Getting to grips with offshore wind turbines

In November last year Prime Minister David Cameron declared that the UK was a world leader in wind energy. Only days later, however, energy giant RWE pulled out of its planned wind farm developments in the Bristol and Devon Channel. The 240-turbine Atlantic Array project would have been capable of producing 1,200 MW of electricity, but the company shelved the plan because of technological challenges and market conditions. Just a few weeks later, changes to the Government's energy policy were announced. Although there is to be no change in overall spending, more financial backing is to be given to offshore wind power at the expense of solar power and onshore wind power generation. The aim is to give energy companies and private investors some stability in the market and show that Government is prepared to provide financial incentive and support to large offshore wind projects. Politics aside, the technical challenges surrounding offshore wind are numerous. The open ocean is a challenging environment and the coastline seabed makes the installation of turbine foundations a difficult task. This is partucularlty the case given the size of many offshore wind turbines, which can have a tower height and even blade diameter in excess of 100m. Building foundations There are many different methods of installing suitable foundations to mount and attach wind turbine towers with the depth of water and size of the turbine key factors for consideration. Methods for setting up suitable foundations in deeper waters are still being developed, with the possibility of floating platforms being explored. Most existing methods of building a solid foundation deal with water depths of less than 80m. From around 40m upwards, preferred methods include gravity base structures, tripod piled type structures, and more conventional steel jacket structures like those commonly used in the oil and gas industry. These methods have the advantage of having more than one anchor point to the seabed, making levelling more straightforward by allowing adjustment after installation. However, when a single column monopile is used – standard for sea depths up to 40m – the columns must remain vertical as they are being driven into the seabed with a hydraulic hammer. Any deviation will make the overall structure lean. And when you multiply this by the size of the towers, the effect is not only noticeable, it can render the entire structure unsafe and useless. There is then no option but to remove these monopiles and redrive them at great expense. This was the challenge posed to marine design and engineering expert, Houlder. The company was contracted to provide a turnkey solution for MPI Offshore, and lead the design, fabrication, installation and commissioning of a suitable method of holding 700-tonne steel monopile foundations in place as they are hammered into the seabed. The solution came in the form of massive hydraulic gripper arms that hold in place the 75m long and 7m diameter piles as they are driven 20m into the seabed. The concept is much the same as using one's hand to hold a nail straight while hammering it in. However, this is on a massive scale, with six giant hydraulic cylinders positioning two clamping claws, each weighing 70 tonnes. These hang over the stern of a purpose built 'jack-up' vessel called MPI Discovery. MPI Discovery can carry up to four monopiles at any one time. Once in position, the jack-up vessel extends it legs down to the seabed to lift its entire hull out of the water. At this point a monopile is lifted up by an onboard crane and lowered in to the water. It is then hammered down in to the seabed using a massive hydraulic ram over the period of a few hours. Keeping the piles vertical in the presence of currents and waves is a serious challenge for Houlder and pinpoint accuracy was needed to control the giant mechanical device. Three pairs of hydraulic cylinders actuate each gripper arm, with one cylinder raising and lowering the arms from a stowed vertical position, while the other two provide motion on the x and y axes. "The gripper arms are an ingenious method of keeping 700-tonne piles vertical by providing a horizontal restraint against environmental forces," says Paul Shaw, marine equipment director at Houlder. "This will significantly improve the efficiency of wind farm installations and overall productivity." Providing the motion control aspect for the project was the job of Glasgow-based Industrial Systems and Control (ISC). The company has a history of control design, particularly for offshore systems such as wind turbines. The team was set a very tight time constraint of just four months to go from initial specification to finalising and delivering the control system, as well as developing effective testing algorithms known as an emulator to see how the control functions of the grippers would perform and react given certain inputs. Dr Andy Clegg, managing director of ISC, says: "One of the key features of this device is that, even though these mechanisms are very slow, they are also very powerful. So, we actually had to include quite a lot of quite careful kinematic constrains to prevent the arms from having the possibility of crushing the pylon or causing them to buckle." ISC developed the control system using a National Instruments NI CompactRIO reconfigurable control and monitoring system with NI LabView software. The NI CompactRIO uses open embedded architecture. Despite its small size it is both powerful enough to be a standalone systems and rugged enough to be used in the offshore environment. Indeed, the CompactRIO controller is approved by Det Norske Veritas (DNV) for safety and meets all the necessary rules for marine operations. The control system was configured to perform real-time data acquisition and is able to monitor and position all six hydraulic cylinders to control the precise movement of the gripper arms. ISC could even individually or synchronously steer the arms to adjust for the vertical inclination of the piles. The use of LabView allowed ISC both the computational power required along with the low level control and kinematic calculations necessary for the application. In addition, its proven reliability both in the software and control electronics was crucial given the safety-critical nature of the application. ISC was able to perform factory commissioning and testing using the available parts of the real system, while simulating pieces that were actually not physically present. This allowed it to then analyse and optimise the performance of the individual cylinder control loops, as well as the overall x-y motion of the gripper arms themselves. "This allowed us to fully test the operational software in the office," says Dr Clegg. "We can test it based on both normal operations and we can also inject fault scenarios to make sure the full functionality of the software works. "This was really critical as these boats are almost on a continuous charter and they have got very expensive day rates. That was why the amount of time we had to commission the software, implement it and get it working was so tight. Even if we had a one day delay, it would be very costly. That is why building the emulator, which is almost as big as the software application itself, was so important, as it would allow us to avoid any problems during the actual implementation." The interface and control of the arms by an operator is achieved using a chest pack that has a joystick controller to move the individual grippers as needed, depending on what is being attempted. Operational logic, monitoring and fault actions are all executed on the CompactRIO and accessed through a touch panel computer (TPC) with the joystick and a set of control buttons all integrated into the operator's chest pack. The TPC is used to provide the operator interface and was also built using LabVIEW. However, behind the chest pack is an enormous number of electronic connections from the gripper arm to the CompactRIO, many of which are monitoring and safety systems. The CompactRIO embedded controller is equipped with numerous Input/Output (I/O) modules to interface with the sensors, actuators and the operator chest pack. The structure of the software included the main real-time (RT) application, an FPGA program including a watchdog, with the HMI software running on the TPC, and the PC-based emulator used during development. The system was tested in August last year, with the first pile being loaded onto MPI Discovery from Vlissingen in Holland. The vessel then set sail for the Humber Gateway wind farm, 8km off the coast of Grimsby, where it was to be fully tested by installing monopiles in the sea. Due to the expensive charter rates for the MPI Discovery there was no time for a test run and commissioning would be done based on the performance on the first set of foundation installations. The Humber Gateway has 18m deep waters and thus acted as an ideal test case for the evaluation of the gripper arms and the control system. Dr Clegg says: "Considering this was the very first monopile to be installed, and this is the first time the software had been used in anger, it was a tense time for us as it would prove whether the system was really working or not." Happily, the system worked without a hitch and it was able to move the piles as required and hold them perfectly in place, as the hydraulic hammer stamped them 20m in to the ground until the top of the monopile was just a few metres from the water's surface. Since the first installation in August, the MPI Discovery has been working continuously and has completed more than 20 monopile foundation installations, which are now ready for a turbine tower to be installed. However, with the array needing a total of 77 turbines installed by 2015, there is still plenty of work to do.