Safety policy is more straightforward to propagate and implement, key performance indicators can be met and risks controlled.
Effective Data Management
So how do offshore wind farm owners manage this ongoing revolution? The first move is usually to buy more PCs to soak up the increased flow of data. This can be a costly mistake, and lots of boxes in lots of locations quickly become a management nightmare. Pre-digitally, old survey reports would be used as door stops; now they are likely to reside in a computer continually consuming resources, in a format that has become obsolete, on a PC that needs upgrading, often put there by a person who has since moved on, so that no-one now can find them anyway.
Perhaps then there is value in managing this through enterprise management systems and a central data centre? Again, there are risks if approached incautiously since this can all too easily have the potential to launch a complicated and very expensive rollout process that can require a complete rethink of existing business processes, resulting in long project timescales and high failure rates. Raw data is also not the natural territory of enterprise data systems and it can be very hard for an IT department to manage a wide variety of unfamiliar data feeds.
Central repositories do have their advantages; different datasets can be cross-indexed so that data that was gathered years ago for completely different purposes can be re-analysed for purposes which could never have been foreseen when the data was acquired.Bulk data is not easy to copy or to use; a highly scalable infrastructure is essential when data acquisition mounts exponentially.
Web Applications
There is, potentially, a very wide constituency for precious information and only the web provides the flexibility required. There should be no installation, no requirement for computer knowledge other than how to fire up a browser, nothing more complicated than a username and password to gain access. Network infrastructures and browser capabilities now support fully real-time web applications; our own SeaPlanner web application has been running 24/7 since 2008, refreshing the screen every 10 seconds.
A graphical presentation is a central pre-requisite of making effective use of highly technical data by revealing interactions and correlations. Graphical presentation allows humans to quickly absorb large amounts of data and spot patterns and inter-relationships. For example, interacting datasets for offshore wind farms can be as diverse as personnel and vessel movements, health and safety records, meteorological data, data from subsea surveys, operational scheduling patterns and SCADA data feeds.
In general, data correlations from different datasets are often either spatially or time-based. Geographical Information Systems (GIS) are tools designed to perform sophisticated analyses in these domains, and those same tools can then be used to provide a clear image to non-technical personnel.
Hosted web applications are very secure, certainly much more so than multiple uncontrolled data sources. The hosting sites will have almost military-grade security and access permissions, so those who attempt to log in can be closely controlled within a global policy.
Managing Data Flows
This is just the human side of the IT story; there is also the question of managing the data flows from a wide variety of sources and devices. It may not always be easy to choose protocols to distribute data through a computer network. In any case the data may flow through a series of connections; for example data might go through a V-24 serial link, a VHF data modem, an RS232 serial port and an Ethernet LAN before it reaches the internet. Major factors that need to be considered during the design phase are the pre-existence of protocols for the type of data source and the quantity and frequency with which that data is generated.
To get a real-world picture of data flows, an automated personnel tracking system might generate 50 records per day, per crew transfer vessel, a wave buoy might create 8,500 records a day, an AIS (Automatic Identification System) receiver at one location could generate around 100,000 records a day, and a sophisticated wave radar around 1 gigabyte per day.
Broadly speaking, the protocol of choice where data flow is not high and the end-to-end parties are both participants in the internet is web services; web services is an industry-standard protocol with wide support, its contents are XML-based and therefore human-readable, the message contents can be easily validated and web services cope well with different versions or partial subsets of data. Sender and recipient can handshake with each other and discover mutually acceptable protocol versions with standard data formats.
Web services are part of the same generic standards as web pages, with all the infrastructure support for hardware failover, integrated data compression, distributed processing in ‘The Cloud’ and universal interoperability with all computer systems. Web services will integrate easily with web-based data delivery.
Web services have the disadvantage of being extremely verbose, even when data is compressed, and they also require a full ‘protocol stack’ to generate them – which may well not be supported by devices with limited processing power. TCP/IP itself, which underpins web services, may not be an appropriate protocol for some transmission media, especially wireless transmission. In these circumstances the probable outcome is that custom binary protocols are implemented at the device level and the data format remains at this lowest common denominator throughout its journey to the recipient.
All protocols have a tendency to degrade, to acquire enhancements and extensions and vendor-specific modifications as the protocol gains general currency, but binary protocols are particularly problematic because they are neither human-readable nor self-documenting. If you are going to cross-correlate a mix of data sources, you must be able to have sufficient programming resource to interpret process and manage these data feeds.
Developing Web Applications
Application service providers have the ability to be extremely agile. As specialists whose sole purpose is to manage and enhance a specialised vertical market, they have the knowledge to incorporate the growing variety of devices and protocols, as well as publicly available datasets and open source software modules. To be successful in managing data from many different sources in a consistent and unified way, a data partner may be the best way to go.
Biography of the Author
Simon Gulliver has worked in IT for over 30 years, developing ground-breaking technologies with leading implementations including CD-ROM databases, digital video, Bluetooth, mobile e-mail and hosted web applications. He designed and developed the SeaPlanner system specifically for offshore renewables.{/access}
As offshore wind farm development, construction and O&M take place in increasingly challenging environments, it becomes ever more essential to harness new trends in IT to improve the safety and efficiency of the project.By Simon Gulliver, Technical Manager, SeaPlanner, UK
{access view=!registered}Only logged in users can view the full text of the article.{/access}{access view=registered}Real-Time Data
As always, IT trends are driven by cheaper and more powerful hardware and communications infrastructure. Increasingly, offshore there are more vessels with sophisticated navigation and electronic equipment, more automated devices with continuous feeds, and more individuals with laptops, smartphones, satphones and PDAs. All of these are generating substantial amounts of data, often in real time. The existence and efficient use of this data demands a different way of working for developers, owners and contractors as they become more heavily involved in data capture and data utilisation, especially where ownership of projects is shared between many different organisations. There is a need for information and documentation to be directly available to all relevant parties and this information is increasingly worked on in a collaborative way.Business Process Management
Because of the variety of consumers, there is a need to present data in different ways for different audiences. It is desirable to be able to present technical data in a way that makes it useful to a non-technical audience. Business processes can now be improved by working with this ever-increasing flow of data; there can be a tighter control of process, more effective use of planning tools, and ‘just-in-time’ strategies become more pervasive. This is particularly relevant offshore where costly plant and metocean constraints demand the optimisation of operations. The ability to monitor service delivery for reliability and cost-effectiveness is also desirable. Indeed, if a developer can provide the tools, their enhanced business processes will enable their suppliers to become more efficient, helping to keep the project on time and to budget.
Lessons Learnt Approach
As offshore work becomes more tightly managed, so too do Health and Safety processes. There is now more information available about project risks and how to employ best practice. This information should be available to everyone who might be affected to ensure effective active safety management. Availability of data also facilitates a ‘lessons learnt’ approach through trending and analysis of areas for concern. With increased availability of information corporate Health and Safety policy is more straightforward to propagate and implement, key performance indicators can be met and risks controlled.
Effective Data Management
So how do offshore wind farm owners manage this ongoing revolution? The first move is usually to buy more PCs to soak up the increased flow of data. This can be a costly mistake, and lots of boxes in lots of locations quickly become a management nightmare. Pre-digitally, old survey reports would be used as door stops; now they are likely to reside in a computer continually consuming resources, in a format that has become obsolete, on a PC that needs upgrading, often put there by a person who has since moved on, so that no-one now can find them anyway.
Perhaps then there is value in managing this through enterprise management systems and a central data centre? Again, there are risks if approached incautiously since this can all too easily have the potential to launch a complicated and very expensive rollout process that can require a complete rethink of existing business processes, resulting in long project timescales and high failure rates. Raw data is also not the natural territory of enterprise data systems and it can be very hard for an IT department to manage a wide variety of unfamiliar data feeds.
Central repositories do have their advantages; different datasets can be cross-indexed so that data that was gathered years ago for completely different purposes can be re-analysed for purposes which could never have been foreseen when the data was acquired.Bulk data is not easy to copy or to use; a highly scalable infrastructure is essential when data acquisition mounts exponentially.
Web Applications
There is, potentially, a very wide constituency for precious information and only the web provides the flexibility required. There should be no installation, no requirement for computer knowledge other than how to fire up a browser, nothing more complicated than a username and password to gain access. Network infrastructures and browser capabilities now support fully real-time web applications; our own SeaPlanner web application has been running 24/7 since 2008, refreshing the screen every 10 seconds.
A graphical presentation is a central pre-requisite of making effective use of highly technical data by revealing interactions and correlations. Graphical presentation allows humans to quickly absorb large amounts of data and spot patterns and inter-relationships. For example, interacting datasets for offshore wind farms can be as diverse as personnel and vessel movements, health and safety records, meteorological data, data from subsea surveys, operational scheduling patterns and SCADA data feeds.
In general, data correlations from different datasets are often either spatially or time-based. Geographical Information Systems (GIS) are tools designed to perform sophisticated analyses in these domains, and those same tools can then be used to provide a clear image to non-technical personnel.
Hosted web applications are very secure, certainly much more so than multiple uncontrolled data sources. The hosting sites will have almost military-grade security and access permissions, so those who attempt to log in can be closely controlled within a global policy.
Managing Data Flows
This is just the human side of the IT story; there is also the question of managing the data flows from a wide variety of sources and devices. It may not always be easy to choose protocols to distribute data through a computer network. In any case the data may flow through a series of connections; for example data might go through a V-24 serial link, a VHF data modem, an RS232 serial port and an Ethernet LAN before it reaches the internet. Major factors that need to be considered during the design phase are the pre-existence of protocols for the type of data source and the quantity and frequency with which that data is generated.
To get a real-world picture of data flows, an automated personnel tracking system might generate 50 records per day, per crew transfer vessel, a wave buoy might create 8,500 records a day, an AIS (Automatic Identification System) receiver at one location could generate around 100,000 records a day, and a sophisticated wave radar around 1 gigabyte per day.
Broadly speaking, the protocol of choice where data flow is not high and the end-to-end parties are both participants in the internet is web services; web services is an industry-standard protocol with wide support, its contents are XML-based and therefore human-readable, the message contents can be easily validated and web services cope well with different versions or partial subsets of data. Sender and recipient can handshake with each other and discover mutually acceptable protocol versions with standard data formats.
Web services are part of the same generic standards as web pages, with all the infrastructure support for hardware failover, integrated data compression, distributed processing in ‘The Cloud’ and universal interoperability with all computer systems. Web services will integrate easily with web-based data delivery.
Web services have the disadvantage of being extremely verbose, even when data is compressed, and they also require a full ‘protocol stack’ to generate them – which may well not be supported by devices with limited processing power. TCP/IP itself, which underpins web services, may not be an appropriate protocol for some transmission media, especially wireless transmission. In these circumstances the probable outcome is that custom binary protocols are implemented at the device level and the data format remains at this lowest common denominator throughout its journey to the recipient.
All protocols have a tendency to degrade, to acquire enhancements and extensions and vendor-specific modifications as the protocol gains general currency, but binary protocols are particularly problematic because they are neither human-readable nor self-documenting. If you are going to cross-correlate a mix of data sources, you must be able to have sufficient programming resource to interpret process and manage these data feeds.
Developing Web Applications
Application service providers have the ability to be extremely agile. As specialists whose sole purpose is to manage and enhance a specialised vertical market, they have the knowledge to incorporate the growing variety of devices and protocols, as well as publicly available datasets and open source software modules. To be successful in managing data from many different sources in a consistent and unified way, a data partner may be the best way to go.
Biography of the Author
Simon Gulliver has worked in IT for over 30 years, developing ground-breaking technologies with leading implementations including CD-ROM databases, digital video, Bluetooth, mobile e-mail and hosted web applications. He designed and developed the SeaPlanner system specifically for offshore renewables.{/access}
