Central Networks uses wind velocity and direction to determine dynamic line ratings.
Wind farms often tend to be located at the extremes of the system where the existing overhead line circuits have insufficient load-transfer capacity to carry the full output of the wind farm. Originally designed to supply relatively small loads, the connection of wind generation may result in a large reverse power flow in excess of the summer and winter static ratings of the existing circuit.
Faced with the connection of existing and proposed wind generation comprising 180 MW offshore and 50 MW to 90 MW onshore, the existing 132-kV, 40-km (24.9-mile) double-circuit overhead transmission line between Skegness and Boston in northeastern England required a load management and protection system to enable the acceptance of this large penetration of wind generation.
Transmission Line Monitoring
The thermal capacity of an overhead line is the maximum current the circuit can carry without infringing on the statutory conductor-to-ground clearances and the annealing onset temperature of the conductor. The majority of utilities monitor the power flow in overhead lines without any real-time data on conductor temperature or sag. There are many climatic variables such as wind speed and direction, ambient temperature and solar radiation that affect the conductor temperature.
The adoption of real-time monitoring allows utilities to achieve a higher use of the load-transfer capacity or ampacity of the line while ensuring all statutory ground clearances are maintained. Several proprietary real-time monitoring systems are available based on two fundamentally different methods to determine the ampacity. One method is by direct measurement using sensors to determine conductor tension, conductor temperature or sag. Alternatively, an indirect approach can be taken by measuring ambient weather conditions from which the real-time ampacity can be calculated.
Various computational methods have been developed in the past to calculate the heat transfer and ampacities of overhead line conductors, but the two most frequently used international standards are those published by CIGRÉ and IEEE for the current-temperature relationship of overhead line conductors.
Selection of the Dynamic Line Protection
AREVA T&D UK, now Alstom Grid UK, developed the dynamic line rating (DLR) relay based on the CIGRÉ equations as part of an Innovation Funding Incentive (IFI) research and development project. Two prototypes were commissioned for evaluation on Central Networks' transmission system at its 132-kV Skegness Substation in 2008. (Central Networks, the electricity distribution business for the Midlands, was acquired by PPL Corp., the parent company of Western Power Distribution on April 1, 2011.)
The DLR is an enhancement of the existing protection functions, with both overcurrent and earth fault protection. The relay allows the user to select the type and current loop input channels to be used for wind speed, wind direction, solar radiation and ambient temperature monitoring. The results are fed into an algorithm that implements the DLR calculations. Three-phase currents are measured and the phase with maximum current is selected as the relaying quantity for the alarm and tripping criteria. The current magnitudes and sensor measurements together with the calculated ampacity are available from the relay as measuring quantities.
Six DLR stages of protection are available, each consisting of its own threshold level and time-delay settings. In configuring the relay, in addition to setting the DLR thresholds and time delays, it is necessary to enter a range of conductor parameters used in the heating and cooling calculations.
Skegness-Boston DLR Scheme
As a result of a proposed increase in wind generation at Skegness, Central Networks has installed a DLR system as a reverse power flow on the 132-kV Skegness-Boston double-circuit overhead line. The load-transfer capacity of the existing Lynx conductors on this circuit could exceed the U.K. Engineering Recommendation P27 winter and summer current ratings of 539 A and 433 A, respectively.
Central Networks has opted to apply the dynamic rating algorithms to its control centre and protection relays using ambient temperature and wind speed measurements. For validation reasons, wind direction and solar radiation also are measured. Two weather stations are connected at Skegness, providing redundancy for the relay, and one weather station is connected at Boston, with all three being connected to the management system.
Power Donuts, manufactured by USi, are installed at Skegness and Boston and at the midpoint of the line to monitor the temperature variations along the line and validate the DLR algorithms. This installation has been in commission for more than a year, during which time data has been accumulated and analyzed. When the measured line current reaches a certain percentage of the dynamically calculated ampacity, the first action is for the load management system to send a signal to the generators to reduce their power output. As a backup to the load management system, the protection relay issues a signal to trip the circuit breaker connecting the wind farm to the Skegness-Boston overhead line after a predetermined time delay. The threshold time delay and trip level can be set within the relay to coordinate with the load management system.
The purpose of the threshold time delay is to avoid spurious tripping during temporary or transient network faults and to provide a means of grading with other protection and control equipment on the network.
Prior to the installation of the DLR, several assumptions and conditions specific to this particular project site had to be agreed.
Wind direction is variable and difficult to take into account because of the changes in direction of the Skegness-Boston overhead line. A simplified approach was taken linking the wind speed measurement to the wind direction factor of sine 20° = 0.34. The factor (sine 20°) is based on the assumption that, if the wind direction is less than 20°, the cooling effect as a result of wind turbulence is roughly as high as when there is no turbulence with a 20° wind direction. Following multiplication by the direction factor, the derived wind speed is limited with a lower limit of 0.5 m/sec (1.64 ft/sec). The resulting limited wind value is used for calculating the convective cooling. Finally, solar radiation is taken into account using a conservative approach that assumes no clouds.
In practice, the dynamically calculated ampacity value is limited by lower and upper limits that can be set. The upper limit can take into account other circuit constraints, for example, the ratings of cables, joints or other circuit components.
Commissioning and Data Analysis
Two relays, each installed in wall-mounted cabinets, were commissioned in Skegness in March and April 2008. Each cabinet contains a data logger to capture data from the weather stations and the inputs from the relays the data is being captured for a period of several months.
The data analyzed showed good correlation between the data from the relay and the data from the Power Donut at the Skegness end of the line. The derived conductor temperature is an iterative calculation based on measurements of ambient temperature, average wind speed, wind direction, solar radiation and line current. A comparison of the results showed minimum and maximum temperatures differentials of -1.24°C (-2.23°F) and 0.96°C (1.73°F), respectively, while the absolute difference for more than 90% of the time was less than 10°C (18°F). The conductor temperature profile was seen to follow the ambient temperature and increase with line current.
A comparison between the relay ampacities, ambient temperature, wind speed, wind direction and solar radiation between the Skegness and Boston ends of the transmission line indicated the maximum absolute differences to be 117 W/m2 (10.9 W/ft2) for solar radiation, an ambient temperature difference of 2.1°C (3.8°F), wind speed difference of 2.3 m/sec (7.55 ft/sec), which, with a wind angle of 680, gave an ampacity of 305 A. The ampacity difference between the two ends of the line is <10% the majority of the time, so it is important to consider the variation in weather characteristics for each particular application. To compensate for these variations, the DLR relay includes correction factors designed to take into account variations and, if required, provide additional safety margins.
Further on-site testing for a two-day period in 2009 revealed the weather parameters having a marked impact on the line rating compared to Engineering Recommendation P27:
Solar radiation, 407 A to 501 A, the maximum being 0% higher than P27
Ambient temperature, 438 A to 526 A, the maximum being 5% higher than P27
Wind speed, 501 A to 960 A, the maximum being 92% higher than P27
Wind speed plus wind angle, 491 A to 1,161 A, the maximum being 231% higher than P27.
Note that P27 assumes zero solar radiation, which is why the solar radiation ampacity is always below the P27 ampacity. The Engineering Recommendation P27 spring and autumn rating for the Skegness-Boston overhead line is 501 A.
Long-term testing from autumn 2008 to winter 2009, which took into consideration all the weather parameters, revealed the calculated ampacity was higher than the U.K. industry's static ratings the majority of the time. Using data from the Power Donut, the analysis confirmed the absolute difference between the conductor temperature as determined by the CIGRÉ equations and actual measured values was negligible.
Central Networks' decision to install DLR protection on the 132-kV Skegness-Boston overhead transmission line, followed by a comprehensive program of commissioning tests, identified and quantified the scale of benefits offered by this form of circuit protection. As a result of taking into account real-time weather parameters on an existing transmission system, a protective relay was developed to provide control commands and backup should the available power output from the wind generators exceed the ampacity of the 132-kV overhead lines. Nevertheless, although the installation of the DLR relay has increased the line capacity of the Skegness-Boston, a satisfactory operational safety margin has been maintained using fixed conservative wind direction and solar radiation values.
Robert Ferris (firstname.lastname@example.org) was the innovation and development manager for Central Networks, responsible for all the company's research and development projects. Ferris has a degree in electrical and electronic engineering from Aston University and is a member of the Institution of Engineering and Technology. During his more than 30 years with U.K. distribution utilities, he has held positions in both engineering (construction, maintenance, planning and an overhead line specialist) and management (automation task force, protection and automation, and assets).
Alstom | www.alstom.com
Central Networks | www.central-networks.co.uk
CIGRÉ | www.cigre.org
IEEE | www.ieee.org
USi | www.usi-power.org