Along with renewables’ growing role in Europe comes unprecedented change in the region’s energy sector. System operators are already calling for flexibility from thermal units to cope with the variability of increasing wind and solar production, and green energy’s low operating costs and subsidies are causing turbulence in electricity markets. Capacity from thermal units will clearly be necessary in the future to provide balance, but will markets be able to deliver it?
Capacity mechanism designs presented to date will not solve this problem. Market mechanisms to attract capacity are still unclear. Wärtsilä has devised a market model for the future that will incentivise flexibility and ensure adequacy of capacity. It is based on two case studies that have shown the value of flexibility in two large power systems: the UK and the US state of California.
Since renewables production generally has feed-in priority, remaining capacity has to adjust its output to balance total electricity production and demand. System operators need to have capacity available that can respond quickly to changes in electricity demand and output from renewables, which can be rapid.
The impact of the deployment of renewables on electricity markets is severe. Such sources generate electricity at low marginal costs and therefore push thermal capacity higher up in the merit order or completely outside it. This means the operating hours of thermal capacity fall and it generates less revenue. Subsidies for renewables also depress electricity prices, which makes the feasibility of thermal plants even more challenging. Thermal capacity is still needed in a high-renewables system for balance, but its profitability is jeopardised.
Several EU Member States have stated that plant closures and a lack of investment in new capacity may prevent the market from bringing forward sufficient capacity under current market arrangements. Allowing electricity prices to reach high levels at peak times would be necessary to allow plants running at low load factors to recover fixed costs. However, it is not simply capacity that is required in a high-renewables system.
Without appropriate price signals, there is an equally important concern around “missing flexibility”. Systems require a sufficiently flexible mix of capacity as well as the right types of capacity.
The importance of flexibility
Transmission system operators (TSOs) and other market players recognise the increasing need for flexibility but the value of flexibility has not been quantified or identified in market arrangements. Wärtsilä has conducted several studies around this topic.
The first step in the process is to define the future power system architecture, which will be based on objectives such as emissions, reliability and costs, which policy makers set. To determine how to achieve these objectives requires the creation of several capacity scenarios with different mixes of technologies. The output of step one is an architecture that can meet future objectives.
The architecture provides input to the second step of the process, i.e. the modelling of power system operations, or despatch. Despatch software PLEXOS was used in recent studies of the UK and Californian systems. Inputs for the model are the expected capacity mix (including the capabilities of these technologies), weather and load data, system requirements (such as required system reserves) and market operation, for example how reserves are procured and at what price. The tool optimises the generating costs of a system in a chosen interval in line with the trading blocks of the system, for example every 30 minutes.
The third step defines the value of flexibility by comparing the results of different scenarios. Power system modelling provides the system operating costs and CO2 emissions as an output for each scenario.
Different generating technologies have different ways of providing flexible electricity. Some can start up from zero output and ramp up within seconds. Others may take hours, but can quickly flex their output up to meet the system needs once they are generating above a stable level. This is typical of large units such as large combined-cycle gas turbine (CCGT) or coal-fired plants. Typically these slower technologies provide a system’s flexibility requirement today.
Part-loading may have been effective in the past but today it is not likely to be the most efficient method of providing the greater flexibility needed in the future.
Part-loading generates extra costs because of increased carbon costs, reduced fuel efficiency, the greater number of generators needed on the system, and the costs of curtailing wind generation to maintain system security.
Given these costs, if conventional sources of flexibility are used in a system with a high level of renewables, the full benefits of decarbonisation may not be achieved and consumers will pay higher prices.
Smart power generation (SPG) describes power plants such as modern gas-fired types that are flexible and avoid the costs associated with part loading. SPG can provide savings for three reasons.
The first is speed. From zero output, SPG can respond almost instantaneously to fluctuations in supply and demand, so they do not need to be part-loaded.
The second is sustainability of output. Unlike other fast-start technologies, SPG can start up quickly and hold output without needing to be relieved quickly afterwards.
Finally, SPG is efficient. Such plants incur minimal costs for being on standby as reserve but can deliver much needed electricity as quickly as conventional flexible technologies and even more quickly in some cases.
Valuing flexibility – UK
August 2012 saw the UK’s Department of Energy and Climate Change (DECC) publish an analysis that estimates how flexibility from a range of sources can generate significant savings to UK consumers, particularly in a scenario of high wind penetration. These sources include demand-side response (DSR), increased interconnection, storage and thermal generation.
Redpoint Energy and Imperial College London followed the report with further analysis of the potential value of flexibility through detailed modelling of the UK power market and balancing costs. The focus has been on supply-side flexibility provided by SPG. The results, however, are more generally applicable to all sources of flexibility, whether DSR, storage or interconnection.
The modelled scenarios are based on projections by the DECC and National Grid, the UK TSO, for demand and capacity mix development by 2020 and 2030. Two capacity mixes came under investigation in the scenarios of high wind and base wind, with and without SPG, for the years 2020 and 2030, as Figure 1 shows. In a ‘No SPG’ capacity mix, efficient gas generation capacity comes from a mixture of combined-cycle gas turbine (CCGT) and some open-cycle gas turbine (OCGT) generation. In an ‘SPG’ capacity mix, 4.8 GW of SPG replaces the same amount of the most fuel efficient CCGT capacity. SPG has a slightly lower net electrical efficiency but superior flexibility compared with CCGT.
|Figure 1: Capacity mixes for power system modelling in the UK, with base and high wind scenarios|
What is the impact of SPG on the provision of system flexibility? Depending on the case, SPG is the least cost option to provide flexibility 35-40 per cent of the time. With SPG providing system flexibility in an optimal way, more room is available for efficient CCGTs and coal-fired generation to run at full load, providing cheap electricity to consumers.
The analysis showed that, depending on the wind scenario, flexible gas generation could save the UK consumer between à‚£380 million and à‚£550 million ($566 million and $820 million) per year by 2020 through reduced balancing costs. By 2030, savings range from à‚£580 million to à‚£1.5 billion, as the volume of wind in the system is expected to increase further.
A comparison with the UK system-wide generation costs is useful to give some scale to the potential savings in balancing costs. With an increasing amount of low-cost renewables generating electricity at almost zero marginal cost, the total generation costs will fall when the output of renewable generation increases. However, the need for balancing actions will increase accordingly, and these costs will have a significant role by 2030.
The savings potential of SPG is as high as 5 per cent in 2020, increasing to an impressive 19 per cent of total generating costs in 2030.
Valuing flexibility – California
California aims to increase generation from renewables to 33 per cent by 2020. However, this development has started a debate about what flexible assets will be required to secure the reliable operation of the power system.
California’s system will face another issue in the near future when new environmental regulation may force the retirement of plants with once-through cooling that total 12 GW in capacity. The state’s system operator CAISO concludes that 5.5 GW made up of equal amounts of new CCGT and OCGT is required by 2020 to secure reliability.
DNV KEMA Energy & Sustainability has analysed the Californian system for 2020 by using dynamic system modelling. The base case for the power system modelling was the Californian system for 2020 with a renewables penetration of 33 per cent, made up of wind and solar but excluding hydro, and 5.5 GW of new gas turbine plants, made up of equal amounts of new CCGT and OCGT. The alternative modelling scenario had the same basic assumptions but 5.5 GW of SPG replaced that amount of gas turbines.
By introducing 5.5 GW of SPG instead of 5.5 GW of gas turbines in the system, California’s consumers save around $900 million per year, representing 11 per cent savings in system-level generating costs. Figure 2 shows the cost breakdown of the total system operating costs for the modelled scenarios.
|Figure 2: Value of flexibility in California in 2020|
The studies conducted by DNV KEMA, Redpoint Energy and London Imperial College make evident that the inclusion of SPG in a generation portfolio reduces total system operating costs in systems with a high penetration of renewables. This is because SPG provides flexibility at low cost.
In addition, by adding SPG to the capacity mix of a power system, other thermal plants no longer need to run in part load and can produced electricity at a higher efficiency, which reduces overall generation costs.
A system without SPG can provide flexibility by running plants at part load, but such actions significantly increase costs to consumers, as the studies show. The value of flexibility in the examined 60 GW UK and California peak load systems with high renewables penetration is greater than €500 million ($642 million) per year.
Translating this to a system the size of Europe’s, the value of flexibility is estimated to be greater than €5 billion per year, even by 2020. Consequently flexibility should be one of the key parameters in the design of a future power system and energy market.
A new market vision
In February 2013, the European Commission asked for stakeholders’ inputs on potential ways to secure capacity adequacy and system reliability in a future system with high amounts of renewables. In a high-renewables power system, flexibility is no longer an invisible and low-cost side product of power generation but a key factor in power system design and optimisation.
Although the studies of the UK and Californian systems clearly indicate the benefit of flexibility in the capacity mix, current market arrangements do not reflect the value of flexibility or incentivise investments in flexibility. They also hide the cost of inflexibility within consumer bills and consequently prevent investments in new flexible capacity. At the same time, energy-only market setups are struggling to keep capacity at adequately healthy levels.
Wärtsilä has studied several electricity market models with the aim of developing one that will incentivise flexibility and ensure capacity adequacy for a system with a high contribution from renewables. The market model should secure capacity adequacy, incentivise the right type of capacity and lead to the least cost to consumer. Figure 3 shows the overall market model design that will deliver this. It is based on two markets existing next to each other.
|Figure 3: A new market design for a power system with high renewable energy integration|
The energy market, consisting of the wholesale electricity markets (day-ahead, intra-day and balancing markets), and a flexibility market, establish a competitive environment. A competitive capacity market would be introduced only if needed, to secure capacity adequacy.
A competitive energy market forms the basis of the market model. The objectives of energy markets are to provide low-cost electricity and low CO2 emissions in all situations via competitive short-term markets.
Cost-reflecting imbalance prices will increase the imbalance exposure of all market participants (where all participants are responsible for balancing), which incentivises balance at gate closure. Supply and demand for energy closer to gate closure is therefore expected to increase because each market player, in order to reduce the risk of out-of-balance penalties, will make efforts to be in a balanced position at gate closure, either through changed positions within its own portfolio of options such as changing the outputs of its own power plants or DSR, or through trading.
This development enhances the liquidity in intra-day markets and provides additional income for flexible assets through balancing and intra-day markets because these units will be in a position to supply energy shortly before gate closure. However, it would be hard or even impossible for providers of flexibility to capture the total value of flexibility through energy prices alone. Therefore, in addition to the energy market, we propose the introduction of a market for flexibility.
A competitive flexibility market would be a day-ahead option market for flexibility to increase or decrease energy the following day. The flexibility market would replace the existing procurement strategies of TSOs and would make the procurement of system services more transparent to market players. TSOs would go to the flexibility market to procure the flexibility, or reserves, required to satisfy the needs of the system for the following day, when the volumes are not locked away under long-term contracts.
The flexibility market would also be open for market participants to procure flexibility to hedge against intra-day prices and imbalance exposure.
There are many key features to the flexibility market. When it comes to buying flexibility, the TSO would always procure it according to the needs of the system. However, procurement by market participants could reduce the amount procured by the TSO.
Market participants determine their own volume requirements depending on their willingness to hedge against price risk, and the TSO acts as a backstop in the day-ahead auctions to ensure that the system has the flexibility needed. The TSO procurement strategy provides stable volumes and liquidity in the flexibility market and makes known the total volume of the flexibility requirement.
Another feature is that multiple products, such as 5-minute or 30-minute ramping, are defined by the TSO. This ensures the needs of the system are met. All products require an option to deliver an increase or decrease in the physical energy in any future settlement period.
Also, the day-ahead timeframe aligns with the energy market or allows co-optimisation with it and provides a daily reference price for different flexibility products. A secondary within-day market for participants and the TSO allows them to trade their options as more information emerges. Clear day-ahead reference prices can allow long-term financial contracts to be struck between flexibility providers and market players or the TSO.
The option holder (i.e. market participant or the TSO) may exercise the option by calling for energy to be delivered prior to gate closure. Self-provided flexibility must provide information to the TSO within-day on whether it will be exercised. After gate closure any unused options would be exercisable by the TSO in the balancing market.
Another key feature is cash flows. Flexibility cleared through the day-ahead auctions, other than self-provided reserve, is paid the market clearing availability fee per megawatt for the contract period. A utilisation fee per MWh is paid on exercise. Unused flexibility must be offered into the balancing market at the fixed utilisation fee for despatch and payment by the TSO.
Ensuring cost recovery is also important. The option holder pays the availability fee to the flexibility providers. The availability fees incurred by the TSO can be recovered via an information imbalance charge levied on out-of-balance market participants.
Finally there is the monitoring feature. The TSO would certify the physical capability of capacity providers who seek to offer into the day-ahead auctions. Any options exercised would be notified to the TSO in the same way as physical energy.
A central capacity market would be established if the energy and flexibility markets are not delivering investments or are not able to keep existing plants in the system. The purpose of the capacity market is to ensure capacity adequacy by providing so-called administrative capacity payment, which compensates the ‘missing money’ from market operations.
While future energy and flexibility markets are volatile by their nature, investors may require stable cash flows to be able to finance new projects. A capacity market could enhance the bankability of new projects. The capacity market, like any capacity mechanism, should concentrate on securing capacity adequacy rather than specifying what type of capacity is needed. It should treat all forms of capacity on an equal basis.
Thus, a well functioning energy market together with a flexibility market would reward capabilities, while a capacity market provides the ‘all-in price’ required by investors to make investments.
Change in market design needed
An increasing penetration of variable renewable generation into a power system changes its operations and impacts market fundamentals. But while system operators are calling for flexibility from the generation side, the thermal fleet takes a big hit as its operating hours are reduced while the average electricity price is lower. The result is increasingly uncertain market-based revenues for thermal plants.
There are potential market-based approaches to incentivise investments in flexibility. These approaches do not require administrative cash flows but call for a reallocation of system costs from the TSO to the market, making the cost of flexibility visible for market players. To develop a reliable, affordable and sustainable power system necessitates several actions.
Firstly, there must be an understanding that more renewable generation has caused dramatic changes in the energy market environment. Secondly, there must also be recognition of the value of flexibility, which must be made visible for market players through cost-reflective imbalance prices and by developing short-term energy markets. Thirdly there must be a transparent market explicitly for flexibility. This will enable efficient procurement of system services and provide clear market signals for investors in flexibility.
Finally, new players must be able to enter the market and new projects must be made bankable by introducing a capacity market if the energy and flexibility markets are not delivering investments.
To avoid the risk of locking in the wrong type of capacity, it is important to take the first three actions before considering the fourth.
Many market players are calling for a market-based approach regarding the EU electricity market structure. We hope we have shown that it is possible to design a market that provides investment signals for the right type of capacity and ensures capacity adequacy at the same time.
Matti Rautkivi is general manager, Business Development, Power Plants, and Melle Kruisdijk is director, Market Development Europe, and Business Development, Power Plants at Wärtsilä. For more information, visit www.wartsila.com
This article is based on a Best Paper Awards winner at POWER-GEN Europe 2013.
Power Engineering International Archives
View Power Generation Articles on PennEnergy.com