Flexibility optimization on Belgium’s passive balancing market

So far, the 2024 power markets’ ‘word-of-the-year’ has been ‘flexibility.’ With the increased integration of intermittent renewables and the resulting price volatility, flexibility has become a critical component of the energy transition – and of a successful short-term trading strategy.

At Dexter Energy, we’ve been developing and optimizing flexibility software-as-a-service since 2018, starting with the Netherlands. Today, we provide solutions to several household names and manage over a gigawatt of renewable assets. Drawing from our experience, we’ve identified the next key market for Flex Optimization: Belgium.

If you’re trading renewables on the Belgian short-term power market, read on and unlock a new profit-making opportunity.

Defining ‘flexibility’ in power markets

In a webinar we recently hosted with Montel, André Bosschaart, Head of Analytics, explained that “everything on the energy market is created as different products.” On the TSO side, products such as FCR, aFRR, or mFRR balance the grid in real-time (in MW). On the market side, traders operate hourly products on the day-ahead market, intraday auction or continuous products, or balancing system products (in MWh).

In André’s definition, flexibility is the option to deliver one of these products, i.e., the right to dispatch the asset but not the obligation.

Determining which products an asset should deliver and when becomes essential to stacking value. It is the basis of flexibility optimization, and is becoming increasingly important.

Revenues from renewables under pressure

The growth in renewable energy has led to a decrease in revenue from power production. As a result, the business case for renewables is under pressure across Europe. We identified three main developments behind this change.

Firstly, subsidies are gradually phased out and the value of GoOs (Guarantees of Origin) is declining, making renewables assets less attractive for developers.

Secondly, negative day-ahead prices have become common, not just with the coming of spring. As the graph below shows, Belgium has seen an upward trend in the occurrence of negative spot prices since 2019. These instances are often correlated with wind and solar generation surpluses, either within Belgium or in neighboring countries such as Germany and the Netherlands, with whom Belgium is highly interconnected.

Of course, 2022 and 2023 have been exceptional years due to the gas crisis, but, overall, we expect the trend from 2019 to continue beyond 2024.

Thirdly, there is a substantial increase in the imbalance price risk and, likely, rising balancing costs. For illustration, the plot below presents the distribution of the EUR/MW spread between the day-ahead and the imbalance prices. To normalize the extreme quarters during the energy crisis in 2022 and 2023, we’ve added the gray-shaded area.

Focusing on the borders of the highlighted area reveals that the spread is increasing. This indicates increased price risk and, consequently, drives rising balancing costs.

These factors explain why flexibility is emerging as a risk mitigator. However, not all energy markets are suitable for flexibility optimization; specific ingredients are required for a market to be fit for flex.

Evaluating market suitability for flex optimization

There are many similarities between the Dutch and Belgian power markets; thus, following our success in the Netherlands, we hypothesized and confirmed that Belgium is also suitable for flexibility solutions.

There are three crucial aspects behind this conclusion:

• Passive imbalance: Being allowed to deviate from your day-ahead position to go into imbalance. (This is legally not allowed in some countries, such as Germany.)
• Data transparency: Another box checked, as the Belgian TSO, Elia, publishes minute data that serves as a base for forecasts.
• Increasing imbalance volumes and spreads: Imbalance volumes and spreads have indeed increased year-on-year.

It is worth mentioning, nonetheless, that there is one major difference with the Dutch market. In the Netherlands, short-term traders must navigate a specific challenge: dual pricing or regulation state 2. It occurs when the TSO needs up and down regulation volumes at the same time. The Belgian market has a single price for imbalances, providing more data transparency and tools for balancing.

In the next section, we examine the balancing market in Belgium more closely.

Passive balancing in Belgium: market behaviors

To further understand the context for flexibility in the Belgian balancing market, let’s focus on imbalance prices, market behaviors, and expected changes.

In 2022, the lowest average weekly imbalance price was -171 EUR; in 2024, excluding outliers, it nearly reached -600 EUR. Including outliers, it would be as low as -800 EUR. The 7th of April even saw an extreme low: -4,500 EUR. This downward trend presents a clear opportunity for passive balancing.

However, there is an associated risk of the market flipping from negative to positive prices. In 2024, there were about 130 weekly instances where prices went from negative to positive, either within a PTU or from one to the next. Holding an imbalance position during these flips (i.e., being on the ‘wrong side’) could result in significant costs, emphasizing the need for a strategy that adapts rapidly to market changes.

Upcoming market changes will also impact developments in Belgium and should be considered. Belgium is joining Picasso (aFRR market coupling) and Mari (mFRR market), which could smooth out price volatility by importing from other markets. Furthermore, Elia is increasing data transparency by implementing an imbalance price forecast, aligning with their goal of providing tools to help parties manage system imbalances and earn revenues.

Equipped with all this information, we can now illustrate what a passive balancing strategy could look like.

A flex optimization strategy: Curtailment

Curtailment – voluntarily ramping down real-time production in reaction to system imbalances – is an excellent opportunity to unlock flexibility from renewable assets in Belgium. By curtailing, you are deviating from your day-ahead or intraday positions. But, in doing so, you are helping to balance the grid (by curtailing during oversupply) and earning revenue (by ‘hitting’ the negative prices).

Curtailment basics

The plot below shows the profile of a wind asset over a couple of days in April. The blue line is the theoretical production – what the asset would produce based on wind speed and other factors, without any restrictions. The turquoise line is the actual power metered.

Note how the turquoise line sometimes goes to zero, particularly around noon. This is when solar production tends to cause imbalance volume due to oversupply, a frequent pattern in both spring and summer.

Strategy components

To minimize risks when curtailing renewable assets, you’ll need a solid strategy with the following building blocks:

1. Imbalance price forecast

Signals from the TSO and other forecasts tend to be on a minute basis, or even shorter; the imbalance price, on the other hand, is settled on a 15-minute block. Thus, your price settlement prediction should be available as early as possible in the PTU. It is essential to get it right in order to avoid being exposed to the risks described above.

2. Asset specifications

Understand how to steer your asset, including ramp-up/down times and costs. The steering signals can be per minute (or shorter), but, as mentioned, the imbalance settlement is per 15 min.

3. Steering strategy

Based on the input data, define the logic for taking action (i.e., stopping or starting production). Choose between a simple, ‘naive’ strategy or advanced strategies that consider asset specifications and market flipping probability.

4. Hardware connection

Finally, turn the steering signals into actions by connecting them to the actual asset. Note that delays in curtailing or restarting your production could result in your steering strategy applying to the wrong PTU. Therefore, we recommend working with a robust hardware controller that offers a minimum response time and high uptime.

By integrating these four components – visualized below – you can fully unlock flexibility on the passive balancing market, resulting in potential imbalance revenues, on top of your day-ahead earnings.

How do you know the strategy works? At Dexter, we measure strategy performance against two metrics: the perfect forecast and the baseline strategy of simply following the Elia signal. The goal is to perform better than the TSO signal strategy and closer to the perfect strategy while considering the market impact of our customers’ trading volume.

The time is now

Flexibility in the power market is here to stay; it will help us manage the increasing pressure on revenues from renewable production, keep pace, and even accelerate the energy transition.

Luckily for short-term traders, Belgium has the right conditions for renewable asset flexibility; curtailment can capture many profit opportunities. But, as with many aspects of the energy landscape, solid trading strategies are complex. Mistakes can lead to painful losses. So, make sure to reach out if you need a partner!