Project PS-B: Wind Power Integration
Manager: Richard Vinter
Investigators: Goran Strbac and Richard Vinter
Research Staff: Aris Kountouriotis (Research Associate)
Collaborators: Martin Clark, David Angeli
Sponsors: EDF Energy Networks
Start date: 01/10/2010
Linked Projects: UT-B and
UT-C
Summary. The UK is committed to reducing emissions-related climate change, and wind power is expected to have a major role in achieving this goal. The supply of wind energy is by nature variable and unpredictable. Clearly, wind generation will displace energy produced by conventional plant, but its ability to displace capacity will be very limited. This will require that a significant amount of conventional plant is maintained on the system, leading to significant asset utilisation degradation. Furthermore, efficient balancing of demand and supply in real time will become a major challenge. This will be primarily driven by the increased need for various reserve and frequency regulation services to deal with wind output uncertainty and variability. The increased need for system management services will not only reduce efficiency of operation of conventional generation, but it may severely limit the ability of the system to absorb renewable output. Our initial analysis shows that in a 2020 scenario with 30 GW of wind generation together with 12GW of less flexible nuclear in combination with energy efficiency measures, more than 25% of wind production may need to be curtailed. This project is concerned with various aspects of wind power generation and integration research that take account of these factors.
Stochastic modelling, control, optimization and forecasting techniques, including those developed in Projects UT-A,UT-B and UT-C, will be directed at some of the key problems associated with the secure operation of an electrical power system with high wind energy penetration. The extent to which conventional power generation can be replaced by wind power, and the minimum reserve generation requirements to compensate for wind generation uncertainty, are two important issues that will be investigated. We shall, furthermore, investigate the dynamic behaviour of power systems with large wind penetration and HVDC transmission links, and improvements to system stability and robustness that can achieved by means of converter control.
Research has also be carried out into wind turbine controllers and their applicability on a new wind turbine facility at Cambridge Systems, and is reported under project UT-C.
Current Status:
At Imperial College:
Dynamic demand management is a promising technology for improving the ability of the power system to absorb increased amounts of intermittent generation. Previous research has demonstrated that, if an appropriate control algorithm was possible to design, dynamic demand may become a very effective solution of providing reserve response in systems with significant penetration of wind energy.
Work has been undertaken on managing power consumption of domestic refrigerators by means of "smart" thermostatic control. In this approach, the operating temperature of these appliances is modified dynamically, within a safe range, in response to mains frequency fluctuations. The framework we adopt assumes that there is no communication between devices, and so each device has to act in an autonomous setting (decentralized scheme). While this is a severe constraint, and complicates the problem, we note that the quantity of interest is the temperature distribution of the whole population of appliances at each particular time point, and pose the problem in a probabilistic framework. The advantage of this approach is that it greatly reduces the dimensionality of the original problem, while it allows for simple, yet successful solutions.
Previous work with simple feedback schemes, in which the operating temperature is varied in a linear fashion with respect to mains frequency deviations, have proved inadequate in achieving desired performance, as individual appliances tend to "synchronize" with each other, leading to unacceptable levels of overshoot in energy demand, when they recover their steady-state operating temperatures.
In contrast, we have developed a viable control scheme, based on the replacement of classical hysteresis-based controllers with controls that randomly jump between "on" and "off" states of the appliances. Careful selection of the jump intensities allows for the decentralized control of individual appliances' duty cycles (and, therefore, power consumption), while the "population" of refrigerators is sufficiently diversified (mixed) with respect to temperature, to avoid undesirable overshoot phenomena.
The proposed control scheme has been developed within a rigorous theoretical framework, confirming its stabilizing properties and the overshoot-reduction capability. Accompanying analysis establishes that comparable performance cannot be achieved by any linear switching strategy. The theoretical results have been verified via extensive simulations. Papers on the control scheme have appeared in IEEE Trans. Control Systems Technology, and were presented at the 18th IFAC World Congress 2011. A paper has also been published in the IEEE Transactions on Smart Grid, on the implementation of the control strategy taking account of the practical constraints encountered.
The algorithm has been analyzed for the case of battery chargers under simple modeling assumptions. Simulations will be carried out to validate the design and test its robustness to different operating scenarios and penetration of the technology.
Attention in this project has now shifted to quantifying the value of dynamic demand management in supporting cost effective integration of large amounts of wind power in the UK electricity system. This approach has the potential to reduce the response requirements from conventional generators substantially, enabling them to operate more efficiently. A range of scenarios with different generation portfolios are considered, and compared in terms of system costs and carbon emission.
Dr Tom Voice, a Research Associate at Cambridge, is also addressing problems of decentralized demand management of smart grids.
Research is also being carried out into control system design for a brushless doubly-fed wind generator. Such devices have a potential for off-shore generation, owing to their robustness and high energy-conversion capability. They are known, however, to be difficult to control, particularly during transients phases of operation. Dynamic models have been developed and are being used for controller design and analysis. The Cambridge wind turbine facilities will be used, at a later stage of the project, for experimental validation.
At Cambridge:
Research is being carried out into control system design for a brushless doubly-fed induction generator (BDFIG). Such devices have a potential for off-shore generation, owing to their robustness and high energy-conversion capability. They are known, however, to be difficult to control, particularly due to the fact that they are time-varying systems. As with most electrical machines, this time variation can be partially removed by appropriate reference-frame transformations. However, the governing equations are still dependent on the operating speed. As the BDFIG is currently aimed at variable-speed applications, current work focuses on adaptive techniques to optimize control across the operating region. Also, it has been shown that the BDFIG is capable of providing good response during frequency deviations and that grid synchronization can be performed quickly and reliably.
Publications:
[BTM14]
A.W. Broekhof, M.R. Tatlow, R.A. McMahon,
Vector-controlled grid synchronization for the brushless doubly-fed induction generator,
7th IET International Conference on Power Electronics, Machines and Drives (PEMD2014), 8-10 April 2014, Manchesterpp1-5,
2014
[BMM13b]
A.W. Broekhof, R.A. McMahon, J.M. Maciejowski,
Reference Frame Re-alignment for Vector Control of the Brushless Doubly-Fed Machine,
39th Annual Conference of the IEEE Industrial Electronics Society, Vienna.,
2013
[BMM13a]
A.W. Broekhof, R.A. McMahon, J.M. Maciejowski,
Decoupling Method for Vector Control of the Brushless Doubly-Fed Machine,
European Control Conference, Zurich, 2013.,
2013
[AKCAS13]
M. Aunedi, P. A. Kountouriotis, J. E. O. Calderon, D. Angeli and G. Strbac,
Economic and Environmental Benefits of Dynamic Demand in Providing Frequency Regulation,
IEEE Transactions on Smart Grid, on-line access, Digital Object Identifier: 10.1109/TSG.2013.2258047, 2013,
2013
[AK11]
D.Angeli and P.A.Kountouriotis,
A stochastic approach to 'dynamic-demand' refrigerator control,
IEEE Transactions on Control Systems Technology, no 99, pp 1-12,
May 2011
[AK11a]
D.Angeli and P.A.Kountouriotis,
Decentralized random control of refrigerator appliances,
18th IFAC World Congress, Milan, Italy ,
Aug-Sept 2011
