A 2.3 MW battery energy storage system is enabling a utility to accommodate wind energy penetration of 85 per cent. Michael Lippert explains how

Located half way between Iceland and Norway, the Faroe Islands is pushing forward with a green policy vision to transition from imported fuel to renewable energy.

The local utility SEV operates an electricity system delivering up to 50 MW to its 50,000 inhabitants and about 60 per cent of this was supplied by renewable sources in 2015, with a vision to increase its renewable electricity production to supply 200 percent of its forecast 600 GWh demand by 2030.

This will be achieved from the islands’ abundant wind and hydro energy resources, together with emerging technologies like solar photovoltaic and tidal stream energy. Simultaneously, household heating is transitioning from oil to heat pumps and increasing use of electric vehicles is also contributing to a changing energy mix.

As a major step in its transition, SEV inaugurated the Húsahagi wind farm in 2014. The new facility added 12 MW to the islands’ previous 6.5 MW wind resources. Recognising that power systems with high wind penetration have many challenges, SEV worked with leaders in wind energy and industrial batteries ENERCON and Saft to ensure stable operation.

As wind energy is being integrated, several challenges are arising. Some are well-known – for example, the natural intermittency of wind generation can impact grid frequency and voltage. Another is that wind generation does not offer the benefit of system intertia.

A less well-known challenge is the potential impact on the grid of shifting from the type of synchronous generation used in SEV’s diesel gensets to the inverter-based generation of wind turbines. These effects are being studied under the EU-funded MIGRATE project.

Recognising that the Húsahagi wind farm would lead to high penetration, SEV was keen to integrate energy storage to ensure system stability to overcome these challenges. It worked with ENERCON and Saft to install a lithium-ion (Li-ion) battery energy storage system (ESS) for the site.

Ramp rate control

Energy storage is a flexible tool with several applications. These include ramp rate control, peak shaving, power shifting and frequency regulation and they can be applied individually or in combination.

The first stage in sizing a ESS is understanding the role expected of it. In the case of Húsahagi, SEV considered combining ramp rate control and frequency regulation and ultimately decided that its priority was to control the ramping rates. This will ensure a smooth flow of electricity into the grid, with no sudden step changes in output caused by a gust or lull in wind.

At Húsahagi, SEV required the ramping rate to be no higher than 1 MW per minute, i.e the combined power flow of the wind turbines and the battery at the point of connection (POC) must not vary by more than 1 MW within a one-minute time interval.

SEV also defined that downward ramp control would be done with storage only (with the battery injecting power to compensate wind power drops), and the battery would absorb energy to control upward ramping, possibly complemented by limiting the power from the turbines.

Finding the sweet spot

Once its role was established, Saft carried out modelling to find the optimum power rating and energy storage capacity to deliver the lowest Total Cost of Ownership (TCO) for SEV.

A smaller ESS will have a relatively low capital cost and take less space. However, it may cost more over its lifetime due to lower revenues, less compliance with the grid code, higher curtailment losses, more regulatory penalties or a shorter calendar life. Because each project is unique, batteries must be sized according to the specific conditions at each site as a collaboration between the project partners.

To find the power capacity at Húsuhagi, SEV provided high-resolution wind generation data from the nearby Neshagi wind farm. This data was accurate to the second and enabled Saft to simulate the operation of the ESS in detail to establish the energy power rating as 2 MVA.

To find the optimum energy storage capacity, an algorithm then mimics the performance of the ESS over time, taking account of the electrical and thermal performance as well as electrochemical aging and the mode of operation, which is ramp rate control in this case. The model also factors in financial and regulatory factors, such as the potential for curtailments and penalties, and greater income from integration of more energy.

After modelling the lifetime cost for several different sizes of ESS, Saft identified the energy storage capacity sweet spot as 700 kWh.

Limiting factors

Modelling also identified two possible limiting scenarios that may arise from Húsuhagi’s wind conditions. The first is down to gusts of wind leading to the ESS delivering more downward regulation and lowering the battery’s state of charge (SOC). The second limiting factor is experienced in the rare instances when very large and sudden wind changes lead to wind power varying by more than 3 MW in a minute.

To avoid non-compliance when these situations arise, SEV’s solution includes a mechanism to reduce power injected into the grid when the SOC is below 30 per cent and increase injected power when SOC is higher than 70 per cent. This has the effect of moderating the rate of charging to ensure the battery has enough capacity to inject or absorb energy when required.

Modelling also had a secondary role in defining the control strategy under the Local Energy Management System (LEMS) to ensure that the wind farm’s output is smooth at the point of connection with the grid. It must also maintain the battery’s SOC within a desired range.

The wind power production data from the Neshagi wind farm was re-used to simulate power smoothing performance against normal, heavy and extreme wind conditions. Results showed that effective power smoothing within SEV’s limits was possible while maintaining the battery’s SOC within the envelope recommended by Saft to ensure an optimum lifetime.

The simulation proved that the control strategy could significantly enhance the wind farm’s output and reduce the impact of power fluctuations on the grid. However, as the ESS has around 20 per cent of the wind farm’s nominal power, under some extreme wind conditions the ESS reaches its limits without being able to completely filter the power fluctuations.

Implementing the ESS

Having confirmed the capabilities of the ESS through simulations, SEV worked with Saft and ENERCON to implement the system.

Saft supplied an Intensium Max 20 HP ESS with 700 kWh energy storage capacity that is capable of delivering 2.4 MW continuous power and 3.3 MW peak power. The ESS is housed in two 20 foot containers that contain battery modules that are built up in series and parallel to deliver the required power, storage and voltage.

Battery management, monitoring, heating and air conditioning, and safety equipment is also integrated into the containerised systems, which arrived on site ready to plug and play. Effective battery management of the battery modules and strings is important to ensure consistent aging of the individual cells that make up the battery. In turn this leads to consist long-term performance and optimised return on investment for the operator.

Another important aspect of Li-ion battery systems is that because they require electronic battery management, they can be integrated into control systems.

In this case, the ESS is linked with the Power Conversion System (PCS) through a communication bus that is based on the CANbus communications protocol.

Power conversion system

As well as supplying the turbines for the site, ENERCON also supplied power conversion and control equipment in the form of an Enercon Smart Container – a 40-foot container that houses seven pairs of 330 kVA inverters, a power transformer and medium-voltage switchgear – to control the wind farm’s output to the Faroe Islands’ 20 kV grid.

The Smart Container acts as the interface between the AC power system and the ESS. As the local grid operator, SEV chose to focus the Smart Container in active power production. Under fault conditions when the grid voltage drops due to a short circuit, SEV specified that the inverters will ride through without injecting current into the grid.

The Smart Container also maintains the SOC limits for the ESS to avoid reducing the battery’s lifetime, with the LEMS determining the power set-points.

When controlling gradient, the LEMS monitors output and produces and opposite power gradient to smooth the power injected. It also regulates the amount of energy charged and discharged by the battery each day to protect the ESS lifetime. The LEMS can also limit the power produced by the wind power during extreme wind conditions.

If desired, SEV can also operate the LEMS in frequency control mode, where the ESS injects and absorbs energy to maintain grid stability.

Testing and operation

Before installation, the complete system was tested at Enercon’s test facility in Germany, where its power smoothing capabilities were tested. Another test evaluated the system’s reaction time in the case of the sudden loss of a wind turbine.

After successful laboratory testing, transport and installation on site, the ESS entered operation with data recorded during commissioning confirming that the battery’s operational profile is close to the initial simulations. Wind power smoothing performed in line with the expectations, which verified the power rating.

Additional SOC testing was carried out under normal and extreme conditions – and this verified that the SOC level is successfully maintained even during extreme variations in wind power. The ESS significantly reduced the variability of the power injected into the grid. During typical conditions, the ESS successfully contained ramp rates below 1 MW per minute window most of the time, with the rate never exceeding 3 MW per minute during the test period.

As an example, see the graph ‘ESS reduces variability of grid frequency’. This shows the variability on a day of high wind generation – typically between 3000 and 400 kW/second in both directions. This was reduced by a factor of 15-20 at the POC due to the ESS

SEV also evaluated the ESS’s influence on grid frequency and found that it reduced the standard deviation of the frequency when the wind farm’s output was smoothed by the ESS. This is illustrated in the graph ‘Impact of the ESS on the grid frequency’ – at the time 10.05 the ESS was turned on and the impact on the grid frequency is clear.

Another benefit is a reduction of curtailed energy, where wind output does not meet the grid code and so cannot be accepted. Thanks to the ESS, curtailment dropped from 22 per cent in the first quarter of 2015 to 9 percent in the same period of 2017.

However, the savings are higher over the winter periods of high wind production between October and March. During winter of 2014/15, 28 per cent of energy was curtailed from a potential generation of 28 MWh. This dropped to curtailment of 19 percent from 23 MWh in winter 2015/16.

Once SEV’s control room operators gained experience in the second year of operation, they were then able to achieve a curtailment figure of only 9 per cent from 23 MWh in 2016/17. This enhanced figure was achieved from greater understanding of the wind’s behaviour at the site, as well as the staff learning how to combine the power outputs of the ESS and the wind farm. It meant that they could reduce power gradients as well as disturbing effects on the grid.

Learning lessons

As Europe’s first fully commercial Li-ion ESS for integrating a wind farm, the project partners gained valuable experience and learnings from Húsuhagi.

The first of these was that the project required close collaboration between the power system operator, battery supplier and control system supplier.

This was essential to understand and analyse the problem and then design, validate, implement and optimise a technical solution with a significant ROI.

The project also confirmed that there is no silver bullet or standard solution for integrating a utility-scale wind farm. The project required a thoughtful combination of solutions and a site-specific iterative design and modelling process.

The project demonstrated that 85 per cent penetration of wind power is technically possible and stable. Controlling wind gradients is the most critical function to achieving this stability.

This function requires mainly short-term power, which can be realised with a limited amount of battery energy, providing dynamic SOC management is in place. Li-ion battery technology is a major advantage to achieving this as it allows dynamic power cycling with a high daily energy throughput.

At the outset, there was a high level of uncertainty about the real-world behaviour of the power plant and the impact of wind variability, inverter-based generation and other factors that affect grid stability. Now that the system is in operation, it is demonstrating the effects of high wind penetration on the grid.

SEV’s wind power plant and electricity grid would operate without the ESS. However, it has a positive impact on the stability of the grid’s frequency and islanders can now make more use of their wind generation.

The ESS is an important step forward on the way to the Faroe Islands’ target of 100 percent renewable energy by 2030. It can also be used as a model for other operators of isolated power systems.

Michael Lippert is marketing and business development manager for energy storage at Saft