Tackling renewable energy integration: five lessons from Sri Lanka
Ambitious renewables targets, movements away from thermal generation, increasing demand for electricity … it’s boom time for renewables around the world. Yet the global energy transformation brings some challenges, particularly for network stability and security.
Sri Lanka offers an interesting case study in the complexities of quickly integrating a large proportion of renewables and how some of the hurdles may be overcome. The lessons we can learn from the Sri Lankan experience are applicable worldwide.
In 2016 the government of Sri Lanka set a very ambitious renewable penetration target, with 100% of power generation to come from renewable sources by 2050. Coal and oil generation, currently accounting for around 50% of all generation, will cease by 2045. If this isn’t challenge enough, add in the projection of maximum demand rising above 7000 MW (from a current daily maximum of 2500 MW) by 2040.
Between September 2015 and March 2016, three nation-wide blackouts threatened to significantly undermine Sri Lankan industry and foreign investment prospects. The challenge for the Ceylon Electricity Board (CEB, which runs Sri Lanka’s power system) now is to achieve the country’s power transformation while still ensuring network security and reliable electricity. Any future unexpected system outage would be likely to erode public confidence in renewable development, regardless of whether the outage relates to renewables or not.
What can we learn from the Sri Lankan experience of the 2015/16 blackouts and Sri Lanka’s current challenges in its journey to a renewable energy future?
1 – Focus on accurate modelling of the power system
The first lesson to be taken from the 2015/16 blackouts is the need to focus on power system modelling. The first blackout occurred after a trip of the largest generating plant under light load. The next was due to a trip of a major transmission line when two parallel circuits were taken offline for maintenance. Neither of these events were envisaged through system model simulations. If a utility’s simulation tools won’t predict such events, planning engineers will struggle to anticipate or guarantee system security under system contingencies.
For system planning and operations, CEB has developed a power system model consisting of major hydropower, oil, coal and wind plants. However, no model accuracy requirements are specified in the national grid code, which leads to the use of generic models with unknown modelling accuracy. Prior to Sri Lanka’s 2015 blackouts, limited tests were conducted to tune the models to accurately represent behaviour. During 2017, CEB embarked on a project to test the model validation for at least some of the major hydro units. These tests need to extend to all major power plants so that engineers can trust the actual performance of every major generator in the system.
2 – Determine a clear, single responsibility for the system model
The second lesson from Sri Lanka’s system modelling is the need for a single owner of the system model files, with responsibility to maintain the model and ensure its accuracy. If different departments of a utility have different system models, results of an event analysis may vary.
Ideally, the entire power system should be modelled accurately and the model controlled by a single team. This is vital for maintaining integrity of the system studies conducted by various parties.
3 – Consider minimum generation
When making decisions about the size of power plants, it is quite common for the focus to be on maximum generation. However, to minimise any unwanted constraints, it is really important to consider the lowest level of generation at which the plant can operate.
Three coal plants in Sri Lanka each have capacity of 300 MW with minimum generation of 180 MW. During periods of low demand, all three coal plants need to remain at their minimum of 540 MW (3 x 180 MW) even if other generation is available in the system.
4 – Ensure the network code looks to the future
When determining the grid code, it is essential to keep an eye on the development of non-conventional renewables. The grid code needs to be able to manage power quality issues in the future while minimising unnecessary costs to developers. In other words, it must have enough regulatory power to impose required technical targets, yet be flexible enough to minimise unnecessary mandatory capital expenses.
5 – Set up an ancillary service market
Sri Lanka does not currently have an ancillary service market, so the fall-back position is mandatory interruption to the system load under certain contingencies. This will not be acceptable in the longer term and customers will demand higher reliability. A way of categorising system support, especially frequency control ancillary services, will enable CEB to maintain system reliability as well as understand the quantities and price to deliver the required reliability.
Some tightening of Sri Lanka’s current technical and connection policies and practices will be needed as the country embarks on drastic change in its generation mix in pursuit of its renewables target. Yet the lessons we’ve explored in this article are readily applicable to any network keen to accelerate substantial integration of renewables, especially solar PV and wind, for a successful energy transformation and a future of reliable and sustainable power.
If you would like to find out more about how Entura can help you overcome network challenges when integrating renewables, please contact Ranjith Perera on +61 3 6245 4272, Shekhar Prince on +61 412 402 110 or Patrick Pease.
About the author
Ranjith Perera is a Specialist Power Systems Engineer at Entura. He has over 22 years of experience in Australia and South-East Asia, working on customer and generation connections and broader power system analysis. Ranjith has provided power system advice on a wide range of network augmentations, network planning and system stability in Australia and internationally. These studies included option analysis in transmission planning, constraint analysis, determination of reactive support (dynamic or static) in system stability / TOV and detail load modelling in voltage stability. Ranjith also developed voltage recovery guidelines to TNSP based on regulatory requirement and customer equipment tolerances.
April 3, 2018